; xyz{}~Oh+'0HP
$TU Delft Repository search results0TU Delft Repository search results (max. 1000)TU Delft LibraryTU Delft Library@<<@<<՜.+,0HPX`hp
x
WorksheetFeuilles de calcul
B=%r8X"1Calibri1Calibri1Calibri1
Calibri 83ffff̙̙3f3fff3f3f33333f33333.TU Delft Repositoryg V,uuidrepository linktitleauthorcontributorpublication yearabstract
subject topiclanguagepublication type publisherisbnissnpatent
patent statusbibliographic noteaccess restrictionembargo datefaculty
departmentresearch group programmeprojectcoordinates)uuid:b7ccbf2752cf4d50ba649282aec248e4Dhttp://resolver.tudelft.nl/uuid:b7ccbf2752cf4d50ba649282aec248e4RSimulation Verification and Optimization of a Vertical Axis Wind Turbine using CFD
Kortleven, M.Kenjeres, S. (mentor); Heemink, A.W. (mentor)NIn this research computational fluid dynamics (CFD) is used to model a vertical axis wind turbine. Well known turbulence models  i.e. the kepsilon and komega SST model  are used and compared in their performance against each other and a set of experimental data. The formal research question is "What models are available for the simulation of vertical axis wind turbines and what considerations should be taken into account to most effectively obtain the power coefficient of those turbines?" The results show that the 2D kepsilon simulations approximate the experimental data the closest, but that the difference between the simulations and the experiment in the komega model can be explained by the simulations being 2D and the experiment being 3D. To improve the results and verify that a 3D simulation indeed does produce better results, another study should be conducted in 3D when more computation power is readily available. As to what considerations should be taken into account: Meshing determines largely how long a simulation will take and to what accuracy a result can be calculated. The sliding mesh method should be used to decrease calculation time, so that the mesh does not need to be recalculated every time step. The orders of the different variables should be set to second order accuracy for the pressure p, the momentum, the turbulent kinetic energy k and the time derivative. The second order calculation of these variables results in a significant increase in accuracy of the simulations relative to simulations conducted in first order accuracy. For further improvement a research could be set up that involves both the experimental and the CFD part. That way the practical limitations to building a wind turbine could be modelled in the simulations, possibly resulting in even better synergy between the measurements and the simulations.9CFD; Simulation; Vertical Axis Wind Turbine; Optimizationenbachelor thesisApplied SciencesChemE/Chemical EngineeringTransport Phenomena)uuid:0da53df4df774afd9bc089ccca593719Dhttp://resolver.tudelft.nl/uuid:0da53df4df774afd9bc089ccca593719gThermal Simulation of Low Concentration PV/Thermal System using a Computational Fluid Dynamics Software
Stylianou, S.Smets, A.H.M. (mentor)bCogenra company has created a low concentration PV/Thermal system that produces both thermal energy and electricity simultaneously, mainly for commercial and industrial applications. The realization of a system that combines photovoltaic modules, solar thermal collectors and concentrating mirrors makes it a complicated system to study. So far, only simple studies have been made on Cogenra s LCPVT system including a 2dimensional model. In this project, the possibility of using a Computational Fluid Dynamics software for analysing the low concentration PV/Thermal system of Cogenra has been studied. The CFD software Ansys Fluent has been used, in which a model was created in accordance to the Cogenra LCPVT system. After validating the results, the model has been used for analysing the system s performance under various conditions in order to realize the system s losses. Furthermore, due to the numerous components of the system, the analysis of the LCPVT system becomes a multivariable problem. For this reason, three main parameters (mass flow rate, optical concentration, PV type) that affect the system s performance has been chosen and studied in order to improve the system s overall efficiency. Since the system has both electrical and thermal outputs, an equivalent efficiency was < determined to express the two different efficiency terms. For the purpose of comparing the performance of the LCPVT system with the traditional photovoltaic modules, one other simple model was created in Ansys Fluent. This model has also been simulated under the same conditions as the Cogenra system in order to observe the difference in output between the LCPVT system and the photovoltaic modules. The low concentration PV/Thermal system has also been compared with other solar thermal systems such as a PV/Thermal system, a Concentrated Thermal system and a simple Solar Thermal System.ECPVT; CFD; thermal; electrical; efficiency; concentration; simulation
master thesis
20160830Electrical Sustainable EnergySustainable Energy Technologies)uuid:bfb6a4c71c60472c858ebc519fc46521Dhttp://resolver.tudelft.nl/uuid:bfb6a4c71c60472c858ebc519fc46521WInfluence of Engineered Roughness on the Flow Instabilities in a Centrifugal Compressor
Kapoor, P.'Pecnik, R. (mentor); Javed, A. (mentor)9 Centrifugal compressors form an integral part of automotive turbochargers. Strict emission regulations and increased engine downsizing in the automotive industry are pushing the turbocharger centrifugal compressors to provide the charged air at very lower mass flow rates. However, at low mass flow rates, the operation and application of centrifugal compressors for a given turbocharger is limited by the fluid dynamic instabilities. These flow instabilities cause the compressor to enter the state of stall and/or surge. Stall defines the lower limit of stable operating range for the compression system. The suppression of this flow instability is a key research focus of turbomachinery aerodynamics. In this thesis, an attempt has been made to reduce the compressor instability and subsequently increase the surge margin by the application of engineered roughness on different compressor parts. The analysis has been split into two parts. Firstly, CFD analysis have been carried out on several test cases to validate the reliability of the commercial solver. The comparison of the CFD results with the available reference data shows a reasonable agreement in terms of prediction of laminar to turbulent transition location and influence of wall roughness on transition onset location. Secondly, a detailed CFD analysis of the centrifugal compressor has been made in two phases. In the first phase, the overall compressor performance has been simulated from stall to choke over a specific turbocharger rotational speed. In the second phase, a parametric roughness study has been carried out by considering the effects of predefined wall roughness and localized roughness strips on the unstable flow in the compressor at low mass flow rates. For the wall roughness evaluation, the roughness has been applied on the impeller blades, impeller shroud, diffuser and inlet shroud. On the other hand, for the localized roughness strips, roughness have been defined at specific locations at the main blade suction side and the diffuser shroud. The analysis reveals a reduction in flow instability in the compressor domain. The steady state simulations show a significant improvement in the flow structure of the diffuser in terms of the reduction in flow reversal. Furthermore, an improvement in the impeller is observed by reduction in the low momentum wake region.ICFD; Centrifugal Compressors; Performance Optimization; Surface Roughness
20160114.Mechanical, Maritime and Materials EngineeringProcess and Energy Technology)uuid:267beaa58b624e58b1adff94355c68f9Dhttp://resolver.tudelft.nl/uuid:267beaa58b624e58b1adff94355c68f9ODesign and Analysis of an Installed Pusher Propeller with Boundary Layer InflowVan Arnhem, N.*Lv, P. (mentor); Veldhuis, L.L.M. (mentor)bBoundary Layer Ingestion is an integrated propulsion concept in which a propulsor operates in boundary layer flow instead of the free streamflow with the goal to reduce the fuel flow for a given operating condition. The objective of this thesis is to obtain a better understanding of the power benefit of an installed pusher prope< ller at the aft fuselage by designing the aerodynamic shape of the propeller and validate the design by means of CFD simulations. A propeller analysis tool for uniform inflow (UI) and nonuniform inflow (NUI) named NXROTOR is developed using the lifting line code XROTOR in combination with XFOIL to calculate airfoil properties. The tool is validated using experimental results and results from CFD simulations of uniform inflow propellers. NXROTOR shows good agreement of the trend of the CTJ and CPJcurves but a constant over prediction of both thrust and power with respect to experimental data is observed and several deviations are explained. A series of CFD simulations in ANSYS Fluent using a reduced wedge shaped domain of one blade of the N250 propeller are performed for several advance ratios including a grid refinement study. Minor deviations between a transient and a steady simulation are found and the steady method is chosen based on computational cost. The trends of NXROTOR in terms of CTJ and CPJ compare well with the CFD simulation with a constant over prediction of the performance quantities by NXROTOR. These over predictions are also noticeable in the radial distributions of thrust and torque with slight over predictions in the high loaded region on the blade. For a moderate advance ratio of J = 0.79 the thrust and power are over predicted by 5.25% and 3.67% respectively. A comparison with the standard kapppaomega SST turbulence model and the SST model with low Reynolds number correction is made. The radial flow on the propeller blade is shown to be quite significant and varies along the blade and shows good agreement with the distribution of bound circulation and the resulting trailing vorticity. A design procedure is developed in which the propeller shape is optimised using shape functions to describe the pitch and chord distribution and a NACA four series airfoil is used to limit the number of design variables for a gradient based optimisation algorithm in Matlab environment. The interaction effects are assumed to be determined apriori and a tapered aft fuselage and the pressure field induced by the fuselage are neglected. Input quantities for the design routine include an inflow field from CFD analysis, the design advance ratio and a thrust requirement. The design objective of all optimisations is minimum power. For the reference design case an axisymmetric body from ESDU is subject to CFD simulations to obtain the inflow profile and fuselage drag for the isolated and installed configuration. Interference effects are approximated using an Actuator Disk (AD) model at the predefined location of the propeller with a pressure jump equal to the defect in total pressure in the boundary layer based on findings from previous research. An 11% increase of drag is found for the equilibrium condition which is primarily due to increased pressure drag. Larger pressure jumps show only a marginal increase in drag. In a comparison study, the number of blades is set to four, an advance ratio of J = 1.50 is chosen and in combination with a radius equal to 99% of the total gage pressure of the undisturbed air yields a tip Mach number of around 0.50. The optimisation results show that the NUI propeller requires 6.93% less power compared with the UI propeller despite the 11% higher thrust. The thrust distribution of the NUI propeller shows a significant increase in thrust in the lowaxial velocity region towards the root and the maximum thrust is shifted inboard. The ratio of thrust to power dT/(dQ Omega) along the propeller blade shows a constant distribution for the UI propeller, while the NUI propeller has a smooth increasing distribution towards the root. This distribution shows that thrust requires a relatively low power when the local axial velocity is relatively low. It is found that this is the main benefit of positioning a propeller in the boundary layer. The bound circulation distribution shows a shift towards the root compared with the distribution of the uniform inflow propeller which is the result of the optimised propel< ler shape which benefits from the favourable thrust to power ratio in the inner radii. The NUI propeller has a significant increased chord compared with the optimal UI and also a higher lift coefficient distribution. The local efficiency defined as eta = dTVa/(dQ Omega) with Va as the local inflow velocity. Optimal UI propellers have a constant efficiency distribution, but the NUI propeller shows a decreasing trend towards the root which is also found in literature. The trend of lower local efficiency is also found when an optimisation for minimum power is performed using a radially varying actuator disk with the same inflow and thrust requirement as for the full blade propeller. Additional analysis on the NUI propeller include a comparison of off design conditions and additional optimisations are performed to quantify the effect of the number of blades, radius and advance ratio. The optimised NUI propeller in the installed configuration is simulated using CFD. NXROTOR over predicts the thrust and power by 4.15% and 4.71% respectively compared with the CFD simulation, which are deviations of the same order as the N250 simulation. The thrust to power distribution shows good correspondence. In the root region this ratio is under predicted by NXROTOR which is expected to be the result of a large pressure and velocity gradient at the junction of the spinner and propeller surface resulting in a region of recirculation. Also the blockage effect of the tapered spinner results in larger angles of attack in the root region. The outer region shows trailing edge stall which is found to be primarily due to the coarse mesh in that region. Improved results are obtained when NXROTOR uses airfoil data obtained from twodimensional CFD analysis of a particular airfoil section. The kappaomega SST model with low Reynolds number correction shows almost exact agreement with XFOIL. The standard turbulence model shows a decambering effect and an earlier stall behaviour. The remaining deviations between NXROTOR with approximated CFD airfoil properties are expected to originate from the radial flow on the blade and the variation in circulation in chordwise directions which are not simulated in NXROTOR. Both the externally induced radial flow by the tapered aft fuselage and the self induced radial flow are expected to result in a decambering of the airfoil due to the influence on boundary layer growth as well as a reduced chordwise velocity resulting in a locally lower dynamic pressure experienced by the airfoil contour. The interference effects of the propeller onto the fuselage are compared with the Actuator Disk (AD) approximation. An over prediction of 0.74% of the drag by the AD model of the fuselage excluding spinner is observed. Downstream of the full blade simulation the pressure is rapidly decreased to a low finite value at the aft end of the spinner. This is the result of the finite bound circulation at the propeller root which releases a strong trailing vortex from each blade. These vortices combine into a strong axial vortex which induces a strong tangential velocity and therefore in a low pressure acting on the spinner. A slipstream analysis is performed of circumferentially averaged flow quantities in radial direction at a plane behind the propeller and the axial development of several averaged flow quantities is shown. Several recommendations for future work are formulated to improve the propeller design, improve the design procedure, reduce the interference effects and increase the power benefit of the nonuniform inflow propeller.vBoundary Layer Ingestion; BLI; pusher propeller; CFD; interference effects; design optimisation; integrated propulsionAerospace Engineering<Aerodynamics, Wind Energy, Flight Performance and Propulsion!Flight Performance and Propulsion)uuid:de63784a11d04cb39e27b32f92415c32Dhttp://resolver.tudelft.nl/uuid:de63784a11d04cb39e27b32f92415c32Determination of the nonlinear roll damping of a barge by means of viscous flow simulations: An advanced approach to compute roll damping characteristics with CFDPSta< mpoultzoglou, Tasos (TU Delft Mechanical, Maritime and Materials Engineering)van der Heul, Duncan (mentor); Metrikine, Andrei (graduation committee); Weustink, Roy (graduation committee); Qu, Yang (graduation committee); Delft University of Technology (degree granting institution)aFor many types of floating structures, roll motion is the most important wave induced motion. More specifically, roll motion is of high significance for barges, in order to correctly predict the acceleration of the structure caused by roll motion and determine the procedure of seafastening and offloading. The correct forecast of roll motion requires accurate estimation of roll damping. At the same time, the main sources of roll damping are the creation of waves, skin friction and creation of vortices (eddies), due to roll motion. For this reason, the physics of the problem cannot be fully described by a flow model based on potential theory, as it is highly dependent on vorticity and viscous effects. As a consequence, potential theory algorithms underestimate roll damping, which results in over prediction of roll motion and thus in a conservative estimate of workability. It is also important to mention that the vorticity and the viscous effects, lead to nonlinear damping characteristics. More specifically, the roll damping moment is controlled by its odd numbered harmonics, the first of which is dominant. Consequently, it is common to express the damping moment in an equivalent linearized form, equal to the first harmonic.<br/>During the last years simulating the flow around the rolling vessel using viscous flow algorithms is becoming more and more popular. Using CFD (Computational Fluid Dynamics) is a cheap and fast way to create a numerical wave tank and perform numerical decay tests, forced roll simulations, and roll response simulations in regular waves. Of course the underlying algorithm should be validated, before any commercial or scientific use. In this thesis, roll damping will be estimated by performing virtual forced oscillation tests, and the computed roll damping coefficients will be presented as a function of roll amplitude. The virtual forced oscillation tests have been performed with the opensource CFD software OpenFOAM. Additionally, numerous numerical experiments are performed in order to choose the optimum discretization schemes, turbulence model, mesh and time configurations. Finally, Ikeda s experimental data is used in order to validate the viscous flow algorithm and the numerical model. Two methodologies have been used in order to determine the nonlinear roll damping. In the first case the free surface is included in the viscous flow model, using the Volume of Fluid method. The second approach disregards any free surface effect and the total damping is calculated as a superposition of viscous (by viscous flow algorithm) and wave (by potential theory algorithm) damping. Both approaches are compared with Ikeda s experimental data and it is concluded that both methods are able to capture accurately the linearized roll damping coefficients, for various amplitudes. Finally, viscous flow simulations of the full scale barge, in order to calculate the roll damping coefficients, are notably time consuming. For this reason, in order to reduce the computational time, the methodology which neglects the free surface effect, is chosen, as it is a good compromise between time efficiency and accuracy.BRoll damping; Roll motion; CFD; Viscous flow simulations; OpenFOAM
20220905!Offshore and Dredging Engineering)uuid:f9a658fe9d0b4a1d808a908698c8e8c2Dhttp://resolver.tudelft.nl/uuid:f9a658fe9d0b4a1d808a908698c8e8c2@Analysis of an over the wing based distributed propulsion system3Khajehzadeh, Arash (TU Delft Aerospace Engineering)oVeldhuis, Leo (mentor); Hulshoff, Steven (mentor); Delft University of Technology (degree granting institution)This project aims to investigate propeller wing interaction of an over the wing (OTW) based distributed propulsion (DP) system with the addition of a secondary wing. This research explored the opportun< ities that OTW configuration provides to DP concept, such as improving the lifting performance of system by inducing the flow over the upper surface of wing. Secondary wing is oriented in a biplane configuration above propeller and intends to improve propeller s performance by decreasing propeller s inflow velocity. This research aimed to investigate the influence of shape and position modification on the system s aerodynamic performance and overall propulsive efficiency. The influence of spacing between the wings, position of propeller, secondary wing s angle of attack and secondary wing s initial lift coefficient on aerodynamic performance and overall propulsive efficiency of DP system was investigated in this project.<br/>The analysis of this particular configuration was performed with the help of Euler calculations. Eliminating viscous effects from the analysis, reduced the demanded computational cost of this study and could help this research to perform a broader investigation. The influence of isothermal flow condition on Euler calculation was examined to reduce the computational cost of optimization study; This study showed that the drag coefficient of system is sensitive to this assumption and that only by simulating the initial and final points of the optimization study with adiabatic flow condition, the use of isothermal flow condition could be justified. The secondary wing s shape was optimized to study the influence of its shape variation on the system s aerodynamic performance. The first optimization study aimed to decrease the drag coefficient of system, and as a result, overall propulsive efficiency of was improved. The second optimization study aimed to improve the lifting performance of system. As a result, the drag coefficient of system significantly increased, however, the overall propulsive efficiency was again improved. The flow decelerated between main wing and secondary wing since the lifting performance of secondary wing improved, which increased the propeller s thrust according to blade element method (BEM) calculation.<brpropeller wing interaction; numerical analysis; over the wing; Optimization; CFD; cfx; distributed propulsion; actuator disc)uuid:21c1e156c8df4ab6863025c6170d3bc6Dhttp://resolver.tudelft.nl/uuid:21c1e156c8df4ab6863025c6170d3bc6FNumerical and Experimental Study of Airfoils with Porous Trailing Edge/Tamaro, Simone (TU Delft Aerospace Engineering)Ragni, Daniele (mentor); Avallone, Francesco (mentor); Shen, Wen Zhong (mentor); Technical University of Denmark (degree granting institution); Delft University of Technology (degree granting institution)The addition of porous material to the trailing edge of airfoils was studied numerically and experimentally. The purpose of this work is to inspect whether the penalization method can be successfully used to describe flows through porous media. In the numerical study, the penalization method was applied to the incompressible NavierStokes equations. The flow has been studied with turbulent, unsteady simulations. The SIMPLE algorithm and a mixed scale model for LES were used in EllipSys2D. The code was validated with a test case on a flat plate. In this case, porous inserts are capable of suppressing vortex shedding by mitigating the pressure discontinuity at the trailing edge. An agreement with wind tunnel data could be observed both in the wake and in the boundary layer. The code was then applied to study a NACA0018 airfoil at zero angle of attack. In this case, porous materials reduce the energy content of turbulent signals in the wake. The power spectral density of pressure fluctuations in correspondence of the porous insert is reduced in the whole spectrum. The same airfoil was considered with the addition a single roughness element to trip the boundary layer at 20% of the chord. The height of the step was chosen from the results of a sensitivity analysis. The presence of the trip triggers turbulence transition successfully. Pressure drag for the porous case was found to be dependent on the presence of the trip. In the experimental stud< y, measurements were taken on a symmetric NACA0018 airfoil at zero angle of attack. Hotwire anemometry and surface microphones were used to measure the flow. Porous inserts with different permeabilities have been studied. Porous materials at the trailing edge increase the shear of velocity profiles in the boundary layer and reduce the power spectral density of pressure fluctuations in correspondence of the trailing edge. According to this study, the penalization method is a promising tool to study flows through porous media. Porous Media; Aeroacoustics; CFD8EWEM (European Wind Energy Masters)  Rotor Design Track#EWEM (European Wind Energy Masters))uuid:b403c8ca64634e8a9e14d6a76fddd297Dhttp://resolver.tudelft.nl/uuid:b403c8ca64634e8a9e14d6a76fddd297YWind Tunnel Interference Effects on Tilt Rotor Testing Using Computational Fluid DynamicsKoning, W.J.F.Vos, R. (mentor)I Experimental techniques to measure rotorcraft aerodynamic performance are widely used. However, most of them are either unable to capture interference effects from bodies, or require an extremely large computational budget. The objective of the present research is to develop an XV15 Tilt Rotor Research Aircraft rotor model for investigation of wind tunnel wall interference using a novel Computational Fluid Dynamics (CFD) solver for rotorcraft, RotCFD. In RotCFD, a midfidelity URANS solver is used with an incompressible flow model and a realizable k? turbulence model. The rotor is, however, not modeled using a computationally expensive, unsteady viscous bodyfitted grid, but is instead modeled using a blade element model with a momentum source approach. Various flight modes of the XV15 isolated rotor, including hover, tilt and airplane mode, have been simulated and correlated to existing experimental and theoretical data. The rotor model is subsequently used for wind tunnel wall interference simulations in the National FullScale Aerodynamics Complex (NFAC) at NASA Ames Research Center in California. The results from the validation of the isolated rotor performance showed good correlation with experimental and theoretical data. The results were on par with known theoretical analyses. In RotCFD the setup, grid generation and running of cases is faster than many CFD codes, which makes it a useful engineering tool. Performance predictions need not be as accurate as highfidelity CFD codes, as long as wall effects can be properly simulated. For both test sections of the NFAC wall interference was examined by simulating the XV15 rotor in the test section of the wind tunnel and with an identical grid but extended boundaries in free field. Both cases were also examined with an isolated rotor or with the rotor mounted on the modeled geometry of the Tiltrotor Test Rig (TTR). A quasi linear trim was used to trim the thrust for the rotor to compare the power as a unique variable. Power differences between free field and wind tunnel cases were found from 7 % to 0 % in the 80 by 120Foot Wind Tunnel test section and 1.6 % to 4.8 % in the 40 by 80Foot Wind Tunnel, depending on the TTR orientation, tunnel velocity and blade setting. The TTR will be used in 2016 to test the Bell 609 rotor in a similar fashion to the research in this report.4tilt; rotor; CFD; RotCFD; wind; tunnel; interference
2016010137.427181, 122.061905)uuid:4e756c0704c8464f9c602c4efcab753cDhttp://resolver.tudelft.nl/uuid:4e756c0704c8464f9c602c4efcab753c*CFD Modeling of Abdominal Aortic AneurysmsVan Kruchten, T.J.G._Poelma, C. (mentor); Westerweel, J. (mentor); Pourquie, M.J.B.M. (mentor); Kalkman, J. (mentor)
An abdominal aortic aneurysm (AAA) is an excessive localized swelling of the abdominal aortic wall. AAAs are often lethal when they rupture and constitute a significant health risk in the developed countries. CFD simulations can help predict formation, progression, and rupture of AAAs by the use of hemodynamic parameters such as the Oscillatory Shear Index (OSI) that indicates the oscillatory behavior of the wall shear stress vector at the aneurysm wall. Ideally, it is e< nvisioned that the risk of rupture of a particular aneurysm can be estimated by patientspecific parameters that are collected from a patient with minimal effort and to classify the aneurysm into different categories that do or do not pose a considerable risk of rupture. The main objective of this thesis then aims to focus on the underlying flow mechanisms in aneurysm flow and tries to take the first steps towards an abstract aneurysm model by means of a proof of principle regarding the prediction of formation, progression, and rupture locations within an aneurysm, based on simple patientspecific input parameters. To this end, CFD simulations of pulsatile blood flow in an abstract abdominal aortic aneurysm (4A) model are performed for the independent ranges of mean Reynolds number 300?Re m ?1200, Womersley number 15.1 ? ? ? 27.7, aneurysm length ratio 2.6 ? Le ? 5.3, and aneurysm diameter ratio 1.81 ? Di ? 2.55. The effect of the variation of the input parameters on the oscillatory shear index (OSI) is regarded and quantified by the 4A surfaceaveraged OSI. General trends show that the average OSI decreases 15 percent over the mean Reynolds number range of 300 ? Re ? 1200, increases 16 percent over the Womersley number range of 15.1 ? Wo, ? ? 27.7, increases 11 percent over the aneurysm length ratio range of 2.6 ? Le ? 5.3, and decreases 5 percent over the aneurysm diameter ratio range of 1.81 ? Di ? 2.55, indicating that the mean Reynolds number and the Womersley number have the largest influence on the average OSI for the 4A. The fluctuating wall shear stress vector at the stagnation points of the vortices present in the aneurysm is pointed out as the origin of the high OSI valued axisymmetric rings found on the surface of the 4A. Additionally, there exists an inverse relation between the surfaceaveraged OSI and the turbulent to periodic kinetic energy ratio, demonstrating the importance of the periodic components. Pulsatile blood flow simulation is also performed on a patientspecific aneurysm geometry and compared with the 4A case with Le = 4.6. The dissimilar flow and OSI results imply that the input parameters alone do not permit to make statements about the OSI values in the patientspecific aneurysm based on the 4A.YCFD; abdominal aortic aneurysm; oscillatory shear index; pulsatile flow; womerlsey numberProcess and EnergyFluid Mechanics)uuid:dbd143ca790844479cf0193011e81984Dhttp://resolver.tudelft.nl/uuid:dbd143ca790844479cf0193011e819848Vapor formation in a flooded lubrication turbine bearing
Roks, G.M.Pecnik, R. (mentor)In this research a case study is performed to vapor formation in a turbine bearing. As a result of centrifugal forces in the rotating lubricant flow, low pressure regions occur in the bearing chamber. It is expected that vapor starts to form above a certain rotor speed. The formation of gaseous lubricant is presumed detrimental to the operation of the bearing. The specific case in this research is the turbine bearing of a commercially available Organic Rankine Cycle (ORC), operating at rotor speed up to 430 Hz. The ORC is a thermodynamic process, applied to recover energy from waste heat. In the machine of the case study the cycle medium is toluene. For technical reasons toluene is also used in the turbine bearing for lubrication purposes. In order to model the flows and heat transfer in the bearing, a two dimensional CFD model is created. The model output is validated against experimental data. After analysis of the results, a clear operating limit is found at 450 Hz. At this condition just outside the normal operating regime of the turbine, vapor formation sets on. Design changes proposed by the company are taken into account as well. The main alteration, moving the feed hole of the bearing chamber inward, proved to be effective. Simulation results show that bearing pressure at high speed is increased by more than 1 bar. The limit of the rotor speed in the alternative design is above 500 Hz.@bearing; turbine; lubrication; cavitation; tilting pad; ORC; CFD
20190131Process & Energy)uuid:< 8b4d06026045437d816709203f8ac6aaDhttp://resolver.tudelft.nl/uuid:8b4d06026045437d816709203f8ac6aa5Prediction of the added resistance in waves using CFDIHulsbergen, Bas (TU Delft Mechanical, Maritime and Materials Engineering)Mikelic, Andrea (mentor); Akkerman, Ido (mentor); Crepier, Pierre (mentor); Delft University of Technology (degree granting institution)Shipyards these days see an increase in customers that specify combined speed and seakeeping ability design requirements. This requires the shipyard to make a prediction of the additional installed power required to maintain a certain speed when waves are encountered. The additional required installed power is directly related to the average extra resistance that the vessel is subjected to when it s sailing in waves. This extra resistance is known as the timeaveraged added resistance in waves. In the maritime industry Computational Fluid Dynamics (CFD) is increasingly used for resistance predictions as it promises cheaper and faster predictions than model testing. The result does come without the comforting truth of the towing tank. In this study, the applicability of CFD for the estimation of the timeaveraged added resistance in regular head waves is researched by assessing the error and uncertainty of the solution. For fast sailing vessels, no standard procedure for the estimation of the timeaveraged added resistance in waves using CFD has yet been developed. Therefore the secondary research objective is to establish such a procedure. For this research the resistance predictions are done for the Fast Displacement Ship (FDS) hull form. Extensive research was conducted on this hull form by the Cooperative Research Ships (CRS) organisation. Their model tests results are used for the validation of the solution. The discretisation error is determined through a procedure developed by L.Ea and M.Hoekstra [25] which is based on a grid refinement study. The timeaveraged added resistance is estimated by simulating the vessel both in calm water and waves. The timeaveraged calm water and total resistance in waves are determined from these simulations. The timeaveraged added resistance estimate is then calculated by subtracting the calm water resistance from the total resistance. First a grid topology is optimised to simulate the incoming waves as well as the vessels response to them accurately and efficiently. Grid sensitivity studies of the simulation of incoming waves as well as simulations of the static vessel in waves and the vessel subjected to forced motion are used to determine an efficient topology. To determine if the vessel s response is accurate, it is compared to the solution from potential flow code solver PRECAL. The comparison proved that accurately propagating waves and accurate vessel response to the waves and motions are achieved on grids with less than 3 M cells in total. Verification estimated an uncertainty that varies between 0.5 % and 1.3 % for the timeaveraged total resistance in waves and between 15.1 and 36.2% for the timeaveraged calm water resistance on grids with a total number of cells ranging between 1.3 and 6.6 M. Comparison with the results from the model test revealed that an error of 1.4 % was present in the timeaveraged added resistance estimate. This error is smaller than the uncertainty margin of the model test result. Using the proposed method, the timeaveraged calm water resistance estimate didn t converge well, resulting in a large discretisation uncertainty. As the added resistance prediction is dependent on the calm water resistance prediction, it is also affected by this uncertainty. Therefore it s concluded that the proposed method for the estimation of the timeaveraged wave added resistance using CFD is not yet applicable in its proposed form. However, by using the proposed method, it is possible to estimate the timeaveraged total resistance in waves accurately and efficiently. Therefore it s concluded that further research is required to improve the uncertainty present in the timeaveraged added resistance due to the unc< ertainty seen in the calm water resistance for the used grids.CFD; ReFRESCO; added resistance; regular waves; 2D waves; radiated waves; diffracted waves; FDS; discretisation uncertainties; verification; validationOffshore Engineering)uuid:b7a76cb18dcc47e18e6f0b46b78f3f50Dhttp://resolver.tudelft.nl/uuid:b7a76cb18dcc47e18e6f0b46b78f3f50Ignition modeling in methaneoxygen rocket engines: Design and modeling of a methaneoxygen rocket engine igniter using reacting flows with computational fluid dynamics2Akkermans, Christ (TU Delft Aerospace Engineering)Zandbergen, Barry (mentor); Gill, Eberhard (graduation committee); van Zuijlen, Alexander (graduation committee); Delft University of Technology (degree granting institution)]The influence of using different chemistry model and different chemical kinetics schemes on modeling combustion inside a rocket engine igniter are investigated. An igniter for a LOX/CH4 rocket engine is designed and a CFD model is developed which is used to model the designed igniter. Verification and validation methods are discussed.<br/><br/><brqrocket; engine; ignition; CFD; chemistry; oxygen; methane; igniter; model; computational; turbulence; finite rate)uuid:7a0b85902b9744818b0d0b832189ad88Dhttp://resolver.tudelft.nl/uuid:7a0b85902b9744818b0d0b832189ad88qWallModelled Large Eddy Simulation of fully rough nonuniform flow for the purpose of predicting stone stability=Nikolaidou, Lina (TU Delft Civil Engineering and Geosciences)Uijttewaal, Wim (graduation committee); Hofland, Bas (mentor); Bricker, Jeremy (graduation committee); O'Mahoney, Tom (mentor); Jacobsen, Niels G. (graduation committee); Delft University of Technology (degree granting institution)>Granular bed protections are a common measure to mitigate scour of the sand bed around hydraulic structures. In view of marching towards costeffective solutions to tackle erosionrelated problems, it is important to accurately predict the loads exerted on the bed and come up with the needed rock grading. To that end, a 3D eddy resolving modelling technique could help in formulation of a new stability formula, based on extreme local flow conditions It is the aim of this thesis to build a hydrodynamic numerical tool, able to predict governing mechanisms in stone stability. Special attention is paid on the way of representing a rough boundary and predicting wall and free turbulence. That being said, a WallModelled Large Eddy Simulation (WMLES) is suggested and is build using the opensource CFD toolbox OpenFOAM. Firstly, simple open channel flow case simulations are setup, to evaluate the performance of rough wall functions in the LES environment. Secondly, a backwards facing step kind of flow from the experimental study of Jongeling et. al (2003) is simulated and the performance of the numerical model is evaluated based on the experimental results. Finally, the suitability of this method for the target applications of this study is discussed, including a comparison with previous numerical studies of this research area.CFD; OpenFOAM; large eddy simulationon; wall modelled large eddy simulation; turbulence modelling; stone stability; flow over fully rough bed; LES; WMLESCivil Engineering)uuid:db908fbf66ca4b0286883d5b35b8deb2Dhttp://resolver.tudelft.nl/uuid:db908fbf66ca4b0286883d5b35b8deb2bSimulation & Analysis of the Aerodynamic Interactions between Distributed Propellers and WingsYFischer, Jan (TU Delft Aerospace Engineering; TU Delft Flight Performance and Propulsion)jOrtun, Biel (mentor); Veldhuis, Leo (mentor); Delft University of Technology (degree granting institution)
The recent advances in complementary technologies have given a new impulse in the research activities related to distributed propulsion. Light electric engines, whose performance is rather insensitive to scaling, improved battery capacities and autonomous flight all contribute to this trend. Promising concepts featuring electrically powered distributed propellers along the wings are expected to improve aeropropulsive efficiency, highlift capabilit< y, vertical takeoff & landing capability and operating costs. Problems associated to the exploitation of these benefits are, among others, the many new design freedoms and the associated growing design space. The design space exploration requires new tools that are capable of dealing with lifting surfaces, interacting strongly with multiple propellers. Furthermore, the tool's ability to include viscous and thicknesseffects is crucial for understanding and assessing the new vehicle architectures. The goal of this thesis is twofold and starts with the implementation and validation of a bodyforce approach to model propellers in a (U)RANS simulation by volumetric source terms. A coupled liftingline code, being executed in the extracted RANS flowfield, provides the propeller blade forces that are acting on the fluid. The validation is performed by comparing the simulation method on closely coupled propellerwing configuration with a meshedpropeller URANS simulation. The comparison shows a very good agreement for the two different approaches. A further comparison to an experimental setup of an overthewing propeller provides further confidence in the methods capability of modelling propellers in the close vicinity of a lifting surface. The second goal is the better understanding of the flow phenomena for selected promising configurations. On the one hand, wingtip mounted propellers in pusher and tractor arrangements are compared for nontwisted and ideallytwisted wings. The thrustdrag balanced configurations show that the total required shaft power is less for a pusher arrangement in case of a nontwisted wing. For the ideallytwisted wing, the two arrangements are equally good in reducing required shaft power. The tractor arrangements have a wing drag reduction of 10% while the pusher arrangements show an increase in propulsive efficiency of 89%. The isolated liftingline simulations showed that the tool is capable of capturing the trends but that it cannot achieve the accuracy of a viscous analysis. On the other hand, liftaugmenting propellers have been installed in front of the leading edge as well as on different chordwise positions over the wing. The overthewing installed propellers showed the highest increase in lift coefficient (up to 107%) whilst strongly decreasing wingdrag. A more upstream positioned propeller achieved a higher drag reduction but a downstream positioned propeller showed higher lift augmentation. The leading edge propellers had the highest lift increase (60%) when being contrarotating. For the highlift propeller configurations, the liftingline code was able to capture major trends. The performed work shows for the first time a viscous analysis on how wingtip mounted propellers can improve the aeropropulsive efficiency of a configuration and compares trimmed pusher and tractor arrangements. It was further shown, how the new bodyforce approach can lead to very accurate viscous simulations of distributed propellerwing configurations with relatively low computational effort.Edistributed propulsion; aerodynamics; CFD; propellerwing interaction
20171208)uuid:4c9d0463b31841ffa7ae94a139464b4eDhttp://resolver.tudelft.nl/uuid:4c9d0463b31841ffa7ae94a139464b4eLFlow reconstruction using Bayesian inference with model reduction techniquesLD' Onofrio, Federica (TU Delft Aerospace Engineering; TU Delft Aerodynamics)VDwight, Richard (mentor); Delft University of Technology (degree granting institution)(RANS equations are nowadays widely used in industry because of their affordability in terms of computational costs. They reached a high level of complexity, as they involve systems of nonlinear partial differential equations. However they still lack of generality as they are based on closure coefficients determined from fundamental real flow cases. Their accuracy drops when dealing with separating turbulent flows.<br/>There is a specific class of flows that separate after encountering a geometryinduced adverse pressure gradient. Periodic hill flows are viewed as the benchmark case of those, p< resenting typical gross features of this class of flows. The separation point is viewed as one of the characteristic features and its prediction is crucial for a fair downstream flow development representation. In literature it was found that different turbulence models produced different flow solutions, as both separation and reattachment points were badly reproduced.<br/>The aim of the project is therefore to reconstruct the solution under uncertainty in separation location and turbulence closure coefficients. A statistical calibration of the uncertain parameters is performed using Bayesian inference; then Markov Chain Monte Carlo Method (MCMC) is adopted to explore the a posteriori information.<br/>The separation location is controlled using a step, that, located in the area of adverse pressure gradient, forces the flow to separate. The geometry parameters of the step are considered as the uncertain parameters to be calibrated. The effect of separation forcing on the flow solution is studied when adopting different turbulence models. A strong coupling between the step influence on flow solution and closure model adopted has been found. Because of this, LaunderSharma kepsilon model was employed for the rest of the work. The calibration of turbulence model coefficients have been performed starting from existing works. <br/>After providing some validated data relative to a particular location of the flow field and performing parameters calibration, the entire field is reconstructed through posterior predictive distribution.<br/>In general it was found that the results of the calibration depend strictly on the location where the inverse problem is performed. The calibrated value are able to reproduce the flow solution at the inverse problem location but unable to accurately predict the solution at different locations. <br/>The calibration of the closure coefficients only, resulted in a fair prediction of the reattachment point, but a bad representation of the separation point. The inclusion of the step resulted in a slightly better representation of the flow when moving away from the inverse problem location, with particular reference to the vertical velocity profile. This proved that the influence of turbulence closure coefficients is predominant.<br/>The analysis is conducted focusing on the velocity profiles. In fact the eventual aim is to start from planar PIV data and post process them for a threedimensional flow field reconstruction.<br/>Because of a future application in real three dimensional cases and the possible highdimensionality of the uncertain parameter space, a mathematical tool to be used during MCMC iterations is developed.<br/>Instead of calling the CFD solver at each MCMC iteration, the flow solution is computed \textit{offline} through model reduction techniques. POD + I and Isomap + I have been tested for this purpose; the first one being used for linear spaces and the second for the nonlinear ones. At the end Isomap + I was also used for examining the intrinsic metric of the flow solution space when varying the turbulence closure coefficients and the step parameters.<br/>The model reduction techniques seemed to work well and this allowed to perform the calibration at cheap computational costs.Bayesian Inference; turbulence modeling; Reduced order model; MCMC; POD; Isomap; kepsilon; CFD; inverse problem; Uncertainty Quantification)uuid:6c590444c6824f1ca271a053c0dc3ba7Dhttp://resolver.tudelft.nl/uuid:6c590444c6824f1ca271a053c0dc3ba7*CFD Study of Piston Cooling Using Oil JetsCVenkatesh, Vishal (TU Delft Aerospace Engineering; DAF Trucks N.V.)Hickel, Stefan (mentor); Hulshoff, Steven (graduation committee); Gangoli Rao, Arvind (graduation committee); Znaien, Jemil (mentor); Delft University of Technology (degree granting institution)For heavy duty diesel engines, piston temperature control is very important in constructing a successful design that meets the demands of increasing power output and stringent emission regulations. Overheating of piston is avoided with engine oil through spray cooling a< nd gallery cooling processes. In this research project, evolution and disintegration of oil jet used for piston cooling by DAF Trucks N.V., is studied numerically using Large Eddy Simulation (LES) and Volume of Fluid (VOF) methodologies. A robust CFD model is built with major focus on the significance of grid resolution in multiphase LES, and is used to reproduce two test cases. This is followed by characterization of physics involved in oil jet breakup through qualitative inspection. Parameters relevant to the two types of cooling techniques are estimated to see the impact of jet development at different flow rates. Based on the results obtained, the need for grid refinement is assessed and performed for certain cases. Finally, turbulent atmosphere within the crankcase is estimated through a separate simulation and its effect on oil jet is investigated. Results show that for oil in quiescent atmosphere jet turbulence is the dominant force and is the primary cause for disintegration. A clear transition to turbulence is captured with increase in flow rate, as the jet behavior is more and more chaotic with droplet formation and spreading. Importance of grid resolution on droplet capturing is recognized and an isolated analysis shows more droplets being captured with fine meshes. The level of refinement necessary to capture all the droplets still remains an open question. With turbulent atmosphere, no significant change in the jets is obtained until primary breakup and the inertial force of the liquid phase is found to dominate the surrounding flow effects. However, secondary breakup is found to be affected, as aerodynamic interactions increase disintegration and spreading, impacting both the cooling techniques. Results obtained from this research work will be used as primary inputs for further studies in the company on spray cooling and gallery cooling (sloshing flow) leading towards optimization of the piston cooling process./CFD; LES; piston cooling; Oil jets; Jet breakup
202407014Aerospace Engineering  Aerodynamics and Wind Energy)uuid:3b257aa35168441f9b168fad291edc1eDhttp://resolver.tudelft.nl/uuid:3b257aa35168441f9b168fad291edc1eNTurbulence in Aneurysms: Numerical Investigation in Abdominal Aortic AneurysmsfRawat, Digvijay (TU Delft Mechanical, Maritime and Materials Engineering; TU Delft Process and Energy)Poelma, Christian (mentor); Pourquie, MAthieu (mentor); Kotsonis, Marios (graduation committee); Delft University of Technology (degree granting institution)An aneurysm refers to the local dilation or ballooning of a blood vessel and if left unchecked, almost every aneurysm continues to grow in size until it eventually ruptures. The abdominal aorta is the most common location for an aneurysm to form and an abdominal aortic aneurysm (AAA) can turn lethal in the event of rupture. Since the current clinical practice to classify AAAs is on the basis of their size (diameter), it was decided to investigate the impact of varying geometric parameters viz., the expansion ratio (ER) and the expansion angle (EA), on the pulsatile flow field in a hypothetical axisymmetric AAA (4A) geometry and in particular, investigate the role turbulence. All the combinations of ER and EA lead to a total of 8 cases. The simulations show that a vortex ring is shed just after peak systole in every case. The vortex ring, at inception, is similar for all cases within visual limits. As the ring travels downstream, an azimuthal instability sets in that grows with time. This is the short wave or the elliptical instability usually associated with a vortex pair perturbed by an infinitesimally small amplitude sinusoidal wave. The number of waves formed along the circumference of the ring as a result of the instability varies from case to case but in general, the number of waves decreases as the ER increases. The growth of this instability, along with interaction of the ring with the remnants of the flow field form the previous cardiac cycle, cause it to break down into smaller vortices thereby generating turbulent fluctuations. Further, subtle diff< erences between various cases regarding the vortex ring breakdown mechanism are observed and discussed. Investigations are also carried out to identify recirculating or dead flow regions as in the context of an AAA, such a region indicates the presence an intraluminal thrombus (ILT) which is a commonly found feature in AAAs. An ILT deprives the surrounding arterial tissue of oxygen thereby weakening the aortic wall and increasing the risk of rupture. From the simulations, it could be seen that the inflection regions of the 4A geometry are most prone to the formation of an ILT which is in agreement with the physically observed locations. Finally, the oscillatory shear index (OSI) is investigated to see the influence of turbulence on a hemodynamic parameter. It was found that for the geometries tested, there is no clear corelation between the two. From this thesis, numerous conclusions can be drawn. But perhaps the most important conclusion is that the flow field in an AAA is very complex and sensitive to various input parameters such as the ER and EA. Although flow parameters such as the Reynolds and the Womersley number are kept constant across all the cases, existing literature clearly highlights the dependence of the flow field on them. As such, the desire to be able to formulate general corelation trends between various flow field quantities in AAAs is wishful thinking at best.OCFD; Turbulence; Abdominal Aortic Aneurysm; Oscillatory Shear Indes; blood flow)uuid:75f3e25ab9914e85ab8a736a32383bb5Dhttp://resolver.tudelft.nl/uuid:75f3e25ab9914e85ab8a736a32383bb5VMesh Deformation Using Radial Basis Function Interpolation With Sliding Boundary Nodes.Mathew, Maria (TU Delft Aerospace Engineering)]van Zuijlen, Alexander (mentor); Delft University of Technology (degree granting institution)This thesis studies a novel mesh deformation method based on radial basis functions to improve mesh quality in regions with small wall clearances. For fluidstructure interaction problems or imposed motion CFD problems, the usually fixed Eulerian fluid meshes have to undergo deformation to conform to the deforming structure. In such cases, it is important that the fluid mesh retains a reasonably good mesh quality after deformation, since otherwise it can introduce additional errors to the numerical simulation. A suitably robust mesh deformation algorithm is required for this. A variety of mesh deformation methods exist, and the radial basis function (RBF) interpolation method is one of the most robust among these. However, in cases with large deformations in areas with small wall clearances, this method can still fail and the chances of degenerate or lowquality skewed cells occurring is high. If a poor quality mesh results, the domain might have to be remeshed which is a tedious and computationally expensive procedure. <br/><br/>This thesis aims to combat this potential failure for low wall clearance test cases by modifying the RBF interpolation method to allow for sliding of the usually fixed boundary nodes along the boundary edges/surfaces. The additional degrees of freedom of the boundary nodes reduces skewing in meshes and allows for higher quality meshes in these types of test cases. In this thesis, the sliding method was first created to work on straight edges and planar surfaces, and then further extended to be able to work on any arbitrary surface. Tests were done with both 2D and 3D meshes. <br/><br/>Two sliding algorithm alternatives were tried during this project, with the first one having the sliding conditions built into the RBF interpolation system directly, while the second one performed the sliding in a more indirect manner by using a projection algorithm which ensures that the points remain on the surface. The first direct sliding method results in the interpolation calculations of the spatial directions being coupled to each other, which is not the case with the classical RBF method. This results in an interpolation system that is up to nine times as large (3Nx3N for N boundary points), and much less computationally efficie< nt. The second pseudosliding method continues to use the decoupled system and is computationally faster to use. While it initially appeared to be promising, giving good final mesh qualities at low computational speeds for simple 2D cases, this second method was ultimately found to be less robust than the direct coupled method, failing especially with 3D meshes and irregular mesh boundaries. <br/><br/>The direct sliding RBF method was found to be more robust compared to the regular RBF method. The method gives good results with straight edges and planar surfaces, as well as smooth curves. However, when used with nonsmooth irregular surfaces, the magnitude of sliding was affected, often resulting in almost no sliding. The method should be avoided in such cases as it is unprofitable. Apart from these cases, the method was able to achieve good improvements in mesh quality compared with the classical RBF method. For applications that have smooth surfaces without large surface irregularities, this method can be used to produce high quality deformed meshes.<br/><br/>Despite its robustness, the direct sliding method is not practical to use due to its large interpolation matrix unless some additional efficiency optimisation method is used. In this case, a control point selection algorithm was used such that fewer points were used to find the interpolation, resulting in a smaller system. With this algorithm, the computational time of the method was considerably reduced to being almost comparable with the classical RBF method (also with the same optimisation applied) particular for large test cases. Another method to reduce the total solution time of the direct sliding method is to use a reduced interpolation system that requires coupling of the interpolation coefficients thereby reducing the size of the interpolation matrix. This has only been briefly investigated but shows large improvements in solution time, while still effectively solving the same system without any reduction in control points.<br/><br/>This means that with the new method, much larger displacement steps can be used to arrive at the same or usually better final quality as obtained with the original RBF method, which results in faster computations. Of course, the timestep size is restricted by the maximum step of the CFD simulation itself. If the magnitude of the deformation is not too high, absolute deformation can also be used instead of relative deformation. This means that the interpolation system will need to be constructed and solved only once at the start of the run based on the initial mesh, resulting in large time savings for the deformation calculation. In conclusion, this method has great potential to be used for mesh deformations, since it can result in an improvement in both computation time and mesh quality.\mesh deformation; radial basis functions; sliding boundary; fluidstructure interaction; CFD)uuid:3dea23d8f101495e8050bf8302f2e609Dhttp://resolver.tudelft.nl/uuid:3dea23d8f101495e8050bf8302f2e609JNumerical simulation of the flow in a scour hole due to a translating jet.Visscher, F.J.C.FVan Rhee, C. (mentor); Keetels, G. (mentor); Van der Hout, R. (mentor)Dredging is a very important activity for mankind, for example deepening waterways or creating artificial islands. The Trailing Suction Hopper Dredger (TSHD) is an important dredging vessel, this vessel uses the so called drag head. It is common practice to use water jets in drag heads to create a more efficient dredging process. Some theoretical and empirical models exist which determine for example the van Rhee jet model which predicts the depth and width of the scour hole created by a impinging jet. The performance of these models is limited such that the production is not estimated correctly. Therefore a lot of effort is done to understand the underlying processes. This thesis project is therefore aiming to provide insight in these quantities using numerical simulations of the flow.Erosion; CFD; Jet
20190322Offshore & Dredging EngineeringDredging Engineering)uuid:5f8a549a443b40aa< b9283c2e205526b0Dhttp://resolver.tudelft.nl/uuid:5f8a549a443b40aab9283c2e205526b0Design Optimization for Enhanced Fuel Mixing and Reduced Combustion Instability: Enhancing Swirler Performance of a Small Turbojet Engine Combustor
Venter, P.Visser, W. (mentor)Aeroengine performance is becoming an increasingly regulated aspect in aerospace indus tries with tighter restrictions on emissions, greater expectations for efficiency and thrust as well as broader requirements for the range of the operating flight envelope. With an increasing consciousness toward these factors during the design of combustors, research led development and improvement of every single aspect of the combustor design needs to be considered in this modern era of aerospace technology. A major contributor to such performance enhancement is the design of flow swirlers used to induce central re circulation zones in the primary fuel/air mixing region. In the current study, the effect of modification to a swirler s vane blade angle on mixing effectiveness and combustion stability is investigated, using flow properties such as turbulent kinetic energy, fuel dis tribution and pressure losses as a measure of combustor performance. The study takes a sensitivity analysis approach and makes use of an existing combustor design that acts as a benchmark for verification of results. A cold flow computational fluid dynamics anal ysis is used to test the effect of blade angle modifications based on a cause and effect methodology. The computational fluid dynamics model is validated against experimental data from a similar combustor. It was found that optimal fuel/air mixing occured in a 70? blade angle swirler however large pressure losses and excessive vortex shedding directly behind the center body indicated a strong likelihood of combustion instability. Good fuel atomisation through strong shear layers and excellent pressure recovery seen in a 30? blade angle swirler was accompanied by poor fuel/air mixing. A swirler design featuring 50? blade angles was found to be the optimum, with good fuel atomisation, stable recir culation zones, promising flame anchoring potential, dispersive but orderly homogenous fuel/air mixing and desireable pressure recovery characteristics.xswirler; combustion stability; fuel mixing; CFD; cold flow; Optimisation; K ? ?; turbulence model; vortex; recirculation)uuid:0fd3498571434110965d8f6f1af59c01Dhttp://resolver.tudelft.nl/uuid:0fd3498571434110965d8f6f1af59c01Changing the crosssectional geometry of a bow tunnel thruster: Effects on the performance of the thruster at slow forward motion using Computational Fluid Dynamics
Schaap, T.Van Terwisga, T.J.C. (mentor); Huijsmans, R.H.M. (mentor); Pourquie, J.B.M. (mentor); De Jager, A. (mentor); Kruijswijk, A.B. (mentor),In this master thesis the flow behavior and the performance of a bow tunnel thruster at slow forward vessel motion is studied using Computational Fluid Dynamics (CFD). The study analyzes the flow and the turning ability for a cylindrical crosssection and the effect of changing the crosssection of the bow tunnel thruster. At Royal IHC it is noticed that trailing suction hopper dredgers experience a significant decrease in turning ability, using the bow tunnel thrusters, when trailing at a speed through the water of 5 [kts] in comparison to zero forward speed. Dredgers are often operating at those speeds, use the bow tunnel thruster to keep course and therefore often experience this effect in practice. To study the flow behavior the commercial CFD solver Numeca FineMarine is used. Computations are made on model scale using a simplified wedge model of a container ship (Nienhuis wedge) for validation purpose and of a trailing suction hopper dredger (Hopper wedge). For the Nienhuis wedge multiple numerical studies are performed that focus on the used setup, nonlinear iterations and convergence, time step, actuator disk modeling and first layer thickness. A grid study together with a verification and validation study close the analysis of the Nienhuis wedge. The settin< gs from the Nienhuis wedge are used for the computations with the Hopper wedge. For the Hopper wedge a systematic tunnel crosssection variation is derived and three different shapes are computed at different ship speeds: a circular crosssection (S1010A), flattened crosssection (S0610A) and a streamlined crosssection (S0602A). The flow of the tunnel jet and the flow around the ship are comparable to the flow of a jet in a crossflow. The flow is unsteady and fluctuates. Once the tunnel jet flow leaves the tunnel the flow interacts with the surrounding flow and is bend into the direction of that surrounding flow. A large wake region is visible behind the jet. The velocity ratio $m$ between the ship speed and the tunnel jet speed is an important factor and characterizes the behavior of the flow. At $m$=0.2 [] a strong jet shows only little interaction with the ship flow, while a weak jet at $m$=0.4 [] is largely influenced by the ship flow. For the Nienhuis wedge a grid study shows large numerical uncertainties. The verification and validation study shows that the computations are qualitative valid and quantitative invalid. Quantitative comparison between two different model tests shows discrepancy in the obtained side force on the wedge. However the quantitative results of this study do agree with a full scale CFD study. In both this CFD study and the full scale CFD study the hub and strut of the thruster are not modeled. It is expected that this has an effect on the side force and is a possible reason for the difference between CFD and model tests. A change in crosssection reduces the wake region behind the jet for the streamlined crosssection (S0602A) in comparison to the other crosssection. The absolute side force of the streamlined crosssection (S0602A) is significantly increased (more than 30 [\%] at $m$=0.4 []), while the resistance is slightly increased (4 [\%]) in comparison to the circular crosssection (S1010A) and the flattened crosssection (S0610A). The aim is an increase in absolute side force as it increases the turning ability of the ship, an increase in resistance however is negative on the fuelconsumption of the vessel. In general the crosssectional variation shows promising results, however the numerical uncertainties of the computations are too high. It is advised to check the CFD model scale results with CFD full scale computations and to validate both with measurements. For a future CFD study it is advised to model the hub and strut of the bow thruster, because they can have an influence on the side force on the wedge.bow tunnel thruster; Nienhuis wedge; trailing suction hopper dredger; jet in a crossflow; Hopper wedge; crosssection variation; CFD; Numeca FineMarine; IHC; streamlined crosssection; forward motion; model scale; performance of bow thruster; systematic series
20151216(Maritime and Transport Technology (M&TT)QMarine Technology  Track Science  Specialization Ship Hydromechanics (MTSCSH))uuid:94162987534c4912ac6855b26df20585Dhttp://resolver.tudelft.nl/uuid:94162987534c4912ac6855b26df20585;Micronozzle Performance: A Numerical and Experimental Study0Ganani, Chaggai (TU Delft Aerospace Engineering)Zandbergen, Barry (graduation committee); Cervone, Angelo (mentor); Cowan, Kevin (graduation committee); Delft University of Technology (degree granting institution)Micropropulsion is universally considered to be a key technology enabling nano and pico satellites to perform more complex missions. However, past research has shown that nozzle efficiencies at the microscale are far inferior to their macro scale counterparts. These low efficiencies can be attributed to the relatively high viscous losses associated with this microscale. This thesis conducted a threedimensional numerical study to investigate the impact of the nozzle geometry on the viscous losses. Furthermore, the designed micronozzles are fabricated and the ground work is laid for experimental testing of these nozzles. Results of the numerical study show that by application of a new double depth micro aerospike de< sign nozzle efficiencies can be improved by as much as 41.2%./MEMS; Micronozzle; CFD; Microthrust; Resistojet
20200509)uuid:54552d42d0b64356b3b4049ad305a505Dhttp://resolver.tudelft.nl/uuid:54552d42d0b64356b3b4049ad305a505)Skydive Cavity Buffeting Noise ReductionFJousma, Werner (TU Delft Aerospace Engineering; TU Delft Aerodynamics),Cavity buffeting noise is the main contributor to discomfort in skydive tunnels. Low frequency noise, generated by selfsustained cavity shearlayer oscillations, similar to automotive sunroofbuffeting noise and sidewindow buffeting noise, is observed in skydive tunnels. Typical low frequencies are not conventionally audible to the human ear: registration of this phenomenon occurs due to the high strength of the oscillations felt on the innerear. Longterm exposure to high intensity low frequency noise can be experienced as fatiguing and annoying. A design Study of Streamlines Design BV has shown a fully effective design solution by simulation, which proved to be ineffective in reallife. The current study deals with validation of a different numerical setup and the analysis of various retrofit designs for the reduction of skydive buffeting noise.<br/><br/>Existing literature has shown the promising results for the use of compressible computational fluid dynamics solvers, combined with detached eddy simulation for similar problems. The computational cost of detached eddy simulation has been found within the computational resources for this thesis, opposed to large eddy simulation or direct numerical simulation. For this thesis, the delayed detached eddy simulation model has been used. For the acquisition of validation data a Scanivalve DSA3217 pressure scanner has been used. For computational fluid dynamics and experiments, pressure has been probed at the same location in the cavity. Four retrofit designs have been proposed for analysis in this thesis: trailing edge extension, wall normal cylinder, span wise cylinder and a wing. Simulations have been conducted for one specific tunnel velocity and experiments for a full range of operating setpoints.<br/><br/>Simulations have shown reductions of 6.0 dB and 2.6 dB for respectively the trailing edge extension and the wing in sound pressure levels of cavity buffeting noise. Furthermore an increase of 3.6 dB and 1.5 dB have been observed for respectively the span wise cylinder and wall normal cylinder. A reduction of the wake by 5.2% has been observed for the trailing edge extension, where the other designs increased the wake.<br/>For the clear reduction in sound pressure level, introduced by the trailing edge extension, the formation of a larger scale vortex has been delayed downstream compared to the current situation. Visualising the vortices in the case of the span wise cylinder, it has been found that the formation of the large scale vortex has been triggered further upstream compared to the current situation, possibly due to interaction with cylinder vortex shedding. <br/><br/>Comparison of simulation and experiment data has shown errors of 11.8% in sound pressure level for the current situation and 17.6% when the trailing edge extension is employed. The reduction in experiment has been observed to be stronger than in simulation.<br/><br/>An evaluation experiment has shown the effectiveness of the trailing edge extension for all tunnel velocities at which high sound pressure levels have been observed. Reductions between 7.7 dB and 22.6 dB have been observed. Over the whole range of tunnel velocities, the sound pressure level has been reduced to 130.1 dB and below.<br/><br/>Concluding on the thesis work, the numerical model is able to predict a reduction. Following the simulation work the trailing edge extension is most effective in both reducing the noise and maintaining tunnel performance. Furthermore, the trailing edge extension has proved effectiveness for all critical tunnel velocities in experiment.mBuffeting; Skydive tunnel; Cavity; CFD; Selfsustained oscillation; Noise; Comfort; Experimental; Validation)uuid:178e2496cf4a440b< a877ed34b2206284Dhttp://resolver.tudelft.nl/uuid:178e2496cf4a440ba877ed34b2206284QA state observer based data assimilation method between RANS and robotic PIV data;Tumuluru Ramesh, Nikhilesh (TU Delft Aerospace Engineering)uSciacchitano, Andrea (mentor); Saredi, Edoardo (mentor); Delft University of Technology (degree granting institution)uExperimental fluid dynamics and computational fluid dynamics have traditionally been treated as disparate fields of study. However, each field has its own unique set of advantages and disadvantages. Data assimilation is a field that can be used to leverage some of the advantages each field offers to help compensate mutual<br/>weaknesses. In this thesis, a state observer based data assimilation method is used to assimilate 3D experimental data obtained in a wind tunnel experiment onto a steady RANS simulation. The experimental data is considered as the ground truth and is used to condition the RANS simulation. An understanding of the working of the method along with a study on the effect of different parameters of the state observer method are gathered by first applying it on the 1D viscous Burgers equation and a 2D CFD simulation. For the 3D case, experimental data is obtained by performing a wind tunnel experiment using robotic PIV to map the timeaveraged velocity field around a bluff body following which the data is assimilated onto a steady RANS simulation of the same body. Application of this method helps to recreate topological features and velocity<br/>fields of the flow with better accuracy than a baseline CFD simulation. Finally, the effects of the different parameters on the success of the method along with recommendations for improving the method are provided.+PIV; CFD; Data Assimilation; State observer
20211231)uuid:9db0dc728ec24d75b1f936dea1f40918Dhttp://resolver.tudelft.nl/uuid:9db0dc728ec24d75b1f936dea1f40918XApplication of the computational fluid dynamics solver FLUENT to keels of sailing yachtsDe Baar, J.H.S.iHuijsmans, R.H.M. (mentor); Keuning, J.A. (mentor); Pourquie, M.J.B.M. (mentor); Gerritsma, M.I. (mentor)
The keel of a sailing yacht has been shown to constitute a significant part of the overall resistance. Where the details of this effect are not yet fully understood, Computational Fluid Dynamics (CFD) analysis might reveal mechanisms unseen to the experimental eye. An important step in CFD application is the simulation of a number of validation cases. In the present study I simulate three different validation cases in the commercial CFD solver FLUENT, applying a Reynold's Averaged NavierStokes (RANS) method with a realizable kepsilon turbulence model and a Volume of Fluid (VOF) freesurface approach. From these three validation cases I obtain five drag coefficients, four of which are within an acceptable range of error of the experimental values. After this validation, I consider several mechanisms related to keel resistance. Simulations indicate that the keel rudder interaction is Froude scaled and that the keel resistance can be scaled by a form factor method, presumably by means of a flat plate skin friction line.8CFD; sailing; yacht; keel; vof; turbulence; freesurface
20100621Ship Hydromechanics)uuid:3d1ff1ee3dc24a98861d37453354867fDhttp://resolver.tudelft.nl/uuid:3d1ff1ee3dc24a98861d37453354867fBDesign of the scoop for the underwater exhaust system: A CFD study/Desai, Sanjeet (TU Delft Aerospace Engineering)Boelens, Okko (mentor); Gerritsma, Marc (mentor); Rietveld, Leon (mentor); Hickel, Stefan (graduation committee); Snellen, Mirjam (graduation committee); Delft University of Technology (degree granting institution)!Feadship yachts are designed for the leisure and cruising across the oceans. These luxury yachts are mostly powered by diesel engines or in some cases, a dieselhybrid system. To prevent the inconvenience and the discomfort arising from the diesel exhaust gases for passengers, Feadship yachts are equipped with an underwater exhaust outlet. These underwater exhaust outlets are located on the side of the hull cl< ose to the dynamic waterline. It consists of an external appendage called "scoop" which creates a low pressure region for the exhaust outlet. During recent sea trials of the Feadship yachts, undesirable variations in the exhaust backpressure were observed at the underwater outlet. These undesirable variations led to a situation with either too high backpressure or too low backpressure. An excessive backpressure will increase the fuel consumption and will damage the diesel engine. Contrary, an extremely low backpressure will give a visible exhaust flow above water thereby discolouring the hull and contaminating the deck with exhaust gases and steam.<br/><br/>An ideal scoop design would substantially reduce the above described problems. To investigate the optimal design for a scoop, a numerical method will be used. The method applicable in this study will be the Mutiphase Flow models from the commercial Computational Fluid Dynamics (CFD) software called Star CCM+. A multiphase fluid interaction between the exhaust gases and sea water will be examined to find the physical phenomenon affecting the backpressure at the underwater outlet. In return, this phenomenon will be useful for a thorough analysis of the different scoop designs and how this design could impact the backpressure. Furthermore, the validation of the numerical method will be carried out against the data procured from sea trials of the yachts with current scoop design.<br/><br/>To conclude, design recommendations for an optimal scoop geometry will be provided such that it can reduce the excessive back pressure, have a low resistance and prevent the discolouring of the hull.(CFD; Underwater exhaust; Multiphase flow
20231025)uuid:88a16dcdf4044d299badd3ffd0cc2023Dhttp://resolver.tudelft.nl/uuid:88a16dcdf4044d299badd3ffd0cc20234CFD of multiphase pipe flow: A comparison of solvers
Peeters, P.T.van Zuijlen, A.H. (mentor)Pipelines with multiphase flow of gas, oil, and water are commonly used in the oil and gas industry. In the presence of offshore platforms or vessels, the pipeline ends with a vertical riser. A possible new concept is to use a single large diameter pipeline along the sea floor that ends into multiple smaller diameter risers. This concept might be of interest for future Floating Liquefied Natural Gas (FLNG) vessels. This report focusses on detailed numerical simulations for a small scale representation of this industrial flow splitting configuration: a twophase flow of air and water though a 0.5 [m] long, 0.05 [m] diameter, horizontal pipe with a Tjunction to a dual 2.5 [m] high, 0.05 [m] diameter riser system. A comparison has been made between two multiphase flow solvers available in the wellknown opensource code OpenFOAM. The interFoam solver utilises a mixture model formulation, while the multiphase EulerFoam solver uses an EulerEuler (two continua) formulation for both fluids. Both solvers use Volume of Fluid as interface sharpening method to solve the equations for incompressible flow. Air and water are flowing into the domain with a total volumetric flow rate of 64.2 [m^3/h]. The volumetric flow rate of water in the simulations is taken as 1, 2 and 3 [m^3/h]. Furthermore, a pressure difference between the outlets of the dual risers is applied. This is meant to analyse the difference between solutions of both solvers with respect to maldistribution between the two risers, pressure loss and liquid holdup. The Pressure Implicit with Splitting of Operator (PISO) algorithm with two corrector loops is used combined with a low CourantFriedrichsLewy condition of 0.25 to obtain sufficient numerical stability. Second order discretisation schemes in space and a first order scheme in time are used. In order to speed up the calculations Geometric Algebraic Multigrid is used to solve the pressure field. Meshing of the domains is done with snappyHexMesh. Three Tjunction meshes are generated with increasing fineness of 23042, 47544 and 95292 cells. Two riser meshes are made with 67720 and 276784 cells. The Large Eddy Simulation turbulence mode< l is used with the Smagorinksy SubGrid Scale model. The fixed flow rates at the inlet are coupled with a variable pressure. At the outlets the outflow velocity is variable with a fixed pressure. The turbulent viscosity at the wall is governed by wall functions. The calculations are done on the hpc12 cluster at Delft University of Technology. The calculation times are in the order of one to three weeks for a typical simulation. Overall, differences between the production at the outlets of interFoam and multiphaseEulerFoam are small. Solutions from both solvers indicate that the influence of a pressure difference over the outlets has more influence on the nonsymmetric production of air than of water. The results from multiphaseEulerFoam look promising and due to its EulerEuler momentum description it shows realistic flow behaviour in the junction. Improvements in simulating the system can be made by progressing the simulations longer in time, by increasing the entrance length of the horizontal pipe towards the junction and by choosing another SubGrid Scale model.NCFD; pipe flow; multiphase flow; LES; OpenFOAM; interFoam; multiphaseEulerFoam&Aerodynamics, Wind Energy & PropulsionAerodynamics)uuid:a47f9e446f684b39b9a989b16b99cf9cDhttp://resolver.tudelft.nl/uuid:a47f9e446f684b39b9a989b16b99cf9c_Analysis of fluid motion in cryogenic propulsion upper stage tanks during launcher ascent phase3Tsavlidis, Ioannis (TU Delft Aerospace Engineering)Hickel, Stefan (mentor); Feindt, Oliver (mentor); Gangoli Rao, Arvind (graduation committee); Zandbergen, Barry (graduation committee); Delft University of Technology (degree granting institution)
class="MsoNormal">The graduation project was conducted at the upper stage liquid propulsion department of ArianeGroup at the facilities of Airbus DS, Bremen. Based on a series of past flights of Ariane 5 launcher, the aim is to analyze the fluid motion and the pressure fluctuations in the cryogenic propulsion upper stage Liquid Hydrogen (LH<sub>2</sub>) and Liquid Oxygen (LO<sub>x</sub>) tanks during the ascent phase. Pressure fluctuations (drops or rises) during the ascent phase are undesirable due to the need for relief or repressurization of the fuel tanks. Tank relief is obtained through relief valves, while repressurization is done using onboard gaseous Helium; both cases increase the failure probability of the system and/or the total weight of the launcher.The objective of the project is to find out why these pressure fluctuations occur, what are the parameters that affect the pressure evolution and at what extend the liquid fuel motion (sloshing) is responsible for this behavior.According to the literature several parameters affect the pressure evolution during sloshing. These parameters are further investigated through flight data analysis. The approach also involves CFD simulations of the kinematic behavior of the liquid fuel focusing on the sloshing angle. Finally, a statistical model is built attempting to predict the pressure change inside the tanks. Higher sloshing angles match with higher pressure rise inside the LO<sub>x</sub> tank. The magnitude of the pressure rise appears to be directly connected to the kinematic profile of the launcher as well as to the ullage volume of the tank. The maximum predicted ullage pressure is below the tank's sizing pressure limit. Regarding the LH<sub>2</sub> tank, no strong correlation of flight parameters to the pressure change is identified; no sufficient statistical model is built. The CFD simulation shows that relatively higher sloshing angle magnitude and duration exists near the pressure drop periods and that strong breaking waves are likely to be formed in the case of a sudden pressure drop behavior. The LH<sub>2</sub>tank is more prone to the formation of breaking/splashing waves. The effect of vibrations, which is not included in the CFD study, is also important for the explanation of the pressure drop magnitude.UCryogenic Tanks; Sloshing; Ariane 5 Launcher; Ascent Phase; Flight Data Analysis; CFD)uuid:44818c8328984e34be3a75efd1a< 71e2dDhttp://resolver.tudelft.nl/uuid:44818c8328984e34be3a75efd1a71e2dkInternal Loads of SemiSubmersibles at an Inconvenient Draft: An investigation into the nonlinear responseFKorte, Ruben (TU Delft Mechanical, Maritime and Materials Engineering)Wellens, Peter (mentor); Keetels, Geert (graduation committee); Jacobi, Gunnar (graduation committee); Vlasveld, Ebert (graduation committee); Delft University of Technology (degree granting institution)
Once a semisubmersible is operating at an inconvenient draft (A shallow draft with limited water column above the pontoons), the passing waves over the pontoons can not keep their linear motion and energy will be transferred to higher harmonic wave frequencies. This means the hydrodynamic response will also contain energy at these higher frequencies. Conventional linear diffraction solvers are not able to solve the combined response of the wave frequency and its higher harmonics, which then result in a nonphysical solution for the wave loads, motions and internal loads. This research aims to obtain a better insight into the hydrodynamic response at the inconvenient draft and ultimately on the internal loads of the semisubmersible. In the first part of this research, model test solutions, linear potential solutions obtained with WAMIT and CFD solutions obtained with ComFLOW are compared. the comparison shows that ComFLOW is able to provide more accurate wave loads compared to the linear potential solver. The higher harmon ics observed in the measured wave loads during model tests are correctly predicted by ComFLOW, although maximum deviations of 30% are still observed between measured and predicted waveload amplitudes.<br/>However, ComFLOW is not able to solve the motions of a freefloating semisubmersible correctly. Due to pressure peaks in the waveexciting forces, solving the equation of motion results in an incorrect motion response and ultimately results in incorrect internal loads. Although a significant effort was made  in close collaboration with the ComFLOW developer  to improve this functionality, the results remained unsatisfactory. Even so, in order to obtain an insight into the higher harmonic response contributions to the internal loads, a parameter study has been conducted, using tests in which the semisubmersible was held captive. This parameter study was conducted by systematically varying wave amplitudes and draft, and resulted in situational limits at which the higher harmonic response contribution becomes significant and the linear relation between the incoming wave and the hydrodynamic response is lost. This limit is shown to be dependent on the Ursell number. Furthermore, it is demonstrated that the significance of the higher harmonic response contribution increases from 10% to 40% throughout the inconvenient draft, while the most severe situations resulted in a higher harmonic response contribution of 50% of the total response amplitude. In the final part of this study, an attempt is made to couple the wave loads from ComFLOW to internal loads. A quantitative analysis is made on the effect of the higher harmonic responses of the wave loads on the internal loads. The time domain simulator aNySim is used combined with the wave exciting forces on a captive semisubmersible calculated with ComFLOW. A multibody analysis is used to obtain the internal loads on the aft and front part of the semisubmersible. This did not provide the correct answers because the stiffness of the spring damping between the two sections affects the higher harmonic response contribution. This method overestimated the higher harmonic response contribution. A better understanding of the joint and the joint stiffness/damping of a dualbody simulation may solve the encountered problems.:Internal loads; Semisubmersible; CFD; ComFLOW; nonlinearMarine Technology)uuid:73c8079a17604fcaa05f166f836163a0Dhttp://resolver.tudelft.nl/uuid:73c8079a17604fcaa05f166f836163a0GVerification and validation of fullscale propulsion analysis using CFDHWieleman, Vera (TU Delft Mechanical, Mar< itime and Materials Engineering)van Terwisga, Thomas (mentor); Schenke, Sren (mentor); Pourquie, MAthieu (mentor); Vaz, Guilherme (mentor); Schuiling, Bart (mentor); Delft University of Technology (degree granting institution)e
An accurate prediction of propeller hull interaction is an important step in the design of a new vessel. The prediction of fullscale flow phenomena, which eliminates scale effects, is becoming available due to increasing computational power. However, the complexity of fullscale CFD calculations combined<br/>with a lack of validation data results in unknown uncertainties. This study contributes to the uncertainty estimation for fullscale calculations by answering the question With what uncertainty can we currently numerically predict resistance and propeller power on fullscale Reynolds numbers?.<br/>The resistance and propeller flow predictions are done for the general cargo vessel MVregal for three cases; a double body, a free surface resistance simulation and an open water propeller simulation. The simulations are performed for the design speed of 14 knots, resulting in a fullscale Reynolds<br/>number of ER = 1.12 " 10. The discretization error is determined by the grid refinement study as presented by Ea and Hoekstra for the propulsion parameters; resistance coefficients and the wake factor. For each case the flow field is analysed, an uncertainty assessment is made and the results are compared to a group of numerical results for the same simulation performed by 20 participants of a case study organised by Lloyd s register on the same vessel as is considered in this thesis.<br/>Modelling the boundary layer of a flat plate on model and fullscale Reynolds numbers encourages the use of unstructured grids for fullscale Reynolds numbers. The uncertainties as predicted by the grid refinement study, vary between 0.6 and 24.5 percent for the friction coefficient. For the ER = 10^7 the values are compared to a structured grid study, which had a better trend over the grid refinement series. Comparison to theoretical friction coefficient calculations confirmed the absolute friction result.<br/>The double body simulation, performed on the fullscale number ER = 1.12 " 10^9, demonstrated the use of the unstructured grids on the fullscale Reynolds numbers. The iterative error had to be closely monitored in order to get a stable solution. While the iterative errors had the same order of magnitude, the uncertainties as predicted by the grid refinement study for the propulsion parameters varied between 1.5 and 140 percent. This is an unacceptable large scatter in uncertainty which calls for another method to determine the uncertainty.<br/>The free surface simulation added complexity, by modelling the free surface resistance of the vessel. The order of convergence is lower for the free surface simulation, which creates a higher iterative error in the uncertainty assessment. The discretization uncertainty prediction varies between 2.7 and 23.5 percent. <br/>The open water simulation, which is based on a hybrid grid of structured and unstructured grids, showed for the monitored parameters a sufficiently low iterative error and a low discretization uncertainty (between 0.1 and 3.4 percent) for the thrust and torque coefficients. Although all different cases showed mixed results for the discretization error, the absolute values are well within the range of results from the test group of 20 participants. This is a good starting point of the repeatability of the flow parameters. It is noted that the current uncertainty estimation is larger than the difference between two grids.tCFD; fullscale; Uncertainty; Verification; Validation; Double Body; Free Surface; Open Water; Grid refinement study)uuid:2e3c46fa175d4580a1768f6c8c5425e9Dhttp://resolver.tudelft.nl/uuid:2e3c46fa175d4580a1768f6c8c5425e9CTimeSupersampling 3DPIV Measurements by VortexinCell SimulationSchneiders, J.F.G.+Dwight, R.P. (mentor); Scarano, F. (mentor)TMeasurement rate limitations of timeresolved 3D3C velocity measurements by tomographic < PIV limit application of the technique to small measurement volumes and low speed flows (~10 m/s). To reduce the challenging repetition rate requirements historically set by the Nyquist criterion, in the present thesis work a novel method is proposed, combining PIV measurements with numerical simulation of the vorticity transport equation using a hybrid vortex particle discretization. The principle of the timesupersampling method is that the spatial information available by the measurements can be leveraged to increase the temporalresolution. The solution of the governing equations is based on the VortexinCell (VIC) method and the unsteady numerical simulation of the temporal evolution of the measured flow is applied within the 3D measurement domain. Both forward and backward timeintegration is performed between pairs of consecutive measurements. The accuracy of the proposed timesupersampling method is studied with two experimental datasets obtained from timeresolved tomographic PIV measurements: a turbulent wake, and a circular jet. The results are compared to linear interpolation, advectionbased supersampling, and measurement data at high sampling rate. In both flows the ability to reconstruct detailed temporal dynamics from data sampled at a rate far below the Nyquist frequency is demonstrated. The study demonstrates that measurement rate requirements can be strongly reduced when the measurements are supersampled with the proposed timesupersampling method, thereby extending the range of application of tomographic PIV. In addition, an alternative application of the approach in the field of noise reduction and application to instantaneous measurements is illustrated. The latter can on the one hand allow for a significantly improved predictor for fluid trajectory correlation methods and on the other hand when validated can pose a radically simplified approach for calculation of the instantaneous and unsteady pressure field from single tomographic PIV snapshots, in comparison to multipulse systems.Tomographic PIV; NavierStokes; Particle Image Velocimetry; CFD; VortexinCell; Turbulence; Unsteady; Temporal resolution; Data Assimilation)uuid:d4c717e8809041eea6a06c7a243ea836Dhttp://resolver.tudelft.nl/uuid:d4c717e8809041eea6a06c7a243ea836;Bilge Keel Roll Damping: Combining CFD and local velocitiesVan Kampen, M.J.4Van 't Veer, R. (mentor); Huijsmans, R.H.M. (mentor)? As hydrocarbon supplies dwindle, technology develops and longterm hydrocarbon prices rise it is becoming more and more economical to develop hydrocarbon fields offshore. An FPSO vessel can be favored as it is flexible, quickly commissioned and costeffective, be used at all water depths and does not require additional pipelines to shore. To predict the roll motion of an FPSO traditional techniques such as the IkedaTanakaHimeno (ITH) method are no longer sufficient due to aberrant dimensions and shapes of bilge keels and riser balconies. The goal of this thesis was to provide a practical method to evaluate the roll damping and motions of an FPSO with aberrant bilge keels and/or riser balconies in (ir)regular waves. For this the ITH method was modified in three manners: 1. by extending current formulations with outofphase terms, 2. by obtaining relevant coefficients from 2D CFD simulations in forced roll oscillations, and 3. by using linear potential theory to obtain local velocities consisting not only of rigid body velocities, but also radiated wave velocities, incoming wave velocities and diffracted wave velocities. This new methodology was compared to forced oscillation and regular wave experiments performed by MARIN of the model scale Glas Dowr FPSO. Forced oscillations were reproduced satisfactory after a correction for additional heave and sway motions was applied. Regular wave results were compared to measurements, simulations using experimentally derived damping coefficients, ITH method damping coefficients, local velocitybased ITH method coefficients and using the proposed methodology. Results were good when compared to simulations based o< n measured damping coefficients but inconclusive when compared directly to measured roll amplitudes. Reasonable agreement at low to medium wave amplitudes and an underestimation at high wave amplitudes were obtained. The underestimation at high amplitudes is faulted to the linear increasing hull pressure coefficient while it is more likely to become saturated at higher local velocities. It is concluded that the combination of CFD and local velocities yield promising results and is more flexible than the traditional ITH method. A more thorough validation should be performed against data at various frequencies, hulls, keels and wave amplitudes before application becomes feasible.3bilge keels; roll damping; CFD; FPSO; riser balconyMaritime Transport Technology)uuid:d1c58f2e40dc43c68383f2041593d1c4Dhttp://resolver.tudelft.nl/uuid:d1c58f2e40dc43c68383f2041593d1c4rEffects of Increasing Aerothermodynamic Fidelity on Hypersonic Trajectory Optimisation for Flight Testing PurposesVan Oostrom, J.GSchrijer, F.F.J. (mentor); Mooij, E. (mentor); Sudmeijer, K.J. (mentor)In a previous study, the optimal reentry trajectory of Hyperion2 has been derived. The mission of the vehicle is to measure hypersonic boundarylayer transition, a phenomena in which the laminar boundary layer turns into a turbulent boundary layer. This is an important feature in hypersonic flow to investigate, as it introduces peak heating and increases drag. A constant Mach 10 flight has been performed, optimising for flight time, whilst maintaining a large Reynolds number range, in which transition occurs (Retrans = 106). Hyperion2 is a theoretical experimental vehicle studied at Delft University of Technology. The aerodynamic coefficients of Hyperion2 in the previous study have been obtained using a modified Newtonian panel method. In present thesis, the aerodynamic coefficients are computed using an open source computational fluid dynamics software called SU2 (Stanford University Unstructured). Simple geometries have been used to verify and validate the solver, focusing on shock position and shape, pressure distribution and heat flux. Using an open source mesh generator called GMSH, a three dimensional unstructured grid of Hyperion2 has been generated. With this grid, the aerodynamic database consisting of 120 combinations of Mach number and angles of attack has been created. The database has been used to create a new optimal trajectory for the Hyperion2 mission. During the presentation, the differences in aerodynamic coefficients are evaluated and the effect on the optimal constant Mach 10 flight trajectory is presented.hypersonic; trajectory; optimisation; CFD; Computational; Fluid; Dynamics; Hyperion; Stanford; University; Unstructured; SU2Aerodynamics and Wind Energy)uuid:7985085a7aad4fce885369eaa5289b6cDhttp://resolver.tudelft.nl/uuid:7985085a7aad4fce885369eaa5289b6cVOptimisation of Flow Distribution for Pipe Pullback in Horizontal Directional DrillingJSaliba, Jonathan (TU Delft Mechanical, Maritime and Materials Engineering)Poelma, Christian (mentor); Pourquie, MAthieu (graduation committee); Delft University of Technology (degree granting institution)Horizontal Directional Drilling permits the creation of tunnels which pass beneath rivers and canals to allow the passing of services from one side to the other. The final stage of this process involves lining the tunnel with a plastic or steel pipe. The pipe is pulled inside into the borehole using the same drilling rig that was used to bore the tunnel. For the case of a plastic High Density Polyethylene (HDPE) pipe, issues of buoyancy may arise during this pullback process since the tunnel is prefilled with drilling mud which now primarily acts as a lubricant. This presents a problem since with this buoyancy, the pipe is lifted up to brush against the tunnel wall, creating issues with the pullback process because of the added pulling resistance. There is the possibility of cutting slots in the pipe wall at the front end of the pipe in order to allow in drilling mud so as to ballast and o< ffset this buoyancy. The drilling fluid is a suspension of bentonite in water and is characterised as a nonNewtonian shearthinning HerschelBulkley fluid, which possesses a finite yield stress. This study aimed to find whether the current slot proportions used for a given borehole diameter, pipe diameter and pipe thickness are sufficient for allowing in drilling mud to ballast the pipe. This was a case where the multiphase flow given by the interaction of air and drilling mud can be simulated using CFD. OpenFOAM is used for this purpose to first simulate the current practice. The multiphase solver interFoam together with the nonNewtonian HerschelBulkley and air model was validated for a series of cases before the main simulations were run. These validations included the Marsh Funnel test and the Slump test. These are two kinds of workability tests used for cement pastes and drilling muds. Cement pastes and drilling muds are characterized as threeparameter HerschelBulkley fluids and the physical setups of the Marsh Funnel and Slump tests were replicated in computational space. The flow time of theMarsh funnel test and the slump diameter from the CFD simulations were compared with experimental data from literature thus validating the model in OpenFOAM. The main simulation setup recreates the situation of an HDPE pipe concentric with the borehole, with the axis of the domain at an angle to the horizontal. There is drilling mud above the slot at time t = 0s. The drilling mud comes into the domain from the annulus from below the slot. The simulations showed that the flow into the slot initially came in from both the drilling mud above the slot and from the inlet. After the volume of drilling mud above the slot is almost drained completely through into the inner pipe, the mud level inside the pipe starts to become comparable that outside the pipe and both interfaces rise upwards at approximately the same pace. Subsequently, certain parameters were changed from the first benchmark case in order to see what is the effect of these individual variables. The effects of a lower drilling mud yield stress, a longer pipe slot, lower drilling mud density, a different slot aperture shape, increased flow and a steeper angle of pipe penetration were all tested in the simulation campaign. The idea is that with better and faster filling, less pipe buoyancy results. An extension of this idea is that the difference between mud levels inside and outside the pipe should be kept to a minimum. That is why the aim is to increase the flow rate through the slot in the pipe. The results show that increasing the slot length by 30% from the current practice increases the throughput of drilling mud by 10%. The results also show that an elliptical slot profile has a neutral effect and decreasing the yield stress of the drilling mud has a slightly beneficial effect.ICFD; multiphase flow; OpenFOAM; Drilling; Horizontal directional drilling)uuid:ee2db063a1c04f0cb8a679f6f3acb749Dhttp://resolver.tudelft.nl/uuid:ee2db063a1c04f0cb8a679f6f3acb749NThe influence of laminarturbulent transition on the perfomance of a propeller
Janssen, R.F.1Veldhuis, L.L.M. (mentor); Eitelberg, G. (mentor)]The influence of laminarturbulent transition on a propeller blade s performance has been investigated in this study, both numerically and experimentally. The computational fluid dynamics (CFD) is performed using the TAU code developed by the German Aerospace Center, as well as an existing propeller liftingline code. The laminarturbulent transition is simulated using the ?Re?t correlation based transition model, which is compared to the results of the SpalartAllmaras oneequation turbulence model. To validate the CFD data, experiments are performed in the Open Jet Facility of Delft University of Technology, where the laminarturbulent transition is measured using an infrared camera. The results show that at high advance ratios, the difference between the CFD simulations using the laminarturbulent transition model and the oneequation turbulence model is large. This can be e< xplained due to the trailing edge separation which is present in the case where laminarturbulent transition is modeled. At lower advance ratios, the difference between the two CFD models becomes smaller. This is due to the fact that a larger portion of the flow over the blade becomes turbulent. Comparing the CFD results to the experimental results it can be seen that the location of transition predicted by the CFD simulations shows some agreement with the experimental results.vpropeller; transition; performance; CFD; computational; fluid; dynamics; infrared; numerical; experimental; RANS; BEMT)uuid:de6e82b528f14a7da073032620a88b60Dhttp://resolver.tudelft.nl/uuid:de6e82b528f14a7da073032620a88b60LA Numerical trimvariation study for ships operating in offdesign conditions
De Jong, R.H.Veldhuis, H.J. (mentor)Fuel efficiency is an important factor for the shipping industry both regarding new build vessels and existing vessels. A method to reduce the fuel consumption of existing vessels operating in off design conditions is to trim a vessel in the most optimum trim condition. The current approach to determine the most efficient trim condition is to perform propulsion tests under different trim conditions in a towing tank. To determine the usability of the CFD program PARNASSOS a trim variation study is performed on a vessel that is already tested in one of the tanks of MARIN. The trend of the propulsion from the tank tests is used to validate the trend of the resistance from the CFD calculations. Before validating the trend of the resistance with the trend of the propulsion it was checked whether or not the change of the resistance is dominant over the change in propulsive efficiency. It turned out that the change of the resistance is dominant over the change of the propulsive efficiency and therefore the trend of the propulsive power can be used to validate the trend of the resistance. The uncertainty of the CFD calculations is determined using a grid refinement study according to the method developed by Luis Ea and Martin Hoekstra [Ea andHoekstra, 2014]. According to this method the uncertainty of the performed calculations is estimated at 3.5%. The uncertainty of the tank tests is estimated at 0.9% according to Martijn van Rijsbergen [van Rijsbergen, 2014]. Unfortunately the uncertainty value obtained using the current method for estimating the uncertainty is not small enough to validate the trend of the resistance with the trend of the propulsion from the towing tank tests. The fact that the uncertainty determined is not small enough to validate the trend obtained from the CFD calculations doesn t necessarily invalidate the assumption that the results of the CFD calculations are correct. It is expected that the uncertainty of the difference is smaller than the relative uncertainty, an expectation that is supported by the fact that the fits of the power series estimation for the resistance coefficients of the even keel condition and the 1.5m aft trim condition show the same trend and the two fits do not cross each other. There are at least two ways to further reduce the estimate of the uncertainty. The first one is to use finer grids. The second one is to further optimize the method to determine the uncertainty of these CFD calculations and reduce the influence of data scatter and nonsimilarity of the grids on the estimate of the uncertainty. The results of the CFD calculations are analyzed as if the data is validated. Trimming the vessel aft resulted in an increase of the total resistance. Trimming the vessel forward reduced the total resistance. The total resistance, the frictional resistance and the hydrodynamic pressure resistance increase when the vessel is trimmed aft while the hydrostatic pressure resistance reduces. These results show that the change in frictional resistance and the hydrodynamic pressure resistance are dominant over the change in hydrostatic pressure resistance. For moderate changes of trim the change in wetted surface is dominant over the change of shear stress regarding the change in frictional< resistance. For extreme changes in trim the cause of the increase of the frictional resistance is a sheet vortex developing at the bow of the vessel and running aft along the bilge of the vessel. This sheet vortex influences the local thickness of the boundary layer and therefore influences the local shear stress. At the transom of the vessel the local hydrodynamic pressure resistance is reduced caused by the submergence of the transom. Trimming the vessel aft reduced the pressure recovery at the stern of the vessel, which has a negative influence on the local hydrodynamic pressure resistance. At the forward shoulder of the vessel the hydrodynamic pressure resistance increases as well when the vessel is trimmed aft. Trimming the vessel aft resulted in an increase of the hydrodynamic pressure resistance, which shows that the effect of the pressure recovery at the stern of the vessel and the increase of hydrodynamic pressure at the forward shoulder of the vessel are dominant over the effect of the presence of a dead water zone behind the transom of the vessel.^CFD; Computational Fluid Dynamics; Trim Variation; Hull Flow; Physical explanation; ValidationMaritime & Transport Technology Ship Hydromechanics & Structures)uuid:8656b6a70c73499e91bea617acf6c28cDhttp://resolver.tudelft.nl/uuid:8656b6a70c73499e91bea617acf6c28cOEffect of Gravity on the Vertical Force of an Oscillating Wedge at Free SurfaceHu, W.5Huijsmans, R.H.M. (mentor); Kapsenberg, G.K. (mentor)A wedge oscillating vertically at free surface with frequency ranging from wave frequencies and slamming frequencies has been studied. The focus of this work has been on the gravity effect on the vertical forces acting on the wedge, especially the definition of the limits of gravitydominant and gravitynegligible conditions and quantification of the gravity effect in between those limits.>wedge; vertical oscillation; gravity; radiation; slamming; CFD)uuid:b3a53d4fa28a486da09a51ce441edc6fDhttp://resolver.tudelft.nl/uuid:b3a53d4fa28a486da09a51ce441edc6f[CFD Modeling of TwoStage Parallel Plate Sedimentation Centrifuge for Microalgae DewateringYu, B.CQu, Z. (mentor); Van der Kraan, M. (mentor); Witkamp, G.J. (mentor)As one of the most fast growing species on earth, microalgae provides great potential to satisfy the ever increasing demand in food, energy and material in a sustainable way. The focus for this thesis work is on one of the most important bottle neck of microalgae harvest process: microalgae dewatering, by CFD modeling of the flow and sedimentation separation in Evodos SPT centrifuge. Various microalgae dewatering technologies have been reviewed and evaluated. Compare to traditional conical disk centrifuge Evodos SPT centrifuge provides 10% to 20% energy consumption, removing up to 95% extracellular water and other benefits i.e. mechanical simplicity and process flexibility etc. In the model, the fluid dynamic behaviors of multiphase flow has been considered. In this research a complete 3D CFD model of the Evodos centrifuge consisting of five sub components have been built. The particle behavior for the centrifugation separation is based on DPM (Discrete Phase Model) in Fluent. The result of the 3D CFD model gives a clear overview of the pathline, flow pattern and pressure profile inside the centrifuge as well as separation efficiency on particle sizes. The model has been validated through visual result from algae separation test runs, theoretical equations and starch test run measurements. A test and sample taking with starch solution has also been carried out on Evodos site in Breda. This thesis work laid a good foundation for future studies in the CFD modeling of Evodos SPT centrifuge and similar machines. The future focus should be on optimizing the geometry of the parallel plates, impeller chamber for separation efficiency; understanding the effects and impacts of operation conditions and further develop the multiphase model.sustainable; micro algae; dewatering; CFD; sedimentation; Evodos; SPT; centrifuge; parallel plate; flow field; flow patte< rn; discrete particle modelPEQ)uuid:9629143d68d444099f69a274690485e5Dhttp://resolver.tudelft.nl/uuid:9629143d68d444099f69a274690485e5Numerical Analysis on Hemodynamics in Intracranial Aneurysms: Proposing a third hemodynamic criterion for predicting rupture sitesPToussaint, Merel (TU Delft Applied Sciences; TU Delft ChemE/Transport Phenomena)Kenjeres, Sasa (mentor); Perinajov, Romana (mentor); Kleijn, Chris (mentor); Bhattacharya, Nandini (mentor); van Ooij, P (mentor); Delft University of Technology (degree granting institution)4Brain aneurysms cause almost 500,000 deaths in the world every year. A better understanding of brain aneurysm genesis and rupture may open new opportunities to prevention and treatment. In this study, a part of the cerebral vascular system, the so called circle of Willis including an aneurysm, was analyzed with Computational Fluid Dynamics (CFD). The importance of boundary conditions in an aneurysm simulation was assessed by comparing the results to 7T MRI velocity data, obtained from the Academic Medical Center (AMC) in Amsterdam. Adequate similarities were found in velocity values, together with qualitative agreement in wall shear stress (WSS) values. <br/>We found that in a patientspecific case with an aneurysm, the velocity profile was able to develop to a parabolic profile because of its location; the aneurysm existed far downstream of the circle of Willis. This implies that it is possible to use a cropped arterial system to simulate the aneurysm and to use parabolic inlet velocity profiles for patients with this aneurysm phenotype.<br/>In previous studies, it proved hard to locate the precise location of rupture in an aneurysm. A risk assessment of the rupture location in an aneurysm can be used as a more accurate tool to assess the need for surgery. The aneurysm geometry of the CFD Rupture challenge from 2013 was simulated to predict the location of a rupture site. This rupture location was predicted by combining the following hemodynamic criteria: the timeaveraged wall shear stress (WSSTA), oscillatory shear index (OSI) and vortexsaddle point structure during systole with accompanying low pressure values. A sensitivity study was performed on these criteria and a critical threshold for rupture risk was proposed. Based on these criteria it was possible to predict the exact rupture site for two analyzed aneurysm geometries. It is concluded that a CFD model could be used to assess the hemodynamics in intracranial aneurysms. Future research should focus on repeating this study on more patient specific aneurysm geometries to verify this hypothesis further.<brDCFD; intracranial aneurysms; circle of Willis; rupture; hemodynamics
20200523)uuid:b8276e19acf74c298203d09d6922e1adDhttp://resolver.tudelft.nl/uuid:b8276e19acf74c298203d09d6922e1ad4Hydrodynamic Forces on Wind Assisted Ships using CFD
Settels, J.W.Van Terwisga, T.J.C. (mentor)With the fluctuating fuel oil prices and increasing environmental awareness more and more attention is given to durable ways of shipping cargo around the world. Reducing fuel consumption by using wind as a source of auxiliary propulsion is one way to achieve that goals. Different projects utilising Wind Assisted Ship Propulsion (WASP) have been investigated and/or realised in the past years. Since these devices often don t exert the propulsive force in line with the sailed course a leeway angle is introduced, which influences the hydrodynamic forces. This thesis focusses on predicting the sail induced hydrodynamic forces on a coastal cargo vessel under stationary leeway angleswith the RANS solver ReFRESCOat model scale. The ship was modelled in three distinct modelling steps: The bare hull without appendages and propeller, the appended hull without propeller and the appended hull with propeller, implemented through a RANSBEM coupling. All configurations were calculated for a range of (small) leeway angles in a double body set up, neglecting free surface effects. The appended configurations were calculated with different rudder angles and different < thrust ratios relative to the thrust required at the self propulsion point of the model. Furthermore a grid refinement study was performed for the bare hull configuration to estimate the numerical uncertainty of the obtained results. The bare hull configuration and the appended configuration with implemented propeller were compared to available model test data obtained from available towing tank test results. From the grid refinement study on the bare hull a high numerical uncertainty became apparent for the lateral force and yawing moment. Visualisation of the calculated flow around the hull revealed an increase in vortex intensity and a change in location of the shed vortices for grids with increasing cell density. The bare hull longitudinal force was not validated. The lateral force and yawing moment were validated with a high validation uncertainty. The calculated bare hull forces on the grid with highest cell density show an under prediction prediction of the longitudinal force, which can be explained by the omission of wavemaking drag in the calculations due to the double body setup. The lateral force is under predicted while the yawing moment is over predicted. The lift and drag due to leeway shows good agreement with the results up to six degrees of leeway but are under estimated for leeway angles of nine degrees. This indicates that asymmetric wave profiles due to leeway contribute to these forces and cannot be captured by a double body setup. The lateral centre of effort is predicted too far forward but the trend with respect to leeway was captured. For the appended condition with implemented propeller good agreement between the calculated and measured values for the lift due to leeway was found with comparison error for higher rudder angles. The drag due to leeway showed good agreement with the measured results with higher comparison errors for lower thrust ratios. All calculations showed an over prediction of the longitudinal position of the centre of effort of the lateral force. Higher thrust ratios and higher rudder angles showed greater under prediction of lateral force and over prediction of the yawing moment, which was likely caused by the incapability of the model to calculate the increased turbulence in the propeller wake, influencing the forces on the aft part of the hull and rudder. The rudder force in longitudinal direction showed values of up to the same magnitude of the force generated by the hull for higher angles of attack on the rudder. The lateral force produced by the propeller due to leeway was found to be negligibly small in comparison to the lateral force produced by the hull and rudder. CFD; RANS; sailassisted; leeway
20200526Maritime Technology)uuid:e5465aeb59334b75b06889b7ad906296Dhttp://resolver.tudelft.nl/uuid:e5465aeb59334b75b06889b7ad906296Wavestructure interaction study, in the context of floating wind turbines, by mean of coupled rigid body code and Fluidity CFD software[Baudino Bessone, Matteo (TU Delft Electrical Engineering, Mathematics and Computer Science)SVir, Axelle (mentor); Delft University of Technology (degree granting institution)Offshore wind energy is a fastgrowing sector in the energy industry, and the cost of electricity per kilowatthour from offshore wind is decreasing significantly. Until now, offshore wind turbines are mainly installed on bottommounted foundations, which are economically feasible in shallow waters. This factor has limited the development of offshore wind to those countries that can benefit from abundant wind resources located in shallow waters. To further extend the wind energy market, floating support structures for wind turbines are being studied and developed. This technology allows harvesting the wind energy resources located in deep waters, avoiding the costly construction and installation of large towers.<br/>A floating support introduces different challenges with respect to a bottom mounted one, not least the fact that the dynamics of the system are significantly different. The aim of this Master thesis is to generate a numerica< l model that can predict the dynamics of a geometrically simple, twodimensional floater under the action of a train of regular waves. The wavestructure interaction problem is solved coupling the CFD solver fluidity with a rigid body code developed in Python that numerically solves the equations of motion for a rigid body and uses NURBS to define the geometry of the body. The immersed body method is used to represent the solid body into the fluid.<br/>The results obtained with this approach are then compared with potential flow results, experimental results and results obtained with other CFD solvers available in the literature. Similarities and differences are outlined for the different numerical experiments that have been simulated. In general, good agreement has been found with experimental results and other CFD solvers results. With respect to potential flow theory, good agreement has been found for lowfrequency waves, while differences have been noticed for the highfrequency waves.<br2waves; Wavestructure interaction; CFD; WindenergySustainable Energy Technology)uuid:731efb5525dc4cfdb972368cea65a2e4Dhttp://resolver.tudelft.nl/uuid:731efb5525dc4cfdb972368cea65a2e4HNumerical modelling of sedimentation in Trailing Suction Hopper DredgersDSloof, Ben (TU Delft Mechanical, Maritime and Materials Engineering)UKeetels, Geert (mentor); Delft University of Technology (degree granting institution)m
Damen Dredging Equipment is a yard dedicated to the dredging industry. The yard specializes in the design, manufacture and supply of a wide variety of dredging tools. One of the tools Damen offers is the Trailing Suction Hopper Dredger(TSHD).<br/><br/>The performance of such a TSHD is described by its production: the amount of sediment loaded in the hopper per unit time. The overflow loss, the material lost overboard, can easily reach up to 30%, which causes a significant decrease of the production. In addition, the turbidity plume caused by these overflow losses can have a negative environmental impact. This turbidity plume reduces light penetration, clogs filter feeders, and disperses contaminants which can be attached to the sediment. Damen is interested in estimating and reducing these overflow losses. Different models to estimate the amount of material lost overboard exist. All these models have their pros and cons. The analytical model of Miedema & Vlasblom(1996) gives a quick and good estimate of the overflow losses, but gives no insight in the flow inside the hopper. The 2DV model of Van Rhee(2002) is able to accurately simulate the flow inside the hopper, but has a large computation time. The 2DV model of this thesis gives a good estimate of the overflow losses, gives insight in the flow inside the hopper, and has an acceptable computation time.<br/><br/>The new 2DV model was developed in OpenFOAM. At the start of this thesis, it was possible to model mixture flow in OpenFOAM, but a sand bed could not be modelled yet. Several features had to be added to OpenFOAM to overcome this problem. The sand bed was regarded as a solid body inside the computational domain. To simulate the influence of this solid body on the mixture flow, boundary conditions were added at the bed interface. Sedimentation was modelled by adding a moving mesh. The closed flume experiments of Van Rhee(2002) have been used to validate that sedimentation is simulated accurately. By comparing hopper simulations with the hopper experiments of Van Rhee(2002), it was shown that the flow in the hopper was also simulated accurately. The computed overflow losses are, however, on the low side. The current version of OpenFOAM can calculate with only one particle fraction. In reality, the smaller fractions of the Particle Size Distribution are pushed upwards, causing the overflow losses to be higher. The 2DVmodel gave a deeper insight to the phenomena in the hopper. It was possible to derive several equations which describe the flow in the hopper. With these new formulas, a simple phenomenological model was developed, which was named the 'Layer Model'.5sedimenta< tion; hopper; OpenFOAM; CFD; overflow losses
20191231OpenFOAM)uuid:87ad4bb361b54a55a9561db361f133c1Dhttp://resolver.tudelft.nl/uuid:87ad4bb361b54a55a9561db361f133c1Efficiency improvement of viscous ship flow computations through use of the Graphics Processing Unit: A performance analysis on different hardware5de Bruycker, Deborah (TU Delft Aerospace Engineering)mMaritime hydrodynamics involves strong inertiadriven flows, including freesurface waves, and with Reynolds numbers as high as 109. Numerical modelling of these flows is therefore a true challenge. Especially for the optimization of the design of the aft part of the ship, where viscous effects cannot be discarded. At MARIN, the viscous flow solver PARNASSOS has been developed for this purpose. Over the years, this solver has been optimized with respect to robustness, accuracy and efficiency, which has resulted in a tool capable of doing fast viscous flow computations. However, if the ship's generated wave pattern has to be taken into account, a complete hull form evaluation can still take hours, or even days. For automatic optimization purposes, where hundreds or thousands of calculations are needed, it is thus desired to further accelerate the computing time of this solver. Most of the CPU time for a computation with PARNASSOS is spent on solving the large sparse linear systems of the form Ax=b. These are currently solved using an iterative solver such as GMRES, in combination with a preconditioner to improve convergence rates. Due to the increase in computational capability of modern computers, the performance of the linear system solver could be further improved through use of high performance computing. Especially the Graphical Processing Unit (GPU) seems to have great potential for speeding up such scientific computations. The aim of the work is therefore to investigate whether it is possible to achieve reasonable speedup of PARNASSOS' linear system solver by making use of GPU computing. This investigation was done using both model and fullscale test cases, both for increasing size of the systems to solve. Performance of different iterative solvers has been analysed on different GPU cards and has been compared against the performance, currently used on the CPU.3GPU computing; Preconditioning; CFD; Krylov solvers)uuid:c2edbffbb2244699984e545ade1b8815Dhttp://resolver.tudelft.nl/uuid:c2edbffbb2244699984e545ade1b8815UCFD based prediction method for the notional permeability of rubble mound breakwaters@van Grieken, Jesper (TU Delft Civil Engineering and Geosciences)Aarninkhof, Stefan (mentor); Antonini, Alessandro (mentor); Hofland, Bas (graduation committee); Henrotte, Johan (mentor); van den Bos, Jeroen (mentor); Delft University of Technology (degree granting institution)
In this thesis, a general method to predict the notional permeability for various structure layouts is developed. A numerical model is used to find a characteristic value for different structure layouts: the three structures for which Van der Meer determined the Pvalue, and a fourth structure of which the Pvalue has to be predicted. The Pvalue of the fourth structure is predicted by interpolation of the characteristic value of this structure between the characteristic values found for the structures of Van der Meer. The numerical model used is the RANSVOF model OpenFoam. The developed prediction method uses the total wave induced water pressure over the armour layer as characteristic value. The prediction method is validated for a structure layout for which (Kik, 2011) found a Pvalue of 0.37 by means of physical model testing. A Pvalue of 0.38 was found by (Kluwen, 2012) for the same structure layout by means of more physical model testing. The prediction method predicts a Pvalue in line with the Pvalue found by (Kik, 2011) and (Kluwen, 2012). The predicted Pvalue is 0.35 with this prediction method. A sensitivity assessment has been performed by varying the wave conditions, the porosity of the porous layers of the structure and the Forchheimer parameters applied < in the numerical simulations. Results show that the prediction is not very sensitive to changes in the wave conditions or changes in the Forchheimer parameters. And although not confirmed by tests, it seems that the prediction method is not very sensitive for changes in the applied porosities of the porous layers of the structure with unknown Pvalue. Therefore, the method seems to be a robust method and can be practically used in engineering projects. With this new prediction method for the notional permeability of a breakwater, it is possible to make a more accurate estimate on the notional permeability for more complex rubble mound structures. A more accurate estimate of the notional permeability will improve the accuracy of the initial design of the armour layer in early stages of a project. A next step to further improve the predictions of the notional permeability and to gain a better understanding of the damage of the armour layer of a rubble mound breakwater, is to research the correlation between the wave induced water pressure over the armour layer, the discharge through the armour layer and the flow velocity on top of the armour stones and to research how the combination of these three parameters has influence on the armour layer stability.&CFD; Notional Permeability; Breakwater)Civil Engineering  Hydraulic Engineering)uuid:acf41200c4f440048d24dcc5f8d6dd7cDhttp://resolver.tudelft.nl/uuid:acf41200c4f440048d24dcc5f8d6dd7cgAn experimental and numerical investigation of the aerodynamic characteristics of a flameless combustor1Huijts, Melchior (TU Delft Aerospace Engineering)Gangoli Rao, Arvind (mentor); Augusto Viviani Perpignan, Andr (graduation committee); Bohlin, Alexis (graduation committee); Schrijer, Ferdinand (graduation committee); Delft University of Technology (degree granting institution)NOx emissions cause harm to the human body, the world around us and our atmosphere, both directly and indirectly. Flameless combustion is a combustion regime that has the potential to reduce these emissions by more than 95%. Operation in this regime is possible when lowering the availability of oxygen and lowering the combustion temperature in the combustion zone. Key for this operation is the recirculation of flue gases using internal recirculation zones and mixing of these gases with the incoming combustion air and fuel. These aerodynamic characteristics have been researched experimentally and numerically for the DUT flameless combustor. Results show recirculation and mixing can be increased with decreasing the jet size, however with an increase pressure loss. Numerical investigation shows that simulation of this setup is challenging using RANS and that LES might be the only way to go forward.YFlameless Combustion; PIV; RANS; CFD; Recirculation; Entrainment; Experimental; Numerical
20190901)uuid:324b6057aeab4ad8ab245aa1a62819a0Dhttp://resolver.tudelft.nl/uuid:324b6057aeab4ad8ab245aa1a62819a0_3Dsimulation of multistage turbomachinery by means of a nonreflecting mixing plane interfaceFrancs Moll, V.Pini, M. (mentor)h Turbomachines have experienced a fast development over the last years, and are present in a wide range of devices involved in the generation of electricity and the transportation sector. Current challenges in both sectors have driven quite a number of developments regarding the increase of efficiency and use of new types of devices. Those changes have also affected turbomachinery design, and new innovative approaches are being used in order to solve current problems. As the design of those machines is sometimes done without lot of prior knowledge, new design techniques are required. Design optimization is usually involving a wide range of disciplines. When it comes to turbomachinery applications, this becomes more challenging as there are lot of parameters involved in the design. The heterogeneous range of applications imposes challenges from the designer point of view as different architectures may need to be considered, different working fluids, different number of stages, etc. In thi< s Master Thesis, the problem of multistage turbomachinery analysis will be undertaken. Simulation of multistage turbomachinery presents a number of limitations when it comes both to modelling of the problem and the use of CFD codes, among which the main one is the modelling of the statorrotor interface. To overcome this problem, a number of solutions are proposed, of which some are presented in the present report. The MixingPlane interface is one of the most used, especially when design optimization is considered. The MixingPlane interface creates an artificial interface between turbomachinery blade rows which forces the match of the average of physical quantities at both upstream and downstream sides of the interface. However, since in general specialised turbomachinery solvers are multiblock structured, current approaches based on the MixingPlane interface are not adapted to unstructured edge based solvers. This imposes limitations in the mesh types that can be used for simulating turbomachinery, and add some design constrains. In this Master Thesis the existing methods are analysed and extended to the edge based unstructured solver SU2. After implementing the proposed approach, a multistage turbine for which experimental data is available is simulated. By comparing the results with the experiments and simulations with another CFD solver, the implemented method is validated.4turbomachinery; CFD; mixing plane; multistage; NRBC<MSc Aerospace Engineering. Flight Performance and Propulsion
134#17#MT#FPP51.9900207,4.373714)uuid:9778a8246d724960a8124615a53c2c41Dhttp://resolver.tudelft.nl/uuid:9778a8246d724960a8124615a53c2c411Numerical modelling of aeratedwater wave impactsvan der Eijk, Martin (TU Delft Mechanical, Maritime and Materials Engineering; TU Delft Marine and Transport Technology; TU Delft Ship Hydromechanics and Structures)UWellens, Peter (mentor); Delft University of Technology (degree granting institution)A substantial part of structural damage for conventional vessels is caused by complex free surface events like slamming, breaking waves and green water. These events lead to the interaction of air and water where air can be entrained in water. The resulting air bubbles can considerably affect the evolution of the pressure caused by hydrodynamic impact loading. Besides a {cushioning effect on the impact pressure and a possible increase of the compressibility of the mixture higher than air, earlier documentation conclude that due to aeration the acting forces can increase due to long lasting time of the pressure and the resonance between oscillating air pockets & generated pressure waves.Nowadays, methods used to predict these forces, assume that the fluid is incompressible. This can lead to an {underestimation of the forces. The aim of the graduation project is to evaluate the effect of a homogeneous mixture of water and air on the wave impact, specifically for a green water event. Based on the documentation of the numerical method ComFLOW, for the evaluation an improved extended numerical solver in Fortran and MATLAB is built.ComFLOW is described as a robust finite volume method which can model these impact loadings by solving the governing laws continuity equation and conservation of momentum for a twophase seperatedflow using the firstorder fractional step method and the secondorder AdamsBashforth timestepping scheme. Even with a coarse grid, a clear distinction between water and air can be maintained by combining the VolumeofFluid approach with the local height function and a constant line reconstructed free surface (SLICiVOF). However, the method is not capable to model the compressibility for a dispersed flow by assuming that inertial forces on the aeration are dominant: a homogeneous mixture of air and water.In this report an extension is developed based on two volume fractions: the entrained air and air above the free surface, and Kapila's fiveequation model. Considering the implementation and improvement of ComFLOW itself, the representation of capillary forces is improved by a< new developed Continuum Surface Force (CSF) model for SLIC. Further, taking into account the transpose of the velocity gradient in the diffusive term, not done by ComFLOW, lead to no significant underestimation of the viscosity around the free surface for Re=O(35) and an error reduction of 2.7% for the 2D rising bubble. Besides the neglected term in the diffusive term, for the interpolation of the density around steep interfaces for SLIC, like wave impacts, the cellweighted averaging method leads to less spurious velocities around the surface than the gravityconsistent averaging method used by ComFLOW. The last revised implementation is a {modified local height function, documented by ComFLOW as strictly mass conserving, which is made more strictly mass conserving due not overlapping of the heights.The modelling of compression waves and bubble interaction is verified with the 1D shock tube and 2D helium shock bubble. For a simplified green water event, the results of the dam break tested for aeration levels 0.1, 1 and 5% showed that the effect of the aeration after the impact is most relevant and can achieve impact forces more than two times higher than the initial impact due to the combination of a plunging wave, mixing of water and air, air pocket oscillation and the longitudinal acoustic mode. The effect of aeration becomes significant for Ma=O(0.10). These observations are based on the dam break results with obstacle.LCFD; Volume of fluid; wave impacts; aeration; Compressible Flow; green water
20190202)uuid:b1c5e142e5244cf594d1bdfeca281897Dhttp://resolver.tudelft.nl/uuid:b1c5e142e5244cf594d1bdfeca281897MicroRamp Flow Dynamics8Casacuberta Puig, Jordi (TU Delft Aerospace Engineering)kHickel, Stefan (mentor); Groot, Koen (mentor); Delft University of Technology (degree granting institution)S
Microramps are passive flow control devices used to delay flow separation. Their use is widespread due to their reduced drag and structural robustness. We reproduce with Direct Numerical Simulations (DNS) recent Particle Image Velocimetry (PIV) experiments of the microramp flow performed at TU Delft to study the wake of a microramp immersed in a laminar and incompressible boundary layer. The microramp is a vortex generator which induces a pair of streamwise counterrotating vortices. The current literature identifies this structure as the main flow feature contributing to the increase of the nearwall momentum. The microramp is also a surface roughness element which can trigger laminarturbulent transition. The action of the induced vortices introduces a strong detached shear layer into the flow field, susceptible to KelvinHelmholtz (KH) instability. We analyse the microramp flow dynamics and the transitional mechanisms which develop in the microramp wake. Furthermore, we intend to contribute to the discussion on the microramp working principle, which has been put into question by other authors. We show the importance of the transitional perturbation development in the microramp functionality. %In the past decade, numerous computational and experimental studies performed inquired into the topology of the flow behind a microramp and its relation to the functionality of the device. <br/><br/>Downstreamtravelling streamwise vortices and transitional disturbances serve to the same purpose of increasing the momentum close to the surface. To examine their relative contribution in this regard, we numerically decompose the microramp flow field into a laminar steady state and a timedependant perturbation field. To achieve that, we apply Selective Frequency Damping (SFD), a numerical technique used to compute the steady solutions of globally unstable dynamical systems. SFD is a popular method nowadays and the preferred approach for aerospace applications. However, it has two casedependant model parameters which are key to the method's effectivity and efficiency, and whose selection remains a challenge in the literature. Not every combination of the model parameters guarantees the success of SFD, and even if so, the required< computational time may be so large that the approach is impractical. We provide the first rigorous analysis of the influence of these parameters to the functionality of SFD, leading to simple expressions and procedures for choosing them optimally. Furthermore, we prove that, under certain conditions, SFD is always able to stabilise a globally unstable flow configuration.ZMicroramp; SFD; Transition; Turbulence; Instability; DNS; CFD; Aerodynamics; Flow Control
2018090151.9900 N, 4.3754 E)uuid:2c7f8595243f4c429c26187bf4976bceDhttp://resolver.tudelft.nl/uuid:2c7f8595243f4c429c26187bf4976bcefModernizing Thruster Design: A Numerical Investigation of a Ducted Azimuthing Thruster in Oblique FlowPavlioglou, S.`Hopman, J.J. (mentor); Godjevac, M. (mentor); Van Terwisga, T.J.C. (mentor); Bulten, N. (mentor),The rudderpropeller, as the azimuthing thruster was originally called, can be rotated 360 degrees and is capable of delivering full propulsive power in any direction. The azimuthing thruster makes use of a mechanical transmission in order to deliver the power from the prime mover to the propeller. The good maneuverability and the absence of a need for a rudder of such a propulsion unit are counteracted by lack of detailed knowledge for this unconventional propulsion device. The field of uncertainty lies primarily in the very nature of the thruster, namely the fact that it is capable of rotating while operating. The oblique angle of the inflow to the thruster can be the source of a series of complex phenomena, not all of which have been systematically monitored and analyzed. The goal of the present study is: to investigate in depth the hydrodynamic characteristics of a ducted azimuthing thruster and to showcase the potential impact of findings on the current detailed design approach. By means of CFD numerical software StarCCM+, the thruster was modelled in a way which would allow the simulation of oblique inflow cases. A realization of a series of operating conditions for various advance velocities, RPM and steering positions of the thruster was followed by a thorough explanation of the observed physical phenomena. The information acquired by the numerical simulations was then compared with characteristic rules of thumb that represent the prevailing method of design nowadays. Finally, a few selected cases were used as the basis for the realization of a force propagation analysis with the purpose of comparing the bearing reaction forces of the propeller shaft to the respective values that arise based on simplistic rule of thumb calculations. Through the course of this project, valuable information was acquired for the behavior of the thruster unit in oblique inflow. The majority of the cases have been found to be in accordance with the imposed rules of thumb. However, a few cases demonstrated divergence from the predicted values in very large inflow angles.azimuthing; CFD; oblique; thruster; steering; bearing reactions; turbulence; computational fluid dynamics; rules of thumb; propulsion
20200401Marine & Transport Technology%Ship Design, Production and Operation)uuid:d795958e2b8d41039723bf55db1b0cb7Dhttp://resolver.tudelft.nl/uuid:d795958e2b8d41039723bf55db1b0cb7xSchiphol Interchange Station  Integrated design research for the wind and daylight performance of the building envelopeVan Kersbergen, D.J.ETurrin, M. (mentor); Heinzelmann, F. (mentor); Cuperus, Y.J. (mentor)The main aim of the research is to passively use the onsite energy to create passive climate comfort. Usually this these passive strategies are not integrated into the design process but added to the building design later on when the building design is already finalized. The main topic of this building technology research is to integrate sustainability concepts into the decisions making process of the building design. Out of all passive ways to us onsite energy, wind and daylight are chosen as the sub topics for this research in order to improve the architectural design. The envelope of the building filters/uses the wind and daylight ener< gy to climatise the building in a passive way. The wind research will transform the building envelope on a large scale while the daylight research will transform the building envelope on a small/component scale. Both of them will work together in improving the architectural design and the energy performance of the building.`wind; daylight; energy; climate; CFD; ecotect; grasshopper; tesselation; parametric; performance
20120127ArchitectureBuilding Technology)uuid:eda271970f434b83a2acd0bb564a7ef4Dhttp://resolver.tudelft.nl/uuid:eda271970f434b83a2acd0bb564a7ef4Prediction of unsteady nonlinear aerodynamic loads using deep convolutional neural networks: Investigating the dynamic response of agile combat aircraft,Papp, David (TU Delft Aerospace Engineering)qVoskuijl, Mark (mentor); van Rooij, Michel (mentor); Delft University of Technology (degree granting institution)New generation combat aircraft are expected to operate over extended flight envelopes, including flight at high flow angles and rapid maneuvers. Conditions beyond traditional limits are giving rise to nonlinear phenomena, such as flow separation, large scale energetic vortices, fluctuations etc. These phenomena have significant impact on aircraft performance and if not resolved accurately design uncertainties are increased risking lack of performance or even costly redesigns. Thus, accurate modelling of unsteady nonlinear aerodynamics is essential for modern and future combat aircraft.<br/><br/>Unfortunately, conventional modelling tools either lack the required fidelity or they are too expensive. Traditional, highlyefficient approaches are not suitable for modelling nonlinear flow phenomena. Concurrently, high fidelity Computational Fluid Dynamics (CFD) simulations are computationally demanding and therefore impractical in many cases. To enhance aircraft design, it is desirable to obtain models joining the best of these two worlds. A common approach is to distill high fidelity methods into ReducedOrder Models (ROMs) that can accurately approximate unsteady aerodynamics at orders of magnitudes lower costs than CFD. Relevant literature offers many different ROM techniques for varying purposes. Nonetheless, constructing such models is still challenging and currently there is no generally agreed method. <br/><br/>In the current thesis a ROM technique that may be applicable to wider ranges of problems and simpler to construct is sought. The objective is to obtain a model that can promote aircraft control design, performance assessment and structural analysis throughout dynamic maneuvers over complete flight envelopes. The thesis proposes a novel approach utilizing modern, deep convolutional neural networks (CNNs). The devised model consists of three main components. First it incorporates a geometry description constituted by coordinates of an aircraft CFD surface grid. Second, a primary encodingdecoding CNN predicts pressure distribution at the grid points of the geometry. The final and third part of the model is an auxiliary encoding CNN deriving integral aerodynamic loads corresponding to the pressure field predictions of the primary network. The model evaluates and produces instantaneous values. Given a maneuver, it proceeds in timesteps. The predictions of the separate instances are computed directly without the need of subiterations (as it would be the case for CFD simulations).<br/><br/>As a proof of concept, the model is applied to symmetric motions in the vertical plane at fixed Mach number and altitude. The subject of the investigations is the MULDICON configuration of the 251 th Science and Technology Organization workgroup of NATO. To fully exploit the advantages of reducedorder modelling, flow characteristics are inferred from a single excitation following an efficient system identification technique using Schroeder sweeps as input signals. The performance of the model is assessed by numerous test cases performed in CFD. First, steady conditions of varying incidence angles are investigated. Second, harmonic pitch and plunge oscillations around d< ifferent angles of attack at different amplitudes and frequencies are considered. Third, additional test cases of a linear pitch updown  and a climbing maneuver are studied.<br/><br/>Considering computational efficiency, the results show robust model performance. GPUaccelerated CNN calculations are conducted roughly 5000 times faster than CFD simulations. The primary network can accurately resolve the pressure distributions over large portions of the geometry. Lower surface predictions are very accurate. However, among certain conditions discrepancies are observable on the upper surface towards the wingtips. Still, the secondary network can predict corresponding aerodynamic forces accurately. In contrast, its moment predictions are sensitive to errors in pressure distributions. Consequently, moment predictions can largely deviate from reference data, especially when nonlinear phenomena are prominent. However, in many cases errors are attributed to insufficient regressor space coverage, i.e.\ certain input combinations are explored poorly by the Schroeder sweeps. Reconsidering system identification practices might mitigate those issues. Nevertheless, the thesis proves the applicability of deep CNNs to the problems at hand. Additionally, the results encourage further investigations.cReduced order model; Surrogate modelling; Convolutional Neural Networks; unsteady flow effects; CFD)uuid:54c220fbfcfb4e2c9dba40afc70e3b7fDhttp://resolver.tudelft.nl/uuid:54c220fbfcfb4e2c9dba40afc70e3b7fHAerodynamic interaction effects of circular and square ducted propellers3Mouro Bento, Hugo (TU Delft Aerospace Engineering)de Vries, Reynard (mentor); Veldhuis, Leo (mentor); Eitelberg, Georg (graduation committee); Avallone, Francesco (graduation committee); Delft University of Technology (degree granting institution)With the goal of decreasing the environmental impact of aircraft, ducted propellers emerge as an efficient propulsive alternative. Ducts are able to increase the thrust to power ratio of a propeller system by both producing thrust and/or lowering tip losses of propellers. In this thesis, RANS CFD simulations were used to analyse the possible impact of modifying a propeller duct shape from a circular to a square geometry. Initially, the two duct designs and the propeller were studied as isolated cases, in order to characterise their aerodynamic performance. In the installed simulations, the propeller was first modelled as an actuator disk, and afterwards with a full blade model. The results indicate two main disadvantages of square ducts. Square duct corners were found to be prone to separation, and to contribute towards the generation of strong vortices. This research can be important for the future study of unconventional ducted propellers, for example with applications to distributed propulsion concepts.Aerodynamic performance; Circular duct; Square duct; Propeller; Propellerduct interaction; CFD; Actuator disk model; Full blade model)uuid:fe21dbbe56ae40a8a4860490dcce9136Dhttp://resolver.tudelft.nl/uuid:fe21dbbe56ae40a8a4860490dcce9136The Development of Downhole Separators in Series, Using Design Models Based on Computational Fluid Dynamics Verified By Laboratory Experiments Saleh, K.AZitha, P.L.J. (mentor); Swanborn, R.A. (mentor); Bos, A. (mentor)
One of the major problems associated with oil and gas production is the large volume of produced water. Operators around the world are facing significant costs with the treatment and disposal of produced water. Downhole separation, a relatively new technique, has been developed to reduce the costs of produced water and increase oil production. Downhole separation is the technique where oil and gas from the produced wa ter is separated at the bottom of the well and reinject some of the produced water into another formation, while the oil and gas are pumped to the surface. The reduction in cost is owed to the downhole treatment of the produced water since most of the topside produced water treatment facilities are reduced in number. Since most of the produced water d< oes not reach the surface this creates an added value of minimizing the opportunity for contamination of underground sources of drinking water through leaks in casing and tubing during the injection process. The goal of this project was to design a downhole liquidliquid separator and to evaluate the performance at downhole conditions with the aid of computational fluid dynamics. The separation performance is evaluated experimentally. A dedicated test rig has been designed and built at ProLabNL, a sister company of Ascom Separation, to test the separation efficiency of the downhole separator. The designed system consisted of three hydrocyclone stages in series to polish the water to the desired injectate quality of 100 ppm oil in water, and was operated under downhole conditions, i.e. high temperature (70  80 C), high watercut (90  95%) and relatively large oil droplets (ranging from 500  1000 [?m]) dispersed in the continuous phase. The system design and the operational method are fully outlined. At the tailend of production, reservoir pressure is depleted causing increased sand production. In the existing commercial downhole separators, the solids that are produced are reinjected downhole leading to potential plugging of the disposal zone. The proposed downhole fluid separation system is equipped with a desander to flush the separated sand with the oil rich stream to the surface. Computational fluid dynamics was used to evaluate the pressure balance and volume flux balance of the internals. An erosion analysis was conducted to investigate the wear due to the sand influx. Furthermore, laboratory tests were conducted to evaluate the influence of a progressive cavity pump (PCP) on the shearing effect of an oil water mixture. The pressuredrop over the pump seems to play a cru cial role on the amount of droplet breakup which leads to a decrease in separation efficiency of the downhole separator.;DOWS; downhole separator; CFD; computational fluid dynamics
20180423!Civil Engineering and GeosciencesPetroleum Engineering)uuid:6286f9e2c24a430ca4fa9fb67b9558b4Dhttp://resolver.tudelft.nl/uuid:6286f9e2c24a430ca4fa9fb67b9558b4The Longitudinal Static Stability and Control Characteristics of a Flying V Scaled Model: An Experimental and Numerical Investigation/Palermo, Marco (TU Delft Aerospace Engineering)Vos, Roelof (mentor); Raju Kulkarni, Akshay (graduation committee); Veldhuis, Leo (graduation committee); Borst, Clark (graduation committee); Delft University of Technology (degree granting institution)Despite widespread research into the possibilities of improving aerodynamic efficiency, a plateau seems to have been reached for the conventional configuration. Hence, the potential of unconventional configurations are being investigated in recent years. Flying V is one such configuration that promises a lift to drag ratio about 24 in cruise conditions, an improvement of 25% with respect to the NASA Common Research Model that was used as conventional configuration benchmark. In addition, the flight dynamic characteristics of such an aircraft must be investigated to ensure flight safety. Subscale flight testing (SSFT) allows the characterization of flight dynamics using subscaled models. In order to mitigate the risk of loss of control situations, the static stability and control characteristics of the model must be investigated. This research work aims to support SSFT by designing a subscale model and assessing its aerodynamic characteristics by wind tunnel testing. The subscaled design is representative of a 4.6% geometrically scaled model of fullscale Flying V design based on Froude scaling laws.<br/><br/>To test the aerodynamics of the future flying model, wind tunnel testing have been conducted. Balance measurements of 4.6% scaled halfmodel have been collected in an open jet windtunnel. A total of three control surfaces are installed on the model and the two inboard have been prescribed, during the design phase, to provide pitching moment control authority. The shift in aerodynamic center at higher angles of at< tack has been registered and large deflections of the control surfaces have been noticed to influence the shift of the aerodynamic center up to 6% of the mean aerodynamic chord.<br/><br/>Aside of the experimental investigations, RANS simulations have been performed using the SpalartAllmaras one equation turbulence model. Although discrepancies have been identified between the wind tunnel and numerical results, especially in terms of drag and pitching moment coefficient, the CFD results are used to get a better understanding of the influence of vortical flows on the genesis of lift and drag over the scaled model. Based on the performed CFD simulations, it can be concluded that the performed CFD simulations are insufficient to reproduce the pitching moment behavior recorded during wind tunnel testing.<br/><br/>The designed subscaled model proved to be able to substain flight loads up to more than 2.5g at MTOM conditions. Based on the performed analyses, the center of gravity is suggested to be located between 1.33 and 1.39 meters behind the nose of the configuration. The deployed control surfaces can trim the aircraft up maximum lift coefficients between 0.6 and 0.7, depending on the location of the center of gravity, with an ultimate static stability margin equal to 4.4%. The results highlight that a reduction in pitching moment control authority would cause a reductions up to 20 % on the maximum lift coefficient achievable in trimmed conditions due to lack of control authority for forward location of the center of gravity. The designed model can therefore be used for future SSFT activities and landing speeds are estimated to be lower than 20 m/s for the proposed range of center of gravity locations at MTOM conditions.iWind Tunnel Experiment; Stability Analysis; Control Analysis; CFD; Center of gravity effects; AerodynamicFlying V Research)uuid:700fdbda04074aaa9a047e1f7a64b03eDhttp://resolver.tudelft.nl/uuid:700fdbda04074aaa9a047e1f7a64b03emDynamic behaviour of the Sognefjord bridge: Analysis and review of the world's largest floating bridge design?Hendriksen, Eduard (TU Delft Civil Engineering and Geosciences)Metrikine, Andrei (mentor); Tsouvalas, Apostolos (mentor); Hendriks, Max (mentor); Amico, F. (mentor); Delft University of Technology (degree granting institution)
A project is under way to replace the ferry crossings in Norway's highway E39 with fixed links, such as highway bridges or tunnels. This thesis research is on the crossing located at the Sognefjord, the widest and deepest of the straight crossings in the E39 highway. Previous thesis projects at IVConsult have yielded a design for a floating bridge supported on twentytwo pontoons. The bridge is moored using a subsea cable system. The bridge design reaches a height of 70 m at its 465 m wide midspan and is dimensioned on the basis of static calculations of the structural elements. <br/>The goal of this thesis research is to calculate the dynamic response of the bridge system to environmental loads and to determine if the current bridge design is sufficient in relation to this response.<br/>To reach this goal, several models have been developed for the structural elements that compose the bridge; the continuous bridge deck girder, the pylons supporting this girder, the floating pontoons supporting these and the subsea cable mooring system fixing the structure in place.<br/>First a structural model for the bridge structure is developed with special attention being placed on the subsea mooring system. For these cables an internal design is made and a calculation method is developed to model and determine the internal hysteretic damping in the cables due to interwire friction. Second, a mechanical model describing the linear dynamic response of the pontoons for small rotations has been developed. The pontoons themselves are modelled as rigid bodies. Third, the bridge deck girder is modelled as an equivalent EulerBernoulli beam.<br/>Finally, a load model is developed for the wave and current loads at the bridge location. Diffraction theory is used to ca< lculate wave loads on the large pontoons and the current loading is identified and modelled according to prevailing design codes. Six critical wave load cases are formulated. <br/>Models of the bridge structure are built using the SACS and Scia Engineer software packages. A nonlinear solver is written in Python to implement the cable model for static calculation, utilizing Scia Engineer's nonlinear solver. Verification calculations of SACS software results are performed. <br/>The steady state response of the bridge structure is calculated using SACS software for the six critical wave load cases formulated. In conjunction with this analysis, the cable damping is calculated according to the cable model. The bridge deck motion and cable fatigue damage are evaluated and are found to be well within design limits, leading to the conclusion that wave loading will not lead to critical failure in the bridge design. <br/>An analysis of vortex induced vibrations of the bridge system caused by crossflow loading of the bridge pontoons is performed. The analysis is performed using Ansys Fluent in conjunction with the SACS Dynamic Response module to model the fluidstructure coupling. A large sensitivity to vortex induced vibrations is found for the bridge system, several potential solutions to this problem are presented and recommendations are made for further research into this phenomenon for the bridge design. <br/>A verification calculation of the FluentSACS model introduced in this thesis is performed using a coupled wake oscillator model. The verification is based on only the crossflow motion of a single pontoon in the bridge system and yields comparable results in terms of load and displacement amplitudes for both models.zStructural dynamics; Vortex induced vibrations; Steel wire rope; Hysteretic damping; Wave loading; Floating structure; CFD61.083753, 5.503247)uuid:1843361f52ce4f80b651eec1e4e5633bDhttp://resolver.tudelft.nl/uuid:1843361f52ce4f80b651eec1e4e5633bwFlow analysis between two bluff bodies in a close distance platooning configuration: A Numerical and Experimental Study3van Tilborg, Frank (TU Delft Aerospace Engineering)van Raemdonck, Gandert (mentor); Sciacchitano, Andrea (mentor); Scarano, Fulvio (graduation committee); Casalino, Damiano (graduation committee); Delft University of Technology (degree granting institution)In the European Union greenhouse gas emissions of heavy duty vehicles make up 30% of the total caused by road transport. By placing vehicles in a platoon configuration the aerodynamic drag can be heavily reduced. The effect of platooning has been studied on both American and European type vehicles. However many of these studies only disclose the drag reductions and do not fully explain the flow behaviour between the models which in the end is what is causing the reduction in drag. Next to this studies done on European type vehicles mostly used timeaveraged simulations so a next step is understanding the effect of unsteady flow on the vehicles. In this study the flow field between two vehicles in a platooning configuration is analysed numerically and experimentally for varying intervehicle distance and front and rear drag reduction devices. The GETS model was used in this study which is a simplified model of a European heavy duty vehicle. For the numerical analysis Exa PowerFLOW is used which is a transient CFD package based on the LBM. Turbulence is modelled using VLES together with a RNG of the  equations. A study of the mesh size showed the sensitivity on the drag coefficient of the single model. <br/>The experimental analysis was performed in the OJF of the TU Delft at a Reynolds number of 3.9 x 105 based on the square root of the model. Next to balance measurements CVV measurements in the gap between the two models were taken in order to visualize the flow field.<br/>The single model showed a toroidal recirculation region in which a vortex ring can be seen. The drag force fluctuates at a Strouhal number of 0.073, side and lift force at 0.058 and 0.102 and 0.130 and 0.160. These fluctua< tions are caused by pressure fluctuations in the wake where the highest magnitudes are seen in the shear layer. The models with a smaller front radius saw a rise in drag with flow separation occurring for the smallest radius for the numerical results and for both smaller radii in the experiment. The addition of a boat tail lowered the drag of the models. With increasing tail angles the drag reduction also increases which is caused by the increase in base pressure. The tails also decrease the force fluctuations due to decreased pressure fluctuations in the shear layer.<br/>At the closest spacing of 0.10 times the vehicle length all tested configurations benefit from the platoon. The flow between the two models is made up of a toroidal recirculation region much smaller in size compared to the single model. In general the flow takes an Sshaped path between the two models. From the underbody of the leading model it either stagnates on the front of the trailing model or moves into the gap where it ends up in the upper of lower vortex or stagnates on the front or rear of one of the models. When a sharper front edge radius is applied to the trailing model more flow is deflected into the gap leading to lower pressure vortices and a higher stagnation pressure. A small vertical misalignment of the trailing model, which was seen during the experimental campaign, leads to a higher stagnation pressure on the bottom of the trailing model, also here more flow is deflected into the gap. At this spacing the sideforce fluctuations are a bit higher due to the vortices that leave the domain passing over the rounded edges of the model.<br/>At the middle spacing of 0.45 times the vehicle length not all configurations benefit from the platoon. The trailing models with the baseline frontal radius saw an increase in drag due to the lack of thrust generated by the rounded edges. The other models did see a drag decrease. At this distance the wake of the leading model is quite similar to that of the single model. Flow enters from the top, bottom and sides of the model and ends up in the recirculation region or it stagnates on the rear of the model before it leaves the wake and stagnates on the front of the trailing model or flows over the rounded edges where it accelerates and leaves the gap. The effect of a sharper radius applied to the trailing model has little effect on the flow field. When a tail is applied to the leading model the recirculation region has been reduced as was seen for the single models. Due to the upwash the stagnation pressure on the bottom front has been increased but the drag has decreased compared to the platoon without a tail due to the increased suction of the rounded edges. At this distance the force fluctuations are much higher compared to the single model. The magnitude of drag and side force can go up to 2 and 6.5 times the values of the single model depending on the configuration. This is due to the stronger vortices from the leading model which pass over the trailing model.<br/>For the last vehicle spacing of 0.91 s/L all the configurations benefit from the platoon again. For the baseline models the gains are only a few percent but for the trailing model with a sharper front radius the gains are higher than those at the middle intervehicle distance. This is due to the reduced stagnation pressure and almost undisturbed suction coming from the rounded edges. At this distance any effect on the wake of the leading vehicle has vanished. The unsteady forces still show an increased amplitude compared to the single model however compared to the middle intervehicle distance they are much lower.<br/>The comparison between the results from the numerical and experimental analysis is quite good. The drag and flow field results are very similar. For the pressure and the unsteady forces this is less the case. The mismatch seen in the pressure comparison can be caused by alignment errors that were observed during the experiment. Next to this while reconstructing the flow field from the experimental measurements the resulting coordinates were ou< tside the expected range. A manual coordinate shift was applied which may have resulted in additional errors. For the unsteady forces it is assumed that the ground plate and its interaction with the model as well as additional vibrations are the cause of the much higher fluctuations seen from the experimental data.<br#platoon; bluff body; wake; CFD; CVV)uuid:5a19d53360e4414ca713f4738236d669Dhttp://resolver.tudelft.nl/uuid:5a19d53360e4414ca713f4738236d669,Redesign of the Solution Algorithms in WandaTHuijzer, Leonard (TU Delft Electrical Engineering, Mathematics and Computer Science)van Gijzen, Martin (mentor); van der Zwan, S (graduation committee); Vuik, Kees (graduation committee); Heemink, Arnold (graduation committee); Delft University of Technology (degree granting institution)The Wanda software package developed by Deltares can used for simulating both steady state and transient fluid flow in pipeline systems. Steady state simulations are used for initial system design and transient flow simulations are used for doing water hammer analysis in pipeline systems. For both types of simulations a system consisting of both linear and nonlinear equations needs to be solved for the main unknown quantities, flowrate and head. This system is solved by linearising the equations using theNewtonRaphson method and solving the resulting system of linear equations. Currently, this is done by using a matrix solver from the proprietary IMSL numerical library which requires a paid license. The problem is that this solver sometimes either crashes or gets stuck in an infinite loop when dealing with singular matrices, while the proprietary nature of the library only allows for limited troubleshooting. The solution method therefore requires a redesign which should improve its robustness and maintainability. On the other hand, no ground should be yielded in terms of solution accuracy and performance.<br/><br/>The singularity of the matrices are caused by quantities being underdetermined either due to user error in network design or phase changes such as a closing valve. In this report, a graphtheoretic approach is taken to detect these structural singularities in the form of determining a maximum size matching in a graph representing the system of equations. This approach gives information about which quantity is undetermined where in the pipeline system. As an alternative, condition number estimation is implemented. Furthermore, the IMSL library is replaced by LAPACK, which is a lightweight and versatile alternative. The open source nature of LAPACK and its permissive license ensure its maintainability. Since thematrices are banded, the band version of the LAPACK algorithms can be used. The new solutionmethod is compared to IMSL and evaluated in terms of robustness, accuracy and performance.<br/><br/>The graphtheoretic method resolves the robustness issues of the IMSLbased method, while showing great performance. The information it gives is used to either correct the matrix and continue the simulation or output an appropriate error message, ensuring a userfriendly experience. Condition number estimation is too slow while also not being useful for further matrix correction purposes and is therefore disregarded. To improve the accuracy of LAPACK, iterative refinement is used. The maximum relative error measured over all the test cases was about 2.5%, resulting from an illconditioned case. Overall, the accuracy compared to IMSL is good, so that users will not be able to notice large difference in solution quality between the two solution methods. To improve the performance of LAPACK, Reverse CuthillMckee (RCM) is applied to reduce the bandwidth of the matrices. Using several test cases, the LAPACK performance is shown to be similar to that of IMSL when using RCM. In that respect, the transition from IMSL to LAPACK should also be flawless. A vendoroptimised LAPACK implementation did not yield any significant performance gains. It can be concluded that the new solution method is an improvement in terms of robustness and maintainability,< while showing similar solution accuracy and performance.<br/><br/>Future work can focus on improving the performance by calculating the RCM permutation only once every time step, as well as keeping the Jacobian constant during each time step. As matrix bandwidth determines the performance of the matrix solver, better methods to reduce the matrix bandwidth could also yield significant improvements. This could be achieved by either ordering the components in the pipeline system a priori, or simply using matrix bandwidth reduction techniques. Optimised LAPACK libraries as well as a different numerical library such asMUMPS could improve the performance even further.Wanda; numerical; linear; Deltares; singularity; detection; algorithm; graph; structural; bandwidth; LAPACK; CFD; pipeline; network; robustness; maintainability; accuracy; performanceApplied Mathematics)uuid:e429748960e2403ea2461b1ea4c4ea63Dhttp://resolver.tudelft.nl/uuid:e429748960e2403ea2461b1ea4c4ea633HighOrder Numerical Schemes for Compressible FlowsSatheesh Kumar Nair, V.Dwight, R.P. (mentor)Highorder numerical methods for Computational Fluid Dynamics have undergone significant fundamental developments over the last two decades owing to combined efforts from the applied mathematics and engineering communities. Even though loworder numerical methods are still the standard in industry, the increased requirements of engineering applications have led to significant scientific interest in developing efficient and robust numerical methods. Applications that would benefit from highorder numerical methods include Direct Numerical Simulations (DNS), Large Eddy simulations (LES), Computational AeroAcoustics (CAA) and vortex dominated flows. The objective of this thesis is to successfully implement and validate a fifth order traditional WENO scheme in a finite volume framework, for a solver currently being developed in the Aerodynamics group of TU Delft. A detailed literature study of classical numerical schemes has been performed along with a study of the traditional WENO schemes. The quality of results using the fifth order scheme is studied for a variety of test cases to study the shock capturing ability of the scheme. Implementing the finite volume WENO schemes includes the calculation of numerical flux at cell faces using Gaussian quadrature formulas. The effect of varying the number of Gaussian quadrature points while calculating the numerical flux is investigated. Also, the effect of the approximate Riemann solvers on the quality of results is studied by implementing four different Riemann solvers and studying the results for different test cases using these Riemann solvers. The test cases are governed by the inviscid Euler equations and deals with flow in the compressible regime. They involve shocks, other discontinuities and often also complicated structures in the smooth part of the solution which tests the design of the schemes to be nonoscillatory at the discontinuities and still gives a high order of accuracy in the smooth parts of the flow. Convergence tests of the error for test cases using the linear advection equation is used to study the order of accuracy of the scheme using different number of Gaussian quadrature points. The tests clearly show that the order of accuracy remains the same irrespective of the number of quadrature points used. This result is important as it allows simulation run with just one quadrature point which is less expensive, and saves memory. This result is highly relevant while running test cases for LES where very fine grids have to be used. WENO schemes have been considered to be too dissipative for LES in their traditional form. This is indeed true as seen by KelvinHelmholtz type small scale vortices (which are characteristic of high Reynolds number flows), even in the test cases using the inviscid Euler equations, due to the inherent dissipation in the schemes. However, this could be seen as motivation for using the WENO schemes for Implicit LES where no explicit subgrid scale models are used to represent the unre< solved scales. The different Riemann solvers exhibit different levels of dissipation and recommendations are made for the choice of Riemann solvers according to the application.;CFD; numerical schemes; WENO; Riemann solver; Finite Volume)uuid:c5ddec13e10349b48b84a8ad013c753cDhttp://resolver.tudelft.nl/uuid:c5ddec13e10349b48b84a8ad013c753cIMultifidelity CoKriging Optimization using Hybrid Injected RANS and LES4Fatou Gomez, Javier (TU Delft Aerospace Engineering)oHickel, Stefan (mentor); Dwight, Richard (mentor); Delft University of Technology (degree granting institution)The computation of complex turbulent flows design optimization processes is currently limited by the lack of accuracy of ReynoldsAveraged NavierStokes (RANS) in massively separated flows and the infeasible cost of multiple LargeEddy Simulation (LES) evaluations. A novel method is presented, injecting data from LES or other highfidelity source such as DNS into the RANS equations, forming a Hybrid Injected RANS (HIRANS) model. The aim is to construct a multifidelity design optimization framework that outperforms singlefidelity RANS and LES variants. Two different formulations, injecting a scaled version of the nondimensional anisotropic part of the Reynolds stress tensor and both isotropic and anisotropic components, are tested in the periodic hill case. A costeffective LES configuration is assessed, and the agreement of the RANS and HIRANS results with respect to the LES reference is investigated. The original geometry is parametrized using a hill width multiplier, computing several LES evaluations. The injection of LES information from the same geometry into HIRANS and the prediction capabilities when using interpolated data from different geometries are tested. A global design optimization process is computed, using singlefidelity RANS, HIRANS and LES Kriging and multifidelity RANSLES and HIRANSLES CoKriging surrogates. The objective function is based on a combination of turbulent mixing and total pressure losses. <br/><br/>The correction of the mean velocity components required of the injection of both isotropic and anisotropic components for the test case. The LES setup analysis yielded similar results to the reference data with one tenth of grid points and forty percent of its averaging period. The locally corrected HIRANS model successfully reduced the L2 norms of the Reynolds stresses with respect to LES to a third part of the original RANS values in the fiftynine LES samples tested, with a modest improvement in the mean velocity components. The nonlocal corrections yielded irregular results for the mean velocity components, with successful corrections of the Reynolds stresses despite the long distances in the parameter space and different flow features of neighbouring LES cases to interpolate from. In the optimization process, the CoKriging LESHIRANS was not able to outperform the CoKriging LESHIRANS and Kriging LES methods. It improved the initial prediction of the underlying function, but the surrogate yielded artificially low predicted errors far away from the LES samples, leading to an overly exploitative method. An error correction formulation combining two HIRANS fidelity levels was simulated using a modified Kriging believer criterion, outperforming the original formulation and achieving similar results as Kriging LES. The computational efficiency improvements for future research of the CoKriging HIRANS are suggested to be linked to an adequate error estimation integration into the surrogate model.RANS; LES; HIRANS; Turbulence; CFD; Interpolation; Optimization; Kriging; CoKriging; Aerodynamics; Periodic hill; Correction; Prediction
20181214)uuid:0caecfdb85ec4f5fb5a9cc98dcb9722aDhttp://resolver.tudelft.nl/uuid:0caecfdb85ec4f5fb5a9cc98dcb9722a7Simulations of steady and oscillating flow in diffusersKSchoenmaker, Lars (TU Delft Mechanical, Maritime and Materials Engineering)Boersma, Bendiks Jan (mentor); Pourquie, MAthieu (graduation committee); Pecnik, Rene (graduation committee); Delft University of Technolo< gy (degree granting institution)WIn this thesis diffuser performance will be simulated with computational fluid dynamics, which will be done with the program Ansys Fluent. The expanding geometry creates an adverse pressure gradient and under certain conditions there will also be separation.<br/><br/>At first a relative simple simulation will be done of a onedirectional incompressible turbulent flow. It will show how different turbulencemodels performat a Reynolds number of 15,000 inside a conical diffuser with an angle of 2=8, no separation is expected in this geometry. The turbulence models investigated are the k, the k and the RSM model. The mesh refinement is tested on produced accuracy of the results. The final results of the simulations are compared with both DNS (direct numerical simulation) and experimental results. All models produce different behavior, due to transport equations, algebraic models and empirical constants used. The deviations are dependent on geometry and flow conditions, where certain turbulence models are better for specific cases. The simulations showed different representations. The flow behavior of the k model was the most realistic, due to the correct core velocity although it showed flow reversal at the wall. Various parameters are reviewed, such as velocity profile, flow reversal, pressure coefficient, friction coefficient and turbulent statistics.<br/><br/>The oscillating flow will represent a case which is more closely to the flow seen inside a thermoacoustic engine. The geometry is a rectangular diffuser and the flow is compressible. It is a much more complex flow with frequencies ranging from 621 Hz during the simulations. Laminar, transitional and turbulent cases are simulated by Re numbers of 380, 580 and 740 with varying displacement amplitudes. The transitional kkl model is used, because of its ability to also simulate laminar and transitional cases besides only turbulent cases. Additional the k SST model is tested. The velocity profiles are not simulated well with the kkl model, which was caused by the under estimation of simulated turbulence near the wall. It was found that for both models separation will induce early and low in the diffuser and expand downstream as the cycle passes. As a result the Reynolds shear stresses show higher values earlier in the cycle. It was also seen that the reattachment would differ in pattern. The trend is found that separation begins earlier with increasing Reynolds number and increasing displacement amplitude. The minor losses, or irreversibilities, vary in accuracy, the effect of the displacement amplitude is not always seen for variables which are dependent on the magnitude of the pressure. In addition turbulence would not show an increase at the point of transition compared to turbulent cases. Both models seem to deliver deviating results, but the k SST models the cases better.<brCFD; model; simulation; kepsilon; komega; komega sst; kklomega; rsm; reynolds stress model; thermoacoustic engine; oscillation; turbulent; transition; diffuser; acoustic power; streaming; fluent; turbulence; flow; separation; incompressible; compressible'Sustainable Process & Energy Technology)uuid:4792857c1c2e4c05bc0f9ca1ece4499cDhttp://resolver.tudelft.nl/uuid:4792857c1c2e4c05bc0f9ca1ece4499c<Numerical modelling of bow thrusters at open quay structuresDe Jong, J.zVellinga, T. (mentor); Verheij, H.J. (mentor); Blokland, T. (mentor); De Koning Gans, H.J. (mentor); Labeur, R.J. (mentor) Introduction Bow thrusters are of great help for the navigation at quay walls, but the high and turbulent velocities can result in a bed load exceeding the strength of the bed or bed protection. To be able to design a stable bed the velocities at the bed need to be accurately determined. In design practise the velocities generated by a propeller are determined with formulae based on a mix of the momentum theory and measurements. The application of the formulae is often limited to cases for which measurements have been carried out and do not allow a secure d< esign for more complicated structures and the different velocity field of a bow thruster. Scale model measurements To improve the calculation of velocities on a slope, a large number of measurements were done by Van Doorn [TU Delft, 2012] for several scenarios with and without piles and resulted in an amplification of the design formula for some of his scenarios. To also predict the velocities for other scenarios these measurements are used to build and calibrate a numerical model. Numerical bow thruster implementation The open source CFD package OPENFOAM is used for the construction of this numerical model. As the implementation of a rotating propeller in the mesh will result in high computational costs and to allow a fine calibration of the propeller efflux, the propeller is simplified to an actuator disc. At the actuator disc an axial and tangential body force, varying over the radius, are added to the momentum equations in the OPENFOAM solver. Functions for both a ducted and a free propeller are simulated and show comparable results, the free Goldstein propeller functions are further applied. The coefficients are estimated based on the measured thrust and torque and calibrated to achieve a good fit to the measured efflux. A local increase of the turbulence at the hub and the propeller tip is not implemented in the numerical computations. Results Comparing the calibrated model to the measured diffusion in axial direction, shows a very good agreement and the numerical model nearly exactly computes the distribution as derived by Blaauw and van der Kaa. When comparing the velocities at the slope to both theory and the scale model measurements, it shows an underestimation of the velocities at the toe for steeper slopes, which is explained by unexpected velocities in the wall boundary layer, as a result of the wall functions in OPENFOAM. The model is exerted on different geometries (with piles) to get insight in the velocities for quay structures.OpenFOAM; CFD; Bow thrusters; Quay structures
20140130Hydraulic EngineeringPorts and waterways)uuid:20e84104201e42b5a67ea91b6f9c5557Dhttp://resolver.tudelft.nl/uuid:20e84104201e42b5a67ea91b6f9c5557ZComputational Method for EarlyStage Design Optimization of Naturally Ventilated TerminalsTrkcan, Okan Fehmi Saban Fred (TU Delft Architecture and the Built Environment; TU Delft Architectural Engineering +Technology)Nourian, Pirouz (mentor); van den Engel, Peter (graduation committee); Tenpierik, Martin (graduation committee); Delft University of Technology (degree granting institution)1This thesis focuses on the development of a computational model for earlystage design decision support of naturally ventilated terminal structures. The model is developed in a parametric Rhino Grasshopper environment paired with Python coding and CFD analyses through OpenFOAM. Optimization of air distribution parameters is performed with Galapagos evolutionary solver, while optimization of the subsequent geometry is done manually by means of the CFD results of the bestperforming variant. Within the thesis a background study is made by means of an interview and literature review, after which analytic calculations and CFD studies are used to test various options and narrow down the domain of solutions for the case of the design of a naturally ventilated terminal structure. Finally, the method itself is developed and validated with a case study that is also used during the development of the model. The result of the thesis is a rapid computational model that allows for the integration of CFD studies into the earlystage design of naturally ventilated terminals with an optimization for the stated objective function. Secondly, the CFD results can be used to give an indication of a more optimal geometry of the hall. The final selection of geometry and the validation are left up to the user to perform afterwards.optimization; CFD; parametric; airport; natural ventilation; thermal buoyancy; ground duct; air distribution; genetic algorithm; Evolutionary Algorithms; evolutionary solver)uuid:2099392381e< 744eaa2961f02893412f1Dhttp://resolver.tudelft.nl/uuid:2099392381e744eaa2961f02893412f1]Laser scanning modelling of a Cessna citation: For Computational Fluid Dynamics (CFD) StudiesKagkaras, A.@Teunissen, P. (mentor); Gorte, B. (mentor); Bucksch, A. (mentor)In Aerospace Engineering the field of Computational Flow Dynamics (CFD) studies the aerodynamic behaviour of aircrafts. What is currently being used to perform CFD simulations are Computer Aided Design (CAD) models of the airplanes, which are usually lowdetail industrial design models. Research of new methods for improving the results of the simulating process is of great importance. One method that can be tested in this direction is the creation of more detailed models of the actual airplanes via reverse engineering techniques. Laser scanning is one of these techniques. Here, a laser scanner is used to measure TU Delft s Cessna Citation. Afterwards, parts of the airplane are modelled by using suitable methods to process the obtained laser points.&laser scanning; CFD; Cessna; geomatics#Earth Observation and Space SystemsMaster of Science in Geomatics)uuid:ab209a8a079b44b1afb8f2159361a906Dhttp://resolver.tudelft.nl/uuid:ab209a8a079b44b1afb8f2159361a906VInvestigation of crown wall stability on top of rubble mound structures with OpenFOAM;Sigalas, Nikos (TU Delft Civil Engineering and Geosciences)IHofland, Bas (mentor); Antonini, Alessandro (graduation committee); Smith, Greg (graduation committee); Zoon, Arthur (mentor); Delft University of Technology (degree granting institution); Norwegian University of Science and Technology (NTNU) (degree granting institution); University of Southampton (degree granting institution)tIn the present study the stability of crown wall elements on top of a rubble mound breakwaters is investigated. The first step was conducting a literature review, in order to identify knowledge gaps. It was found that current design methods do not take the freeboard of the crown wall into account when calculating the vertical force acting on it. Recent research has established that when the freeboard is increased, this force is reduced while the portion of the crown wall base that is wet is reduced as well. However, there have been different approaches to increasing the base freeboard, that could lead to different effects on the loading. Moreover, parametric investigations regarding the vertical force have been very limited. Furthermore, the current design formulas assume that the maximum horizontal and vertical forces occur simultaneously but research suggests that there is a time lag between the two.<br/>Based on these knowledge gaps, the study goals were defined. In order to achieve these goals numerical model simulations were prepared and ran, where the freeboard was varied with two different approaches, as well as simulations with varied breakwater slope. The selected CFD numerical model is OpenFOAM making use of the waves2Foam toolbox, implementing the volume of fluid (VOF) method. <br/>It is found that the currently used empirical methods fail to predict the changes in loading for increasing freeboard. For an increasing base freeboard, less part of the base becomes wet and that the vertical force, as well as the critical weight are reduced. On the other hand, the horizontal force increased. It is concluded that when increasing the base freeboard by means of lowering the water level results in lower loading compared to an increase of freeboard by elevating the crown wall element. Additionally, for the latter approach, a larger portion of the base slab is wet. Also, it was confirmed that a recently proposed reduction coefficient by (Bekker et al., 2018) for calculating uplift pressures can provides more accurate results in the case of freeboard increase.<br/>Examining the uplift pressure distributions, it was found that for a zero base freeboard the pressure distribution follows an Sshaped profile, which with increasing base freeboard reverses. The peak pressure is located slightly inwards instead of the seaward end of the base, followe< d by a reverse peak. In order to propose a generally applicable profile shape more data are required.<br/>The results indicated the presence of a time lag between the maximum horizontal and vertical forces. This time lag results in lower critical loading on the structure than when assuming simultaneous maxima. Nonetheless, further research is necessary, as these findings are a result of only one wave condition.<br/>A finding which contradicts the predictions made with empirical methods is that gentler breakwater slopes resulted in higher loading. This is considered to be a result of an increased internal setup for gentler slopes.<br/>Further research is recommended, especially on tests with varying wave conditions and geometries, which should make these conclusions more generally applicable.crown wall stability; crown wall loading; uplift pressure; freeboard; time lag; phase lag; numerical modelling; CFD; VOF; waves2FOAM; OpenFOAM; wave impacts5Coastal and Marine Engineering and Management (CoMEM))uuid:7fce49ea3da0451494de15515f616190Dhttp://resolver.tudelft.nl/uuid:7fce49ea3da0451494de15515f616190*A CFD study of the Actuator Cylinder modelSaes, Thomas (TU Delft Aerospace Engineering)uSimao Ferreira, Carlos (mentor); Madsen, Helge (mentor); Delft University of Technology (degree granting institution)8CFD; VAWT; Vertical Axis Wind Turbine; Actuator Cylinder)uuid:e675fadcfd8d4cedb59e45b6faba6fddDhttp://resolver.tudelft.nl/uuid:e675fadcfd8d4cedb59e45b6faba6fdd>A Study of GlobalCoefficient NonLinear Eddy Viscosity Models\Dpke, Max (TU Delft Aerospace Engineering; TU Delft Aerodynamics, Wind Energy & Propulsion)rDwight, Richard (mentor); Schmelzer, Martin (mentor); Delft University of Technology (degree granting institution)%In this research a globalcoefficient nonlinear eddy viscosity model (NLEVM) is studied. This model stems from the inherent inability of the Boussinesq approximation to model anisotropy and therefore flow features such as: swirl, streamline curvature and secondary motions (Lumley, 1970; Pope, 1975; Craft et al., 1996). The focus lies on the limitations of using globalcoefficients calibrated on a squareduct flow when applied on a rectangularduct and a wingbody junction. The calibration is done with the Direct Numerical Simulation (DNS) results from Pinelli et al. (2010) at a Reynolds number of Re = 1, 100. It is shown that using a globalcoefficient NLEVM the velocity prediction on a squareduct and rectangularduct is successfully corrected, i.e. secondary motions are present. On an attempt to improve the corner flow separation on a wingbody junction poor performance is observed. Stability issues led to only 3 models converge out of 21. Differently to Bordji et al. (2014) who found a large corner flow separation reduction with a SpalartAllmaras Quadratic Constitutive Relation (SAQCR) turbulence model when compared to a standard SA model, the globalcoefficient NLEVM only showed limited corner flow separation reduction. Apart from correcting the anisotropy the nearwall resolution and treatment is found to be of large importance for flow field predictions. In the square and rectangularduct a wall damping function destroyed the secondary motion prediction, whilst in the wingbody junction improving the junction and corner flow prediction.1CFD; RANS; Turbulence Modelling; Machine Learning)uuid:5ffe749f81aa42d1aed52f07580e8078Dhttp://resolver.tudelft.nl/uuid:5ffe749f81aa42d1aed52f07580e8078JA numerical investigation into the aerodynamic characteristics of AeroCity
Van Sluis, M.Gangoli Rao, A. (mentor)ZAeroCity; WinginGround; WIG; ground effect; highspeed transportation; CFD; aerodynamics
20180310)uuid:3dd54665f48c4e489f57dc285cece612Dhttp://resolver.tudelft.nl/uuid:3dd54665f48c4e489f57dc285cece612wImpact of Turning Induced Shape Deformations on Aerodynamic Performance of Leading Edge Inflatable Kites: Master Thesis3Sachdeva, Shivaang (TU Delft Aerospace Engineering)qSchmehl, Roland (mentor); Folkersma, Mikko (mentor); Delft University of Technology (degree < granting institution)With growing energy demands and a need to switch to a sustainable source of energy key stakeholders are considering the use of high altitude wind energy systems. TU Delft and its start up company kitepower are key stakeholders investigating the commercial viability of this technology. With this goal the research group has developed several kite systems capable of accessing high altitude winds. It is believed that the low investment cost and high performance of kites could lead to a lower cost of energy. Concepts currently being considered involve a leading edge inflatable (LEI) kite that is controlled by an onboard control unit and is connected to a groundbased generator. Once the kite is deployed to the required altitude it enters a power generation stage where it is flown in a figure eight routine. This routine is controlled by the onboard control unit that pulls on tethers that are connected to the tip of the kite. This process is followed by a retraction phase where the kite is pulled back in. The goal of the system is to maximize energy production in the generation phase while limiting the energy consumed in the retraction phase. It is critical to assess and improve the kite design and its performance at all stages of operation such that the net power production can be maximized. While significant advancements have been made into the performance for normal flight there is a lack of research on the aerodynamic performance when there is control input to initiate a turn. The shape of the kite and the high angle of attack at which the kite is flown results in complex flow behavior involving separation, flow vortices, flow reattachment etc. This poses several challenges to maintain accuracy. A computational approach involving a steadystate Reynoldsaveraged Navier Stoke (RANS) simulation is believed to be a computationally viable mode of analysis to capture the flow behavior. This thesis details the approach used to improve results attained using this method and understand the influence of deformations associated with control inputs on the aerodynamic performance of the kite. A control input is simulated using a finite element model (FEM) with the Abaqus software by reducing the length of the right steering and increasing the length of the left steering line by 0.5 m. This results in the right side reducing its curvature and being pulled towards the kite control unit (KCU). Whilst on the left side the bridle lines are less tensed, leading to an increase in curvature. The turning performance is governed by the offset and variations in magnitudes of forces. Meshes are generated that attempts to minimize geometry alterations whilst still maintaining high quality. The influence of boundary layer parameters are investigated. A tradeoff is made where the influence of key parameters on accuracy and computational time is evaluated; where applicable improvements are made. Both the global as well as the local parameters of the kite in normal as well as turning orientation are analyzed. The results show that control induced deformations lead to a percentage reduction in the lift, whilst the effect on the drag is minimal. It is further seen that the kite initiates stall at an angle of approximately 40 degrees. The stall behavior is initiated at the midspan of the kite and gradually moves to the tip. The turning performance is measured by looking at the yaw moment. The magnitude of this parameter is linked to the offset and magnitude in force vectors at the tip. Due to the delayed stall at the tips, it is observed that the yaw moment increases beyond 40 degrees. Accuracy issues using this method were seen when performing a validation study on a profile similar to that of the kite. These issues could be due to limitations of the method or potential errors in the reference study. It is recommended to reevaluate the validation study before using this method for detailed flow analysis. It is concluded, that in order to fully trust the relevance of the results, one would have to have to conduct a validation study. However the< method s ability to address nonlinear flow effects within a limited time frame makes it a viable option for design optimization/system modelling.GLeading Edge Inflatable; Kites; Deformation; Turning; CFD; Aerodynamics)uuid:fdcf842311f04b33956e3e761635ac41Dhttp://resolver.tudelft.nl/uuid:fdcf842311f04b33956e3e761635ac41.Single Skin Kite Airfoil Optimization for AWES.Coenen, Roger (TU Delft Aerospace Engineering)VSchmehl, Roland (mentor); Delft University of Technology (degree granting institution)dAirborne Wind Energy is a technology where wind energy is harvested with tethered flying devices. Kitepower uses flexible leading edge inflatable kites, but these have a scaling disadvantage in that they become heavier with size. A single skin kite has the potential of negating this disadvantage while at the same time being more aerodynamically efficient. <br/>An airfoil of this type is therefore investigated using Computational Fluid Dynamics and optimized using Surrogate Modelling techniques. <br/>A hybrid mesh was generated with hyperbolic extrusion and triangulation. The RANS solver that was used produced good results.<br/>The results of the optimization were unsatisfactory. The parametrization did not provide enough local control and unique airfoil shapes. The surrogate modelling approach is promising due to the computationally expensive CFD analyses.2Kites; Airfoil optimisation; CFD Optimization; CFD
20201221)uuid:957b762614584c84a7ce80ab64a7a121Dhttp://resolver.tudelft.nl/uuid:957b762614584c84a7ce80ab64a7a1215Numerical Analysis of Bow Tunnel Thruster Performance Mohan, A.fHuijsmans, R.H.M. (mentor); Van Terwisga, T.J.C. (mentor); Godjevac, M. (mentor); Munts, E.A. (mentor)
A previous work at Royal IHC, following the observation from sea trials of trailing suction hopper dredgers (TSHD) that the turning performance of bow tunnel thrusters significantly reduced at slow forward speeds, studied the flow behaviour and performance of bow tunnel thrusters at slow forward speeds using Computational Fluid Dynamics (CFD). Simulations were performed, on model scale, first with a simplified wedge model of a containership and then with a simplified wedge model of a TSHD with the thruster modelled by an actuator disk. The numerical results deviated significantly from experimental results from literature. The present work aims to investigate this deviation. Simulations are also performed with a full thruster unit to compare with the actuator disk approach. Also, possible improvement in thruster effectivity is explored by altering the tunnel geometry of the TSHD wedge. First, a grid convergence study is performed on the containership wedge. Next, simulations are carried out for a range of forward speeds  for selected thrusts for the actuator disk approach, and corresponding rpm s for those thrusts for a full thruster unit. In case of the actuator disk approach, the numerical results show a steady increase of transverse force (Fy) with increasing forward speed, contrary to observations in sea trials. The circulation around the wedge, resulting from interaction of the jet from the tunnel and the crossflow due to the forward speed, results in the wedge to behave as a slender body in a flow. This results in a transverse force, analogous to KuttaJoukowski lift, that results in increased Fy. A full thruster unit in place of the actuator disk result in a different flow behaviour and variation of Fy with forward speed. In this case, an initial decline in Fy is observed before it increases monotonically. Comparison with experimental result of the containership wedge and numerical results by both actuator disk and thruster unit show considerable deviation. The transverse force due to the circulation around the hull presents a plausible explanation. The conclusions from the grid convergence study for the containership model are utilized in meshing the TSHD wedge. Simulations performed for a range of forward speeds, first with actuator disk approximation, indicated lowered thruster performance as was observed in sea trials. The pressu< re contours over the hull imply that the resultant of the forces around the inlet due to low pressure caused by high velocities of the flow into the tunnel and those around the outlet due to low pressure caused by high velocities of outflow is such that there is a net thrust deduction. Simulations with thruster unit also show lowered thruster effectivity. The tunnel of the TSHD wedge is then given a forward bend of 45o at the tunnel inlet and outlet in order to explore the possibility of improving thruster performance. It is expected that the changed geometry will result in better inflow at tunnel inlet, and the action of the crossflow on the jet will result in a straightened jet with lesser interaction with the hull compared to a jet from a straight tunnel. The CFD results, with actuator disk approximation, indicate that thruster effectivity improves in case of higher forward speeds and reduces for lower speeds. It is advisable to perform the study with smaller tunnel bending angles such that the jetflow interaction is favourable for the entire range of speeds.CFD; Bow tunnel thruster; ANSYS CFX; Nienhuis; trailing suction hopper dredger; thruster unit; actuator disk; model scale; tunnel thruster performance; thrust deduction; IHC!Maritime and Transport Technology)uuid:09dd8ba2303b470c8a8e4eeac2c44200Dhttp://resolver.tudelft.nl/uuid:09dd8ba2303b470c8a8e4eeac2c44200FlowMock Up: Feasibility and Method Analysis of the Flow Field Reproduction of an Open Cabriolet Vehicle in order to Subjectively Assess Draught PhenomenaGabrielse, B.C.Gerritsma, M.I. (mentor)TIn the automotive industry the development process is driven by a highly competitive market, calling for shorter and more costefficient development procedures. Computer aided engineering (CAE) is one possibility to achieve these reductions and becomes an inevitable tool in the development process. However, CAE does not provide the subjective evaluation, which still have to be conducted. In the development of the thermal comfort of open cabriolets, the subjective evaluation is dependent on expensive experimental methods and can only be conducted in a late phase of the development process, at which freedom of decision is low. Especially the thermal comfort in open cabriolets is very challenging as the passengers are surrounded by a turbulent flow field, exposing the passengers to velocities up to 40% of the driving speed. This leaves room for optimization, which is why the idea of a flow mockup is investigated. The procedure of a flow mockup is a combination of digital development and physical simulation. A flow mockup should simulate the flow field inside a passenger compartment of an open cabriolet from the results obtained from CFD simulations. The research objective is to concretise the idea of a flow mockup by identifying the requirements a flow mockup has to fulfil in order to be advantageous for the development process as well as to be able to replicate the flow field of an open cabriolet. This is done by analysing the actual development process and proposing a modified development process, which includes the use of a flow mockup. Furthermore, the characteristic factors, which influence the human perception while driving an open cabriolet, are identified from researches found in literature and analysis of measurement data. Finally, to investigate the technical practicability of a flow mockup, a preliminary CFD investigation is conducted to investigate the possibility of replicating the flow field, which is present inside the passenger compartment of an open cabriolet. The flow field simulation should be replicated in a closed space and without the dependency on the vehicle geometry.JCFD; Cabriolet; Thermal Comfort; Simulator; Flow MockUp; Program StarCCM+)uuid:57693312431545ffbf5682551871dc52Dhttp://resolver.tudelft.nl/uuid:57693312431545ffbf5682551871dc527Modelling of Scour Depth at Quay Walls due to ThrustersVan den Brink, A.J.W.]Vellinga, T. (mentor); Verheij, H.J. (mentor); Verhagen, H.J. (mentor); Blokland, T. (mentor)
Introd< uction Inland and seagoing vessels are equipped with (bow) thrusters. The use of these thrusters can cause scour of the bed alongside a quay wall. In order to assess the consequences of local scour due to bow thrusters on the design of the structure it is desirable to know the dimensions of the scour holes as a function of in time. Analytical relations for the calculation of scour, only determine the scour depth, but do not involve remaining dimensions of the scour hole and the development of scour in time. Model description A three dimensional quasisteadystate numerical model is developed, which describes the scour development alongside a vertical quay wall induced by thrusters of a vessel. The flow simulations, using the Realizable k? Model in the Computational Fluid Dynamics package OpenFOAM, provide the flow properties near the bed in order to calculate the bed shear stress. In the flow simulation the influence of the rotation of the propeller is neglected. A boundary adjustment technique is applied in order to move the mesh near the bed. In case the critical shear stress is reached, the morphology plays a role and the bed changes. This critical shear stress and the bed change are calculated, by applying an empirical relation for the erosion rate of sand, which is valid for both the high and low velocity regime of the flow. Every time step the bed level is updated and the hydrodynamics are calculated for the updated bed level. Results & Validation The hydrodynamics are validated separately from the model with erosion by applying a fixed (i.e. nonerodible) bed and comparing the results regarding flow velocity in the numerical model with physical experiments and analytical expressions. Several cases are elaborated with different distances between quay and ship, and keel clearances. It appears that for relative large distances between ship and quay the numerical model overestimates the near bed velocities. The erosion itself is validated with full scale tests regarding thruster induced scour, performed in the Port of Rotterdam. A maximum scour depth of 2.2 meter after 6 departures of the vessel is calculated with the numerical model, which is rather conservative compared to the measurements where a maximum depth of 1.751.85 meter is measured. The difference in results is probably due to presence of clay and silt in the Port of Rotterdam, however the numerical model is developed using an empirical relation for sand. Sensitivity From a sensitivity analysis it appears that the maximum scour depth is sensitive for the amount of fine sediment which is present in the soil. Coherent with this the porosity of the soil and the dilatation of the soil during erosion play an important role in the magnitude of the erosion rate. Recommendations It is recommended to improve the modelling of the propeller induced flow and to integrate this flow model with the current erosion model. Beside that an improved mesh near the bed is recommended. In this case flow properties, regarding flow velocity and turbulence, can be averaged over the water column in order to involve the higher order moments for the calculation of the extreme bed shear stresses. The length of this water column depends on the mixing length of the turbulence. In addition, it is striking that the width of the scour hole in both calculations and measurements, are smaller than the width of common applied bed protections. From an economical point of view it might be interesting to assess whether the current bed protection design is not too conservative.lScour; Bow Thruster; CFD; OpenFOAM; Soil Water Structure Interaction; Quay Wall; Bowthruster; Bed Protection)uuid:244669d3f4be4df8baa5dca9ba9fc323Dhttp://resolver.tudelft.nl/uuid:244669d3f4be4df8baa5dca9ba9fc323Reducedorder modelling for prediction of aircraft flight dynamics: Based on indicial step response functions investigating agile aircraft undergoing rapid manoeuvres3Ketelaars, Martijn (TU Delft Aerospace Engineering)Voskuijl, Mark (mentor); van Rooij, Michel (mentor); Sodja, Jurij (graduation committee); Eitelberg, G< eorg (graduation committee); Delft University of Technology (degree granting institution)During aircraft design,multiple tools are utilised to inspect the performance of the configuration. As the design matures, higher fidelity analyses are conducted to predict the flight dynamics of the aircraft. These analyses are conducted by using semiempirical relations, numerically analysing flow behaviour, conducting windtunnel tests and performing scaled test flights. However, Semiempirical relations might not hold for next generation aircraft and wind tunnel testing and scaled test flights are extensive and are also prone to accuracy issues. A best of both worlds can be found in numerical analysis. However, an increase in flow fidelity modelling comes with an increase in computational cost. Besides, complete analysis of all possible manoeuvres of a design increases the number of computations significantly. Current methods cope with this issue by using flight dynamics models based on so called stability derivatives, instantaneous values which couple flight state parameters to aerodynamic loads to predict aircraft flight dynamics. However, these models do not take into account time dependency. Therefore, these methods do not accurately predict the flight dynamics of agile aircraft, such as unmanned combat aerial vehicles, undergoing rapid manoeuvres where unsteadiness dominates flow behaviour. This conventional reducedorder modelling method, in which samples of the fullorder model are taken in the form of stability derivatives, causes design iterations to be analysed inaccurately. The objective of this report is to investigate reducedorder modelling for flight dynamics prediction, thereby comparing conventional techniques to a method which does take into account unsteadiness in flow behaviour.<br/>The method investigated is based on indicial step response functions, which are samples in the form of unsteady aerodynamic flow behaviour functions of the fullorder model. The idea is that once these samples are known, any flight manoeuvre can be analysed within minutes. Research found in literature has assessed some of the capabilities and limitations of this method, but not yet applied this to flight dynamics prediction. The research described within this report will address this gap by using two test cases. The first testcase is used to assess the assumptions made in literature, on aerodynamics loads modelling, by applying the method on a twodimensional airfoil in subsonic flow conditions. It was found that the indicial step response functions are indeed representing the fullorder model, thereby taking into account unsteady flow behaviour in aerodynamic loads prediction. In longitudinal motions, the angle of attack and pitch rate effect need to be taken into account to predict lift, drag and pitching moments. Multiple frequencies of the same manoeuvre can be analysed within minutes once the samples are calculated. Results show that the accuracy of the predictions becomes a tradeoff issue between samples calculated and accuracy required. The second testcase is used to apply the indicial response functions to flight dynamics prediction of an agile<br/>unmanned bomber aircraft undergoing fast manoeuvres. A longitudinaldirectional climbing manoeuvre was calculated by developing a flight dynamics model based on stability derivatives. The flow behaviour encountered during this manoeuvre was analysed to include highly unsteady and nonlinear phenomena (e.g. vortices and flow separation) at higher angles of attack. By comparing the results of themethod under investigation to the fullorder solutions, it was shown that aerodynamic flight dynamics predictions were accurate in capturing unsteady behaviour and weak nonlinear flow behaviour. However, the samples proved to be inaccurate in representing behaviour in highly nonlinear regions. Concluding, this means that indicial step response functions provide more accurate flight dynamics predictions than conventional stability derivatives in representing unsteady flow behaviour. The accuracy of the< predictions are highly dependent on the samples chosen. Several samples suffice to predict the unsteady behaviour for linear and weak nonlinear flow regions of the flightmanoeuvre. If surrogate modelling is applied, the method can become more computational efficient than conducting multiple fullorder timemarching numerical calculations. It is recommended that more research is performed on indicial step response functions<br/>in capturing highly nonlinear flow behaviour, as the research showed that the size of the samples affects the flow behaviour representation.cReduced order model; Flight dynamics; Surrogate modelling; Unmanned Aerial Vehicles; CFD; Modelling)uuid:ef4b9c51111f445fae272966a31e9d44Dhttp://resolver.tudelft.nl/uuid:ef4b9c51111f445fae272966a31e9d44HTowards the CSMCFD modelling of membrane wings at high Reynolds numbers`Steiner, Julia (TU Delft Aerospace Engineering; TU Delft Aerodynamics, Wind Energy & Propulsion)iVir, Axelle (mentor); Rajan, Navi (mentor); Delft University of Technology (degree granting institution)hThe field of Airborne Wind Energy is concerned with harvesting highaltitude wind power; the corresponding group in Delft, in particular, is investigating the suitability of a kite for this purpose. Due to the high complexity that such a system ensues, it is of interest to have a fundamental understanding of the critical aeroelastic modes of the structure in flight.<br/>The aim of this project was to further extend on the existing research efforts in this area by coupling a high fidelity ReynoldsAveraged NavierStokes solver with an improved version of the existing structural solver framework and validate the model on simplified test cases using a partitioned approach. The developed methodology uses the opensource Computational Fluid Dynamics solver foamextend with an adapter to the coupling library preCICE. For the structure model, an inhouse Python code based on a nonlinear shell element formulation is utilised.<br/>A literature survey revealed that while there exists an abundance of publications on strongly coupled FluidStructure Interaction problems, the majority of them are targeted at applications set in completely different flow regimes. Thus, the most difficult part of the project was to find appropriate validation data for membrane wings at high Reynolds numbers<br/>Finally, the capabilities of the method have been successfully showcased on a classic FSI benchmark case, and a partial validation on benchmark cases at more realistic Reynolds numbers has been carried out as well. Hence going forward a full quantification of the accuracy of the method for its target application range is recommended.<br3membrane wing; sailwing; FSI; CFD; kite; Windenergy)uuid:6640f075f0b94c0f9b4a756a8ea085fcDhttp://resolver.tudelft.nl/uuid:6640f075f0b94c0f9b4a756a8ea085fc^Testing the application of CFD for building design: Towards a CFD application as a design toolHunte, S.R.Vambersky, J.N.J.A. (mentor); Geurts, C.P.W. (mentor); Sluys, L.J. (mentor); Van Bentum, C.A. (mentor); Schipper, H.R. (mentor); Van Rooij, R.P.J.O.M. (mentor)_The goal of this thesis is to contribute to the understanding of the method CFD by engineers in determining wind loads on structures and ideally contribute to the development of a future design tool. The field of wind engineering is explored and wind tunnel and CFD modelling is discussed. Results determined with wind tunnel tests and CFD simulations are compared and verified. This is the focus of this thesis. Recommended actions for a guideline on postprocessing steps are presented. Conclusions that are drawn concern the walladjacent cell height, the use of turbulence models and simulation methods.4CFD; turbulence modelling; walladjacent cell heightDesign and Construction#Structural and Building Engineering)uuid:bd04b1ffc245403a9149b027d03a9a95Dhttp://resolver.tudelft.nl/uuid:bd04b1ffc245403a9149b027d03a9a95nExtending a RANS Solver with heat and pollution modules for dispersion problems in urban areas with vegetationOTierolff, Espen (TU Delft< Applied Sciences; TU Delft ChemE/Transport Phenomena)UKenjeres, Sasa (mentor); Delft University of Technology (degree granting institution)` Currently, 54% of the general population lives in urban areas. This number is estimated to increase to 66% by 2050. Urbanization is generally linked to several phenomena that are detrimental to the overall quality of life; the urban heat island effect, and the decrease of air quality due to pollutants. The addition of vegetation to urban areas is generally seen as the best measure to combat these phenomena, due to their suggested filtering and cooling capacity. Currently, both phenomena are studied as separate processes, but research suggest that the cooling power of vegetation is linked to the amount of pollution that is present. This calls for a numerical implementation that is able to accurately model both the filtering, and cooling effect such that the interplay between them can be studied in the future. In this work, an existing RANS k" solver is extended with the drydeposition model to determine the filtering capacity of the vegetation, and the leaf energy balance model to determine the cooling effect of the vegetation. Using the dry deposition model, we were able to accurately reproduce experimental measurements of the filtering capacity of a hawthorn hedge in an open field, obtained by Tiwary et al. [2006]. We assumed that the filtering took place due to both needlelike and broad leaf collectors. The same case was studied byS 1p and Benes [2016], who obtained similar results, but used a different mixing parameters of the collector types. We concluded that additional studies are needed to determine the importance of the collector mixing parameters for multiple species, before the dry deposition model can be considered as valid. Using the leaf energy balance model, we were able to determine the cooling power of a simple vegetation block, exactly reproducing the results obtained by [Manickathan et al., 2017]. Using the leaf energy balance model, we were not able to reproduce experimental measurements of the leaf temperature of potted impatiens, obtained by [Kichah et al., 2012]. This was caused by the nature of the flow, which proved to be barely turbulent and outside of the scope of our RANSk" solver. We concluded that the leaf energy balance model needs to tested against other experimental measurements, obtained under different flow conditions. As of yet, we cannot determine if the model is able to accurately reproduce experimental measurements.NCFD; Vegetation Flow; NavierStokes; Turbulence; Drydeposition; Transpiration
20200618;Applied Physics  track: Transport Phenomena and Fluid Flow)uuid:5e2ab613335c427784a68b55c7ee5bbcDhttp://resolver.tudelft.nl/uuid:5e2ab613335c427784a68b55c7ee5bbc?Simulations for an expanding gas jet with JouleThomson coolingdde Muijnck, Stan (TU Delft Mechanical, Maritime and Materials Engineering; TU Delft Fluid Mechanics)Henkes, Ruud (mentor); Schrijer, Ferdinand (graduation committee); Pourquie, MAthieu (graduation committee); Delft University of Technology (degree granting institution)This study was aimed at improving the accuracy of the model predictions for the minimum fluid and inner wall temperatures for cold, lowpressure startup of wells that produce oil or gas. Due to the large pressure drop over the well head choke, the socalled JouleThomson cooling will give a very low temperature of the<br/>expanding gas jet. Lowtemperatures can give brittle fracture of the material in the piping downstream of the choke. Models are needed to verify whether the material temperature remains above the lowerdesign temperature. For the model validation, Imperial College in London (on request by Shell) has carried out lab experiments with argon gas that expands through an orifice from 120 bara to 1 bara. Awaiting the results of the lab experiments, detailed simulations were carried out in the present study using the Fluent CFD programme.<br/><br/>The 3D, steady, compressible ReynoldsAveraged NavierStokes equations were solved with the SST k< " model for the turbulence. The considered configuration is the same as in the lab. It consists of an upstream chamber with argon at 120 bara, that expands through a 5 mm long orifice with 1.55 mm diameter, into a square outlet section with 50 mm sides and 500 mm length. The inlet temperature is 17 oC and the outlet pressure is 1 bara. The supersonic flow leaving the orifice reaches a maximum Mach number of about 9, just before a shock to subsonic flow is found. The jet reaches very low temperatures due to isentropic expansion, and reaches the isenthalpic expansion temperature of 196 K (or 77 oC) downstream of the shock. The jet reaches the sides of the outlet at a distance of about 100 mm.<br/><br/>The maximum Mach number of about 9 predicted by Fluent is higher than the value of about 6 found in a previous simulation study that used the STARCCM+ CFD programme. To verify the Fluent results, the distributions of grid cells was varied and the number of grid cells was increased. Also, a MATLAB programme was written that solved the inviscid compressible equations (Euler equations) for an axisymmetric jet. <br/><br/>This confirmed the Fluent results. In addition to the 3D square outlet section, also 3D and 2D Fluent simulations were carried out for a cylindrical outlet (using a hydraulic diameter of 50 mm). The maximum Mach number and the jet structure (velocity, temperature) are not affected by the side walls. This is because the side walls are sufficiently far from the jet.<br/><br/>Furthermore, the temperature and the heat transfer at the walls of the outlet section were investigated. Thereto both adiabatic and nonadiabatic walls were considered. The ambient temperature is 20 oC. Thermal boundary layers are formed along the side walls, that are exposed to a temperature of 196 K (the isenthalpic expansion temperature) in the centre of the pipe, up to a distance of about 1.5 m, where the outer edge of the boundary layer reaches the centre of the pipe. Thereafter the centre line temperature increases due to heat inflow from the ambient.Joule Thomson cooling; Underexpanded jet; Cooling; Mach number; Shockwave; gas dynamics; High pressure to low pressure; CFD; Computational Fluid dynamics; Fluent; Shell
20230118Solid and Fluid MechanicsFluid Flow team Shell52,3864 4,9018)uuid:df3b1b0294f6427cad844c4342dddf57Dhttp://resolver.tudelft.nl/uuid:df3b1b0294f6427cad844c4342dddf57hModeling of the exhaust plume of a submerged exhaust system: A numerical analysis of a submarine exhaustcKlapwijk, Maarten (Mechanical, Maritime and Materials Engineering; Marine and Transport Technology)van Terwisga, Thomas (mentor); Rotte, Gem (graduation committee); Pourquie, MAthieu (graduation committee); Nienhuis, B. (graduation committee); Kerkvliet, M.S. (graduation committee); Gerritsma, Marc (graduation committee); Delft University of Technology (degree granting institution)The Royal Netherlands Navy (RNN) operates four dieselelectric submarines, the Walrus class. The submarines sail submerged on electric engines and periodically recharge the batteries with diesel engines at periscope depth. The exhaust gases are dispelled at the back of the sail below water level. During the sea trials it was observed that the rising exhaust gases elevated the surface locally with a height of 1 to 2 m. This elevation limited visibility through the periscope, water flooded the air intake and the submarines could be easily detected. To remedy this problem model tests were performed. To enable evaluating several exhaust configurations for a replacement of the Walrus class a numerical model is studied to predict the surface elevation. <br/><br/>To model this situation a numerical method is used. The method applied in this work is the Volume of Fluid code ReFRESCO, which is a Reynolds Averaged NavierStokes (RANS) solver. Three test cases are studied, a rising bubble, a buoyant jet and an exhaust plume from a submarine sail. <br/><br/>The simulations for the buoyant jet show that the momentum in the jet is dominated by buoyancy rather than by the initial in< flow momentum. The influence of different turbulence models on the width of the jet and the corresponding surface elevation are investigated. The choice of turbulence model influences the distribution of air in the domain, but does little to the surface elevation. It is concluded that the KSKL model yields the most physical results where the air spreads out in the domain. This model also has a satisfactory convergence behavior. It is concluded that both the density of the exhaust gas and the velocity profile at the nozzle have little influence on the final result. A parabolic velocity profile instead of a uniform outflow does improve the convergence. The dominant numerical error is the discretization error, which is in the order of 12% for the surface elevation. To this end the L" norm of the iterative error must be reduced to a satisfactory value, in the order of 10^3 or smaller. Comparing the results with experiments is difficult due to a lack of proper experimental data, however it is concluded that the results are in a similar order of magnitude.<br/><br/>Finally the exhaust gases on a submarine sail are modeled. For the modeling three simplifications are made: only the flow surrounding the sail is modeled (the hull of the submarine is not modeled), the control planes on the sail are not modeled, and no incoming waves on the free surface are taken into account. Both in the experiments and in the simulations a pulsating behavior can be observed in the rising air. L" norms for the submarine modeling are generally in the order 10^3, but occasionally less. The numerical results are validated against the experimental results. It is concluded that the use of the RANS code ReFRESCO is possible for the modeling of a submerged exhaust of a submarine. The estimated uncertainty for the result is in the order of 15%.<brKSubmarine; Multiphase flow; Turbulent jet; ReFRESCO; CFD; Submerged exhaustShip Hydrodynamics)uuid:2f18b7fe598146689777851513ca3acfDhttp://resolver.tudelft.nl/uuid:2f18b7fe598146689777851513ca3acf>Analysis of a Hybrid RaNSBEM Method for Predicting Ship PowerRotte, G.M.xPredicting the equilibrium between a ship s required thrust and resistance is a complicated problem. Shipyards have to pay big penalties when a ship does not run at its design speed at a specified required power. An accurate prediction of the ship s selfpropulsion point is thus of huge importance. In the determination of the selfpropulsion point model scale tests are still decisive. An alternative for these model tests is found in CFD methods. This thesis work has provided an analysis of such a method: a hybrid coupling between a viscous (RaNS) solver for the ship s hull and a potential flow method (BEM) for the propeller.ZRaNSBEM coupling; CFD; power prediction; propellerhull interaction; effective wake field)uuid:bfcf5b18428848b48587918cc7899b86Dhttp://resolver.tudelft.nl/uuid:bfcf5b18428848b48587918cc7899b86dNumerical analysis of blood flow patterns in simplified aortic root with prescribed valve kinematicsFagioli, Giorgio (TU Delft Mechanical, Maritime and Materials Engineering; TU Delft Process and Energy; TU Delft ChemE/Transport Phenomena)Kenjeres, Sasa (mentor); Poelma, Christian (mentor); Delfos, Rene (graduation committee); Gijsen, Frank J H (graduation committee); Delft University of Technology (degree granting institution)7
The aortic valve is a valve made of three thin flexible leaflets which open and close in order to allow the unidirectional passage of blood from the left ventricle to the aorta. Because of this repetitive motion at each cycle, the valve is constantly exposed to large variations of pressure and hemodynamic forces. Deterioration in its peculiar functioning could lead to the arising of valvular diseases such as stenosis or regurgitation. A widely spread procedure to cure severely damaged heart valves is the replacement with artificial heart valves. However, up to now, the natural aortic valve properties and functionalities have never been equalised by any manufactured prototypes in terms of efficiency < and durability. This limitation has inspired many engineering researches with the purpose of gaining a complete understanding on the working principles of the valve and the evolution of the flow through it in order to enhance the performances of these medical devices. <br/>In this thesis, numerical simulations of blood passing through the aortic valve have been performed. In particular, this study proposes an alternative investigating approach which consists on prescribing the rigid motion of the leaflets a priori as taken from experimental values and without the employment of the structural solver. To execute this strategy, the Arbitrary LagrangianEuler method along with remeshing techniques has been employed in Ansys Fluent. The main objectives concern the study of the resulting hemodynamics in the aortic root and the validation with other investigating techniques such as experimental work and FluidStructure Interaction simulations.<br/>First, the methodology was tested in a twodimensional geometry which embodied a simplified representation of the aortic root. Secondly, a more realistic threedimensional case was analysed with the employment of a commercial bileaflet mechanical valve. In this case, two different methods were adopted to deal with the transition of the flow to a turbulent regime. Namely, the analysis was conducted both by the resolution of the unsteady ReynoldsAveraged NavierStokes equations with the k model and by maintaining a laminar approach. <br/>Validation of the method with available experimental data showed that the twodimensional case is well reproduced. On the other hand, mainly qualitative agreement is obtained for the threedimensional case. This suggests that the proposed approach is able to capture the main features of the flow to a certain extent and future investigations are required to obtain more accurate predictions of the flow.<br1aortic valve; valve kinematics; CFD; dynamic mesh
20190628)uuid:1a0b82791283470d87b2107e3ed90ec2Dhttp://resolver.tudelft.nl/uuid:1a0b82791283470d87b2107e3ed90ec2jSteel structure in an open car park  The influence of trapped smoke on the fire resistance of steel beams
Wong, Y.C.A car in an open car park is on fire. The fire and the smoke unleash a heat that causes the temperature to rise. The open car park is made out of a steel frame and to be designed for a fire resistance of 60 minutes. This car fire is a local fire whereas flashover doesn t occur. The crucial point of the fire is when the steel frame reaches the critical temperature. If this occurs, the steel will fail and might lead to the collapse of the building. Because it is an open car park, it is assumed the natural ventilation will lead the smoke out of the building very fast. But the smoke can be trapped between the beams, which are present below the ceiling as a support structure for the floor system. The aim is to study whether the influence of trapped smoke between beams in an open car park is significant or not. And is it necessary to take the smoke into account during the design of an open car park when making use of the Bouwen met Staal guideline (BmS guide), a guideline for the design of the fire safety of an open car park? In this study the influence of the smoke on the steel structure of the open car park was investigated by making use of a spreadsheet Car Park Fire (CaPaFi) and the software, Fire Dynamic Simulator (FDS). CaPaFi is a spreadsheet which calculates the steel temperatures without taking smoke into account. It is based on real car fire tests and the Eurocode. FDS is a program that uses computational fluid dynamic calculations to run a simulation (model) of a car fire in an open car park. It calculates the steel temperature of the steel beams taking the effect of smoke into account. All the information for FDS is written in a script beforehand and run afterwards. A single car fire and a 3 car fire were modeled and studied. To verify the calculations in FDS with CaPaFi, a single car fire model was put up. Because there were still some differences in the results, 4 more< models were set up with small adjustments in the first one to fine tune the model (model 2 to 5). According to the study 95% of the car fires are limited to maximum 3 (fully) burnt cars. So, 3 car fires were also taken into account and the smoke. There were two cases with 3 car fires: one with beam sticking out (model 6 with obstacles) and one with beams hidden in the ceiling (model 7 without obstacles). Based on the results of this study, it is found that the smoke which is trapped between the beams can be neglected when making use of the BmSguide to design the fire safety of an open car park for the steel beams with a height smaller than 500mm. Also a reasonable explanation could not be found of why the temperatures of the steel beams of FDS are lower than the ones calculated by CaPaFi. And in a 3 car fire model points that are further away from the car fire can be disregarded because their steel temperature calculated by FDS, even though higher than CaPaFi, will never reach the critical temperature. At last, it is still unclear why the increased number of soot doesn t have any effect on the temperature of the steel beams in the single car fire model.0Car park fire; Steel structure; FDS; CaPaFi; CFDStructural EngineerSteel and Timber Structures)uuid:fbf93d6e211f4b92a057956d694db315Dhttp://resolver.tudelft.nl/uuid:fbf93d6e211f4b92a057956d694db315ALatent Space Modelling of Unsteady Flow Subdomains: Thesis ReportEMulder, Boris (TU Delft Aerospace Engineering; TU Delft Aerodynamics)WHulshoff, Steven (mentor); Delft University of Technology (degree granting institution)Very complex flows can be expensive to compute using current CFD techniques. In this thesis, models based on deep learning were used to replace certain parts of the flow domain, with the objective of replacing wellknown regions with simplified models to increase efficiency. To keep the error produced by the deep learning model bounded, a traditional CFD model and deep learning model were coupled using a boundary overlap area. In this overlap area, the flow computed by the traditional CFD model was used by the deep learning model as an input. It was demonstrated that since traditional CFD model continuously feeds in reliable information into the deep learning domain, the error remains bounded. Furthermore, it was found that the accuracy of the deep learning models depends significantly on the random initial weights. Therefore, deep learning models trained differently must be carefully compared.UAerodynamics; CFD; Deep Learning; Latent Space; Autoencoder; Recurrent Neural Network)uuid:61b58a943a0141aebf8d2d105840bef4Dhttp://resolver.tudelft.nl/uuid:61b58a943a0141aebf8d2d105840bef4NThe effect of sloshing in partially filled spherical LNG tanks on ship motionsVan Twillert, M.J.Huijsmans, R.H.M. (mentor)This report describes two methods to analyse the effect of sloshing on ship motions. The first method applies linear potential theory on both the barge and the internal tanks. For the barge the radiation, diffraction and incoming wave potentials are solved. For the internal tanks only the radiation potential is calculated, as there are obviously no incoming or diffracted waves in the tanks. The potentials are then used to solve the added mass, damping and wave forces for the coupled bargetank system in the frequency domain. By comparing the frequency domain results with the scale model tests results a good agreement was identified for all loading conditions. On top of this the frequency domain approach is very fast and relatively easy to set up. However, linearity is assumed while in the scale model test nonlinear sloshing was observed. It is expected that nonlinearity will damp out the roll motion of the vessel and it is therefore expected that for increasing wave height the frequency domain solution will overestimate the response. It was also found that both roll RAO peaks are very sensitive to the wave direction. By comparing the roll RAO s for liquid and frozen cargo in the tanks it was found that sloshing significantly decreases the height< of the main roll RAO peak and also creates a lower barge natural frequency. The second method is a more complex time domain model that is based on a coupling between Volume of Fluid solver ComFLOW and ship motions solver aNySIM. In the coupled model the motions of the barge due to waves, without internal tanks, are obtained by using linear potential flow and are calculated in aNySIM. The ship motions calculated in aNySIM are used as input in ComFLOW that calculates the more complex and potentially nonlinear motions of the liquid in the spherical cargo tanks. The resulting sloshing loads are again used as input in aNySIM, creating a twoway coupling between the dynamics of the ship and cargo tanks. The time domain results showed a reasonably good agreement with the scale model tests. However, the second, sloshing induced, roll RAO peak is underestimated in the coupled model. This might be due to the type of waves used in the test, uncertainty in the RAO s and due to the properties of the coupling between ComFLOW and aNySIM. The time domain method is time consuming and complex to set up, but offers more insight in what is really happening. It was shown that nonlinear motions of the barge occur due to sloshing in the tanks. It was also found that with partially filled tanks the response of the barge is irregular when exposed regular waves. The influence of the pump tower was also investigated, which showed that due to the damping created by the pump tower the overall roll response is lower and the barge natural frequencies moves to a lower frequency.sloshing; spherical; tank; offloading; LNG; FLNG; anysim; wamit; comflow; CFD; diffraction; radiation; potential; flow; carrier; diffraction; radiation; waves; linear; nonlinear; RAO; roll; frequency; time; domain; barge; ship; cargo; liquid; scale; model; tests
20160317"Ship Hydromechanics and Structures%MSc Offshore and Dredging Engineering)uuid:9bc88456e5374965a874666e5edae0c0Dhttp://resolver.tudelft.nl/uuid:9bc88456e5374965a874666e5edae0c0(Numerical simulation of a cavitating jetSchouten, T.D.Van Rhee, C. (mentor); Keetels, G.H. (mentor)>cavitation; free jet; CFD; OpenFOAM; cavitating jet; RANS; LES
20170713Dredging & Deep Sea Mining)uuid:9ef11baf576c4eb0a84db507590b6000Dhttp://resolver.tudelft.nl/uuid:9ef11baf576c4eb0a84db507590b6000JSloshing: Topside Storage Tank Application on Floating Offshore Structuresvan Dijk, N.Offshore structures with partially filled storage tanks may experience sloshing of the cargo when exposed to waves. Inventive use on the topside result in storage tanks which are built as an integrated part of the deck structure. Weight control and available space is often a critical issue for offshore projects and can be improved by this application. CB&I have decided to carry out a research related to the occurrence of sloshing and impact pressures for these, so called, indeck tanks. The sloshing assessment procedure is an important part of the structural strength checks. Sloshing occurs when the natural period of the fluid coincides to the motions of the storage tank. Four factors mainly contribute to the sloshing phenomenon. Namely, tank dimensions, fill, fluid properties and motion characteristics. However, the complex, chaotic and nonlinear behaviour of sloshing makes it hard to predict or estimate impact pressures. Indeck tanks are applied at the topside of the Aasta Hansteen SPAR project, carried out by CB&I. The application of these tanks faced difficulties concerning the sloshing assessment procedure. There is no method applicable related to this situation. Therefore, a conservative method has been defined as a temporary solution. For future implementation of these tanks, better understanding and knowledge of fluid behaviour is essential. In order to tackle this problem, a CFD analysis is carried out in two phases and concludes with a statistical analysis in order to estimate sloshing impact pressures. The first phase relates to a general 2D CFD simulation for various cases. The second phase includes 2D long time simulations< of sloshing cases extracted from the first phase. The results of the first phase show that no sloshing occurs for the Aasta Hansteen SPAR related cases. Where the motion period of 60 seconds is too far away from the period of the 1st wave mode, which is around 8 seconds. FPSO related cases contain a period around 10 seconds and show sloshing impact behaviour. The impacts occur specifically for longer tank lengths and higher filling levels as these cases coincide better with the motion behaviour of the tank. Noted that the combination of input parameters for which sloshing occurs is highly dependent on the forced excitation on the tank, where sloshing behaviour is sensitive to changes of these parameters. Furthermore, a motion case analysis is added and different sea states are assessed from mild to harsh tank motion excitations. Resulting in sloshing for harsher sea states and higher accelerations. Overall, sloshing impacts conclude in the order of 100 kPa  300 kPa. The impact area includes the vertical wall and 2.4 meters on the top of the tank. In the event of nonimpulsive oscillating behaviour (no sloshing), one can apply the linear theory for an accurate prediction of the pressures. However, when the fluid motion becomes chaotic and nonlinear, there is no method able to accurately predict the impact pressures. The results of the second phase contain the sloshing impact order of magnitude for eight individual sloshing cases. With difference in fluid, tank length, fill and motion type. Six of these cases can be compared to one another and resulted in a fill/length ratio of 0.063 for the highest impact pressures. A lower viscosity of the fluid seems to increase the sensitivity to sloshing behaviour. Filling levels of 50\%  70\% show high sloshing impacts, where 80\% fill does not result in sloshing anymore. The increase of tank lengths results in higher sloshing impacts. Briefly summed up: 7m no sloshing, 12m semi sloshing, 15m sloshing impact order 150 kPa  200 kPa and 20m sloshing impact order 300 kPa  500 kPa. Two fitting curves are used in order to establish the Exceedance Probability Function. Namely, the Generalized Pareto and Kernel Smoothing. Both show a good fitting, but present different behaviour in the so called 'tail' of the Probability Density Function. A good distribution of this 'tail' result in a better Exceedance Probability Function. A decision on the best fitting curve is not made due to the lack of sufficient simulated statistical data. The sensitivity analysis proved the Kernel Smoothing fitting more robust compared to the Generalized Pareto. Also, the reduction of statistical data resulted in the highest sensitivities within the sensitivity analysis. Which underlines the need of more simulation and statistical data for improvement of the results.6Sloshing; CFD; ComFLOW; Indeck Tank; Impact Pressures
20170823Marine and Transport Technology)uuid:632afd2e81de42e3a25e3e2d4eb1aad4Dhttp://resolver.tudelft.nl/uuid:632afd2e81de42e3a25e3e2d4eb1aad4*Numerical Study of Flow around Bypass Pigs Liang, X.KHenkes, R.A.W.M. (mentor); Breugem, W.P. (mentor); Hendrix, M.H.W. (mentor){In the oil and gas industry, pipeline networks are used to transport the production fluids from wells to production plants. During normal operation, the pipelines need regular cleaning and inspection. Typically, the pipeline maintenance is performed by pigging, which refers to using devices known as pigs (Pipeline Inspection Gauges) to perform various maintenance operations of the pipeline. In order to describe the motion of the pig in the pipeline, detailed understanding of the flow around the pig is required. In this research, a CFD (computational fluid dynamics) approach was applied to model fully turbulent flow (Re 107) around various types of bypass pigs. We especially focused on the relation between the overall pressure drop, which was represented by a dimensionless pressure loss coefficient, and various dimensionless parameters describing the flow and the configuration. The pressure loss coefficient is caused by the < fluid that passes through the bypass area. If the pressure loss coefficient is known, together with the friction between the moving pig and the pipe wall, the motion of the pig can be described. Moreover, often the flow in the pipeline is in multiphase (stratified flow) condition. Therefore, in this research the effect of multiphase flow around a bypass pig was also investigated. For the single phase study, two types of bypass pigs were investigated: the disk pig and the complex bypass pig. The disk pig has a fixed and relatively simple geometry, and it is based on the conventional bypass pig, with a deflector plate in front of the pig body. The complex bypass pig geometry is based on the disk pig, though now the bypass area is created by holes which can be adjusted. In reality, for these complex bypass pigs, the bypass pig velocity is controlled by adjusting the bypass area. For the conventional bypass pig, previous studies have shown that the Idelchik s correlation for thick orifices can predict the pressure loss coefficient accurately. Thus a similar approach was applied in the disk pig study in order to obtain a theoretical correlation to predict the pressure loss coefficient for the disk pig. Indeed such a correlation was found which gives an accurate prediction for a certain parameter regime. In the complex bypass pig study, we mainly focused on the influence of the bypass area fraction on the pressure loss coefficient. Two correlations based on two approaches were suggested. It was found that these correlations can predict the overall pressure drop across the complex bypass pig accurately, especially when the opening fraction of the bypass adjusting holes was relatively large. Furthermore, for the multiphase study, the simpler pig models were investigated. First of all, the flow in front of a pig without bypass region was investigated. One of the practical purposes of this study is that we want to investigate under which condition the full pipeline perimeter gets wetted with liquid. This is important for the distribution of corrosion inhibitors. Moreover, the multiphase flow around a (conventional) bypass pig was investigated, to obtain a better understanding of the multiphase flow behaviour for bypass pigs.CFD; Bypass pig)Sustainable Process and Energy TechnologyEnergy Technology)uuid:71f03228175e40b09fd25be4480dcfecDhttp://resolver.tudelft.nl/uuid:71f03228175e40b09fd25be4480dcfecSmart Wind Designing Ras, T.T.2Homans, T.C. (mentor); Van der Zaag, E.J. (mentor)A research on finding a method to integrate the wind engineering process to the design process. Wind engineering is seen to be a rigid and slow process of CFD modeling and wind tunnel testing. New frames of work and data analysis make it possible to integrate this process in the, much faster, design process. Proof of this integration is given in the design of a dwelling complex in the centre of Amsterdam Teleport.SWind Engineering; Architecture; Building; CFD; Membrane facades; Amsterdam Teleport
20140408&Architecture and The Built EnvironmentArchitectural Engineering)uuid:f8b6c47cb08e46d79358488602c50973Dhttp://resolver.tudelft.nl/uuid:f8b6c47cb08e46d79358488602c50973`Numerical Study of the Flow and Heat Transfer in Supercritical Water Based Fluidized Bed ReactorZeng, C.*De Jong, W. (mentor); Harinck, J. (mentor)Public version. This version is only limited to the summary. Supercritical water gasification (SCWG) is a novel process for the thermochemical conversion of wet organic waste to gas and minerals. It is an alternative to the anaerobic digestion process, dewatering followed by drying and incineration and conventional dry gasification. Its advantages include no requirement for drying, higher syngas yield and much shorter residence time. From societal aspect, SCWG offers a solution to environmental problems caused by wet waste and fossil fuels, through the production of renewable gas and minerals. TU Delft and Gensos B.V. collaborate in fundamental research on supercritical gasification using a fluidized bed reactor concept. This stud< y focuses on the hydrodynamics and heat transfer of the fluidized bed reactor of the supercritical gasification process. It aims to develop CFD models that can correctly predict the main heat transfer and fluidization phenomena in the supercritical fluidized bed reactor. CFD models are developed to conduct numerical studies of singlephase heat transfer without fluidization, multiphase fluidization without heat transfer and fluidization with heat transfer. The CFD models are validated by using experimental data from Yamagata, Mokry and Lu et al.Bsupercritical water gasification; CFD; heat transfer; fluidization
20200901Energy Technology Section)uuid:ed6cc506622d49be9b3c0ab396b9c3d5Dhttp://resolver.tudelft.nl/uuid:ed6cc506622d49be9b3c0ab396b9c3d5ANumerical study of boundary layer transition behind a zigzag trip
Kerkvliet, M.gElsinga, G.E. (mentor); Pourquie, M.J.B.M. (mentor); Westerweel, J. (mentor); Boermans, L.M.M. (mentor)5
In this thesis a numerical investigation was performed to increase our understanding of a low Reynolds number transition behind a zigzag trip. This was done with the use of Direct Numerically Simulation (DNS). This means that the NavierStokes equations are solved numerically without any use of a turbulence model. The DNS solver of the CFD package ANSYS Fluent is used. The results from the DNS are compared to experimental data from an earlier study on the transition region behind the same type of zigzag trip [Elsinga and Westerweel, 2011]. In this experiment the undisturbed laminar boundary layer was formed over a smooth flat plate with a nominally zero pressure gradient. At a distance of 145 [mm] downstream the leading edge the laminar boundary layer was then tripped by a 1.6 [mm] high zigzag trip. In our DNS study the Reynolds number with respect to the experiment was increased in order for turbulent flow to develop behind the trip. The Re k based on the freestream velocity and the height of the trip were 672 and 336 for the DNS and the experiment respectively. In general the statistical data of the DNS in terms of mean velocity and velocity fluctuations showed good agreement with the experiment. The shapefactor of the boundary layer, which is a more robust integral parameter, showed very good agreement for the total domain. In this thesis particular emphasis will be on the study of the generation and the shape of the large and small scale turbulent structures and their dynamic behavior directly behind the trip. Small scale vortices are detected on the leading edge of the trip as well as streamwise directed large vortices behind the trip. Further it was found that the generation and breakup of hairpin vortices, as identified by the second invariant of the velocity tensor gradient, showed variations for different inflow angles and geometrical changes. The shedding frequency and vortex amplitude are affected by the recirculation bubble behind a downstream pointing spike. After the shedding of the hairpin structures they are propagated downstream, where they grow in spanwise width and wallnormal height till they breakup into smaller structures, i.e., smaller hairpin structures or small streamwise directed rolls. Also the scatter distribution of the second and third invariant of the velocity gradient tensor, respectively Q and R, are studied in the transition region directly behind the trip. This scatter distribution shows the expected teardrop shape which is described by Ooi et al. [1999] and is thought to be related to a certain state of the development into a fully turbulent boundary layer flow. The teardrop shape appears already closely behind the trip, which indicates the production of threedimensional structures by the trip. Further studies on this subject are desired. Furthermore it is found that the angle of incidence of the mean flow, in the range from zero till twelve degrees, had a marginal influence on the transition to turbulence. However, changing the spanwise wavelength of the zigzag to 50% and 200% of the original wavelength, resulted in significant differences in the transitio< n into turbulence. In this study, the wide trip, i.e., larger spanwise wavelength, had a shorter transition region compared to the original trip and the narrow trip had a longer transition region compared to the original trip.\CFD; zigzag trip; transition; boundary layer tripping; speedskaters; wake reduction; gliders
MEAH  263)uuid:581b8923d82c470d9b1e8cd6b8a3b44bDhttp://resolver.tudelft.nl/uuid:581b8923d82c470d9b1e8cd6b8a3b44b{Analysis and Simulation of an Anode Supported Solid Oxide Fuel Cell Single Channel for Operation with Biosyngas and MethaneDimitriou, E.D.Aravind, P.V. (mentor)A threedimensional thermofluid model coupled with electrochemical reaction for an anodesupported planar SOFC has been developed to investigate the internal processes and temperature distribution within a single cell unit for the design proposed by the Energy research Center of the Netherlands (ECN) in cooperation with TU Delft. The SOFC developed is for operation with natural gas. In this work, operation with biosyngas is evaluated. The electrochemical reactions, heat and mass transfer phenomena between the solid and gas phases have all been included in the cell model. The object is to investigate the complete cell using the models developed and make comparisons for the cell performance when fed with hydrogen, methane and different biosyngas compositions, come up with suggestions for efficient and ideal SOFC operation conditions when fed with different fuels and investigate safety issues under different working conditions. In order to clarify the goals of this work, the following questions have to be answered Is operation with biosyngas safe for the SOFC (Ni oxidation, carbon deposition)? Is operation with methane safe for the SOFC (Ni oxidation, carbon deposition)? What is the impact of different fuel compositions on SOFC performance? What is the impact of the steam reforming reaction of methane on the SOFC performance for operation with biosyngas and methane? What constructive suggestions can be made for further development of SOFCs?BFuel Cell; SOFC Solid Oxide Fuel Cell; Sustainable; Biosyngas; CFD!MSc Sustainable Energy Technology)uuid:43faf68c7e5f459c9d1b88ec7c4c71b9Dhttp://resolver.tudelft.nl/uuid:43faf68c7e5f459c9d1b88ec7c4c71b9STowards a physicsbased understanding of fruit frost protection using wind machines0Heusinkveld, Vincent (TU Delft Applied Sciences)van de Wiel, Bas (mentor); van Hooft, Antoon (graduation committee); Delft University of Technology (degree granting institution)
Wind machines are used to prevent or mitigate the adverse effects of night frost in spring. These frost events occur during clearsky, lowwind nights in which a thermal inversion builds up from the surface. The machines work by mixing and transporting warm air from aloft downward which consequently erodes the thermal inversion. Various studies have been conducted regarding this frost protection method, which have resulted in empirical regression models that relate affected area and temperature enhancement (i.e. the performance) with inversion strength. In this study, we assessed heightdependent temperature responses and the sensitivity of wind machine performance to various physical processes. Both of which have not been studied thoroughly in earlier works. In this regard, a large field experiment was conducted and experimental analysis was augmented with sensitivity studies using turbulent resolving Large Eddy Simulations (LES). Experimental observations showed that the temperature response strongly depends on the radial distance to the fan and the height above the surface. In agreement with previous studies, the wind machine was able to achieve rotationaveraged temperature increases of up to 50% of the inversion strength (H"3 K) in an area of 35 ha at 1 m height. Furthermore, it was observed that lowspeed ambient winds (<1 m/s) can cause strong upwinddownwind asymmetries in the protected area, the downwind area being larger. The LES model, inspired by the field experiment, showed similar spatial temperature response< s as compared to observations. Sensitivity studies using a simpli_ed case showed that the a_ected area strongly increased for slower axial rotation times (ranging from 3 to 6 min) while the temperature enhancement stayed relatively constant. Furthermore, variation of the horizontal tilt angle showed that, in our model, temperature enhancement was maximized between 8 and 16. Presumably, those angles, corresponding to near horizontal ow, are very effective in generating strong shear layers which in turn generate KelvinHelmholtz like instabilities. These processes increase mixing and vertical transport of heat. Finally, analysis on the ambient wind showed that, in agreement with observations, strong upwinddownwind asymmetries in the affected area exist._Frost protection; Fruit forst; Wind machine; Inversion; Agriculture; CFD; largeeddy simulation
20200701)uuid:ae6fc82c152d442899d5d531cc3ae6baDhttp://resolver.tudelft.nl/uuid:ae6fc82c152d442899d5d531cc3ae6ba#CFD Analysis of Piston Cooling Jetselik, Halil (TU Delft Aerospace Engineering)The piston is the only moving part in the combustion chamber of an internal combustion engine. With the ever increasing demand for higher power output, the control of piston temperatures has become one of the determining factors in a modern successful engine design. Because of the highly dynamic environment created by the reciprocating piston, experimental testing of the cooling process is expensive and difficult. In this research project, the first steps are made towards a CFD model, that can be used to analyse the piston cooling process at DAF Trucks NV. This thesis concerns the numerical investigation of the oil jet by using LES and the RANS analysis of the impingement heat transfer. A more realistic situation is analysed by combining these results. The results show that both the jet behaviour and the resulting heat transfer can be improved by varying the operating conditions of the piston cooling jets.+CFD; Impingement; LES; RANS; piston cooling
20230301)uuid:ec68055b990a45b5af269dd67372f6d1Dhttp://resolver.tudelft.nl/uuid:ec68055b990a45b5af269dd67372f6d1DLES and Unsteady RANS simulations of Multiple Jet Impingement SystemPenumadu, P.S.Rao, G.A. (mentor)DJet impingement is a subject of extensive research over the years due to its industrial importance and the fundamental physics of heat transfer and turbulence. These jets generate high heat transfer rates with better uniformity on the surface which is to be heated or cooled compared to other heat transfer techniques. However, to study the flow phenomenon and heat transfer rates, conducting experiments overtime for change in design has become expensive. Hence, to reduce the cost and time, the best possible way to study the jets behavior is to perform numerical simulations. With numerical simulations, one can predict the flow physics inside the domain, which is difficult to obtain from the experiments. In the recent times, there have been tremendous developments in terms of computation power and numerical models that have proved to produce good and accurate results for many applications. With these approaches, the flow characteristics can be studied in depth at each and every time step which provides better understanding about the jets behavior inside the array. However, these new models have to be tested and should be compared with experimental data to see how good these approaches can fit with the experimental results. Moreover, LargeEddy Simulations (LES) have not been carried out in the past to investigate the flow features in an impinging jet array. So, this master thesis is focused on performing LES and Unsteady Reynoldsaveraged NavierStokes Simulations (URANS) for multiple impinging jets and validate these CFD results with the experimental data. The primary objective of the thesis is to predict the pressure drop characteristics across the nozzles and the flow channel as the pressure drop directly affects the efficiency of the system. Simulations with Reynoldsaveraged NavierStokes (RANS) is a good method to an< alyze these quantities, but it was found that through RANS approach, the results obtained from CFD simulations are over predicted than the experimental values and this deviation increases with increase in Reynolds number. So to analyze these quantities, transient simulations are performed to predict the flow physics and heat transfer characteristics. From the unsteady RANS and large eddy simulations, it was understood that the major pressure loss oc curs inside the nozzles and due to contraction effect at the nozzle s inlet. It was also observed that inside the nozzles the pressure drop occurs differently for the particles near the wall and particles which are in the mean flow. The pressure drop values obtained from the CFD simulations are validated with the analytical and experimental results. It was found that, the results are in good agreement with the analytical results. Fur thermore, the heat transfer characteristics obtained from these transient simulations also show a substantial improvement compared to RANS models. The deviation in the results were found to vary between 7%  10%. To study the effect of geometrical parameters on the heat transfer and pressure drop characteristics, a sensitivity analysis was performed by varying the nozzle to plate distance and hole diameter. It was noticed that, with the change in nozzle diameter by 10 microns, the total pressure drop of the impingement system is affected significantly. Therefore, when designing these precision systems, it is important to manufacture them accurately as, a minor change in the jet diameter would affect the system performance on a large scale. Finally, It can be concluded that, the unsteady RANS simulations would be a good approach to study the heat transfer characteristics and flow physics inside the array. However, it would be a difficult task for any turbulence model to accurately predict the pressure drop characteristics in the impinging array, as the pressure drop inside the array is extremely sensitive with the change in geometrical parameters.7LES; Jet Impingement; Unsteady RANS; CFD; Multiple jets)uuid:57fecdc34987439ca6fcf2cf6a1d70c3Dhttp://resolver.tudelft.nl/uuid:57fecdc34987439ca6fcf2cf6a1d70c3Numerical Investigation of Spray Formation in AirBlast Atomizers: Numerical study of airblast atomization using a hybrid volume of fluid/discrete phase solverHPal, Botond (TU Delft Aerospace Engineering; TU Delft Space Engineering)Zandbergen, Barry (mentor); Roekaerts, Dirk (mentor); Cervone, Angelo (graduation committee); Delft University of Technology (degree granting institution)The conducted study investigates the potential of a newly released multiphase solver to simulate atomization in liquid rocket injectors. The "VOFtoDPM" solver was used to simulate primary and secondary atomization in an airblast atomizer with a coaxial injectorlike geometry. The solver uses a hybrid Eulerian/EulerianLagrangian formulation with a geometric transition criteria between the two models. The conducted study assumed isothermal, nonreacting flow at room temperature. The primary focus was predicting Sauter Mean Diameter and droplet velocity data at a sampling plane downstream of the injection site. The results showed that the solver is able to produce the expected data and to predict trends similar to those found in experimental measurements. The accuracy of the produced droplet diameters was roughly a factor 2 off compared to experiment. This is attributed to mesh resolution. Measurements were obtained via a cooperative agreement between TU Delft and The University of Sydney. It was concluded that the solver has the potential to predict atomization at a reasonable computational cost, but further study is needed to confirm its full capabilities.,Atomization; Multiphase flow; CFD; Injector)uuid:1b6037cf6cb94ebfb9c225e7fa6cf575Dhttp://resolver.tudelft.nl/uuid:1b6037cf6cb94ebfb9c225e7fa6cf575TNumerical investigation of nearshore wave transformation and surf zone hydrodynamics:Patil, Akshay (TU Delft Civil Engineering and Geosciences)Reniers, Ad< (mentor); Bricker, Jeremy (graduation committee); Lashley, Chris (graduation committee); Jacobsen, Niels G. (graduation committee); Lowe, Ryan J. (graduation committee); Osterloo, Patrick (graduation committee); Delft University of Technology (degree granting institution)%Rapid climate change and the corresponding estimated sea level rise can affect the performance of the coastal defense structures such as breakwaters, seawalls, and dikes. In order to improve these coastal defenses, a detailed understanding of the processes which contribute to wave runup and overtopping over the coastal defenses needs to be established. Following the exponential growth of computing capacity around 1970 s, a wide variety of computational models were developed to study fluid flow. Traditionally, three computational paradigms have existed in order to study wave transformation and surf zone hydrodynamics: phase averaged models, phase resolving models, and Computational Fluid Dynamics (CFD) models. Limitations posed by the underlying linear wave theory in phase averaged and other simplifications in the phase resolving models, may not provide sufficient detail in wave breaking, wave energy dissipation, wave runup, wave overtopping, and potentially other detailed hydrodynamic processes. This lack of resolution in depth averaged models for wavebreaking, wave runup, and wave overtopping processes motivates a detailed investigation using CFD based models, which can correctly mimic wavebreaking and other hydrodynamic processes.<br/><br/>The recent growth in available computational capacity has greatly improved the applicability of CFD based models for large scale transient flows such as waves near a coast. Additionally, the developments in wave generation and wave absorption boundary conditions by Jacobsen et al. [2012] in the opensource CFD toolbox OpenFOAM, have facilitated the use of OpenFOAM in coastal engineering applications. This encourages investigating the coastal environment using relatively complex models, thus providing insights into fundamental processes which contribute to coastal safety. To that end, this thesis focuses on investigating wave overtopping and the underlying processes which contribute to the aforementioned hydrodynamic aspects.<br/><br/>Overtopping demands accurate capture of the free surface (interface between water and air). The waveFoam solver suffers from numerical diffusion of the interface, consequently requiring a different approach to mimic the sharp interface. In order to cater to this deficiency, a new solver which combines the capabilities of waveFoam [Jacobsen et al., 2012] and isoAdvection [Renby et al., 2016] which has the ability to capture sharp interfaces by means of a subgrid approach has been integrated (waveFlow) and used in this study. In addition to the new solver, a new set of Reynolds Averaged NavierStokes (RANS) closures developed by Larsen and Fuhrman [2018] for wave modeling applications have been employed to correctly capture turbulence levels under breaking waves. The preliminary steps include calibrating and assessment of the newly integrated waveFlow solver. Using a relatively simple conceptual test case, a comparison of the free surface behavior and overtopping discharge was carried out. This calibration test was followed by a comparison of numerical results with the experimental investigations carried out by Ting and Kirby [1994]. Following this benchmarking study, experimental studies carried out by Flanders Hydraulics investigating wave overtopping over dikes in shallow foreshore environments was validated. A comparison of waveFlow and waveFoam was made to assess the qualitative and quantitative differences between the two interface capture methods on overtopping. Using this new solver, OpenFOAM was able to reproduce the surface elevation and significant improvement in the overtopping results were obtained for identical model setup in comparison to the waveFoam solver. A coupled approach using a potential flow solver named OceanWave3D aided simulation of large domain wave propagation and he< lped to cut down the computational time.COpenFOAM; isoAdvection; Wave Breaking; Overtopping; CFD; Turbulence)uuid:e55916dbe9464bc099f3df3b768760cfDhttp://resolver.tudelft.nl/uuid:e55916dbe9464bc099f3df3b768760cf>Fire dynamics and spalling mechanism in tunnel infrastructures<De Poli, Matteo (TU Delft Civil Engineering and Geosciences)Hordijk, Dick (mentor); Blom, Kees (mentor); Ravenshorst, Geert (mentor); Lottman, Bas (mentor); Snel, A.J.M. (mentor); Delft University of Technology (degree granting institution)Underground infrastructures and the importance of the people s safety and structures robustness are relevant and contemporary issues in the civil engineering industry. The demand for this infrastructural typology has developed to such an extent that the need for a better and costbeneficial knowledge of the risks is necessary.<br/>This thesis aims at bridging and developing further the knowledge on Fire safety engineering and structural engineering on the particular topic of spalling failure. Another objective of this thesis is the discussion over the possible replacement of the used procedures used to assess and guarantee tunnel safety. Both from a fire safety and structural engineering point of view, prescriptive measures and solutions are mostly proposed to accomplish a safe tunnel design. This causes the design to be non optimised and in some cases more costly than what is actually needed.<br/>From the fire safety side, the use of pregiven fire curves is put under discussion. Research has been conducted to asses which are the origins of the most widely used curves. Studies have also been performed to develop a practical analytical engineering method to analyse and estimate the consequences of a given fire scenario tailored to the specific tunnel under consideration. Subsequently the results of the analytical models have been compared with the results obtained with advanced Computational Fluid Dynamics tools. Finally, more complicated scenarios have been studied with the use of this software.<br/>On the other hand, from a structural engineering point of view, a new model able to describe the spalling mechanism has been proposed. The model predicts the spalling time for NSC elements and at the same time verifies which is the optimal thickness for the piece to spall. On top of that the possibilities for further use of the model in the description of the spalling mechanism for HSC and PPFRC elements have been investigated.<br/>Finally this two topics have been combined together and conclusion have been drawn.<brDTunnels; Underground infrastructures; Tunnel Design Guidelines; Fire safety engineering; Design fire curves; Fire scenarios; Performance based design; Prescriptive design; CFD; FDS; Fire temperature curve; Spalling failure; Spalling analytical model; Concrete failure mode; Concrete tunnels; Fire protection; NSC; HSC; PPFRC,Structural Engineering  Concrete Structures)uuid:1ff626fb177d44b18095091923eb8f3aDhttp://resolver.tudelft.nl/uuid:1ff626fb177d44b18095091923eb8f3a{Fluiddynamic characterisation of smallscale organic Rankine cycle radialoutflow turbine for renewable energy applicationDruzdzel, P.Since the problem of global warming has been broadly considered as vital and due to constantly rising oil prices, further fossil fuels exploitation requires diametrical changes immediately. The demand for environmentally friendly improvements does not only entail searching for new resources but also utilizing those which are available and more importantly, renewable. Organic Rankine Cycle (ORC) is possibly the most flexible technology in terms of temperature level and capacity nowadays. It is often the only applicable means of conversion for external energy sources and is therefore a very active research area in the field. Project investigates performance of an unconventional, smallscale, threestage radial outflow, ORC turbine as this component has proven to be critical in the system. Main content embraces steady and transient computational fluid dynamic simulations by means of Ansys CFX software < and evaluation of loss mechanisms. The main project objective is to evaluate the performance of the machine at its design conditions and fluiddynamic characterisation of the flow. Geometrical data has been obtained from TU Delft inhouse meanline code zTurbo for preliminary fluiddynamic design of turbomachinery. The main findings of this work involve performance evaluation of an innovative miniturbine configuration and the applicability of available loss estimation models. Although there are guidelines on designing conventional, larger, axial power plants, still little is known about research in the field of smallscale, radial outflow machines working with molecularly complex fluids. Overall totaltostatic efficiency of the machine is 65.3% for a 0.1 mm tip leakage gap and sealed stators. It is higher, and in close proximity to the one of 63.1% obtained by meanline prediction tool zTurbo. Thermodynamic processes undergone by the working fluid in miniORC ROT can be represented by velocity triangles at midspan with a 2.2% discrepancy in overall efficiency w.r.t. twoequation, steady RANS. Relatively uniform blade load, ability to apply tight clearance, tangential deflection in rotors due to Coriolis force and the relative motion w.r.t. the casing contribute to a decrease in 3D effects. This efficiency is expected to be slightly lower if tip leakage is imposed also on the stationary domains, possibly with even better match. Total pressure loss coefficient for the first stator, for the steadystate case without tip leakage is of exactly the same value (23%) as the one for transient simulation with freeslip endwalls, accounting only for twodimensional effects, averaged over the oscillation period. For the rotor, where the endwall boundary layer is more disturbed, steady state case predicts this loss by 3 points higher w.r.t. transient case, which gives 28%. Together with small secondary losses predicted in the stator, accounting for only 14% of the overall loss, 2D estimation/optimisation methods can be possibly used to predict the performance. Naturally growing passage area, allowing for smaller flaring angle, contributes to a decrease in spanwise velocity components and diminishes 3D effects. Stage 1 proved to be more susceptible to tip leakage increase. Small aspect ratios, in the order of 0.5, in miniORC ramp up the impact of tip leakage vortex compared to larger blades. The poorest performance (totaltototal efficiency of 51.2%) has been noted for the third stage, which requires profile optimisation and stagger angle modification. Expectedly, increase in tip leakage gap in rotational domains, leads to rise in entropy, which is even slightly more perceivable in lowaspect ratio blades, such as Rotor 1. Good match has been found between CFD and zTurbo in terms of overall loss estimation. Discrepancies for Rotor 1, Stator 3 and Rotor 3 are 0.1%, 0.1% and 0.2%, respectively. For Stator 1, Stator 2 and Rotor 2 these discrepancies are higher, of 3.3%, 7.7% and 3.0%, respectively. The main design problems are: profile losses and 2D design, including shock interaction shown in the results from the transient simulation. It is expected that this work will make a longstanding contribution to the body of knowledge on loss mechanisms in smallscale ORC machines and it will help to build a more sustainable and cleaner world.ORC; ROT; organic; Rankine; cycle; radial outflow; renewable; energy; miniturbine; turbine; centrifugal; CFD; RANS; unsteady; transient
20170130)uuid:0ab1c5b34a674b21b572a0384abdc576Dhttp://resolver.tudelft.nl/uuid:0ab1c5b34a674b21b572a0384abdc576gThe influence of underhood flow on bluff road vehicles in a platooning configuration: A numerical study4van Rijsingen, Mart (TU Delft Aerospace Engineering)]van Raemdonck, Gandert (mentor); Delft University of Technology (degree granting institution)The transport sector is making a significant contribution to the global CO2 emissions causing Global Warming. A large portion of this is caused by heavyduty vehicles like tractor semitrailer combinations. Sin< ce tractor semitrailer combinations are often operated at relatively high speeds for long periods of time their fuel consumption, and with that their emissions, can be strongly reduced by reducing the aerodynamic drag. This can be done by improving the aerodynamics of individual vehicles, by carefully rounding their leading edges or application of drag reduction devices, like boat tails and side skirts. Another option is to use the benefits of drafting by operating two or more vehicles closely together in a platoon. The benefits of this have already been proven in multiple studies and real world experiments. When platooning will be implemented on a large scale, it will be beneficial to optimise the platoons for maximum drag reduction. To be able to do this, the effect of different truck design parameters have to be investigated. This study focuses on the influence of underhood flow on the drag of a tractor semitrailer in isolation and in a platoon. This is done by using simplified models adapted to have a underhood model consisting of one porous medium and four ducts to replicate the mass flow, pressure drop and flow field of a real underhood area. Simulations were performed on fullsize and highway speeds using the commercially available PowerFLOW solver, based on the Lattice Boltzmann Method. For an isolated vehicle it was found that an increased underhood mass flow gives a higher total drag, mainly because of the drag contribution of the porous medium. Due to the mass flow entering the underhood, the suction over the leading edges is slightly reduced. On the other hand, smaller leading edge radii with higher suction give less mass flow through the underhood. Besides this the underhood flow actually has beneficial effects on the parts surrounding the tractortrailer gap and in the trailer underbody area. The highest total drag was found for the models with the most underhood flow and the smallest leading edge radius. Platoons of two vehicles were tested with three different inter vehicle distances, 3.75, 7.5 and 15 meters. The leading vehicle, which did not have underhood flow in all cases, has the strongest drag reduction for the shortest distance. The trailing vehicle has the lowest drag at the largest tested distance, while it is highest for the middle distance. This can be explained by the reduction in pressure in front of the vehicle. This reduces the drag contribution of the front surface, but also reduces the suction over the leading edges. The models with underhood flow experienced a stronger drag reduction, meaning that the absolute drag values were closer than for the isolated vehicle. This is caused by the reduced underhood mass flow in a platoon. At the shortest inter vehicle distance only 35% of the mass flow of an isolated vehicle is available, while this is 50 and 70% when the distance is increased. The beneficial effects of underhood flow on an isolated vehicle are still present in a platoon, although they are reduced in strength. When a boat tail is mounted on the back of the trailer of the leading vehicle the drag of this vehicle is strongly reduced due to the increased back pressure. However, this might not be beneficial for the trailing vehicle. The increased stagnation pressure indeed increases the total drag of the trailing vehicle at an inter vehicle distance of 3.75 m. As discussed before, an increased stagnation pressure also gives increased leading edge suction. Therefore the drag is actually reduced for the two larger inter vehicle distances. It was also found that the tail gives higher flow speeds over the top of the trailing vehicle, and lower flow speeds around the bottom. This reduces the drag for the underbody parts like the wheels and slightly decreases the underhood mass flow. When the platoon is placed at a yaw angle, the total drag of the leading vehicle is increased. The contribution of the front part is decreased, but this is more than compensated for by the drag increase for the rear and all other parts, which are no longer perfectly aligned with the flow. The drag increase of the trailing veh< icle is stronger, due to an increased contribution of the front part. The underhood mass flow is also increased compared to the platoons without yaw, this effect is strongest at small inter vehicle distances. The leading vehicle causes the flow to be more aligned for the trailing vehicle, therefore the drag increase for most other parts is less strong. The same effect can be seen for the side force, which is way lower for the trailing vehicle and increases for increasing inter vehicle distance.8Platooning; Trucks; bluff body; Underhood flow; CFD; LBM)uuid:d42b40e023c14e49abb696a05a3dcd06Dhttp://resolver.tudelft.nl/uuid:d42b40e023c14e49abb696a05a3dcd06A Study on the Sloshing Motions within a Rectangular Moonpool and the Effect of Implementing Appendages: A Computational Fluid Dynamics ApproachDingemans, N.T.eHuijsmans, R.H.M. (mentor); Akkerman, I. (mentor); Gerritsma, M.I. (mentor); Stofregen, M.M. (mentor)Moonpools are vertical wells used on most offshore vessels at Huisman Equipment. The oscillations in the moonpool lead to a significant increase in resistance for vessels in transit conditions. In the concept design phase it has proven difficult to predict the moonpool behaviour, as well as the effect of implementing appendages within a long rectangular moonpool. The ability to optimize the moonpool design is desirable, but difficult because there are no proper tools available. Solutions are now found by trial and error and experimentally validated by performing model tests. This research investigated the complex turbulent behaviour within long rectangular moonpools in transit conditions, using a commercial Computational Fluid Dynamics (CFD) package called STARCCM+ by CDAdapco. The capability and feasibility of applying this type of approach for the concept design phase of a moonpool is questioned..Moonpool; Sloshing; CFD; STARCCM+; URANS; DES
20210110Floating Offshore Structures)uuid:aa09e42ab148468cbd7e0ec7cbfcdd97Dhttp://resolver.tudelft.nl/uuid:aa09e42ab148468cbd7e0ec7cbfcdd97NDevelopment of a transient CFD methodology for the optimization of CNG engines/Maselis, Robbe (TU Delft Aerospace Engineering)UHickel, Stefan (mentor); Delft University of Technology (degree granting institution)Compressed natural gas (CNG) has been gaining attention in the automotive industry since it allows for a substantial reduction in carbon dioxide emissions. Recent advancements in directinjection technology have contributed to an increase in volumetric efficiency and engine power, but the physical modeling of such gaseous injection with Computational Fluid Dynamics (CFD) is still challenging. This work describes the development of a CFD methodology for the simulation of CNG engines, an evaluates its predictivity by varying certain simulation parameters. <br/>Firstly, turbulence modelling has been investigated by comparing the RNG k RANS, Dynamic Smagorinsky LES and Delayed DES approaches. The DES provides the best tradeoff between computational cost and accuracy during fuel injection, but the model makes abrupt transitions between its RANS and LESlike regions, leading to inaccuracies during combustion. The expensive nature of the LES and the adequate accuracy of the RANS model lead to the preference of the latter. Secondly, the turbulent Schmidt number was investigated. By lowering the number, mixing is more dominant in the simulation and the mixture formation becomes more stoichiometric. The influence on combustion is even larger because of turbulent transport across the flame, with extremely low values even leading to falselypredicted engine knock. Thirdly, by comparing multiple consecutive engine cycles to each other, it was found that the first cycle still contains a large amount of initialization error in the turbulent field. The second and third cycles show much better agreement to each other. Finally, the amount of turbulent fluctuations at peak power and their influence on laminar flame speed partially lead to flame extinction at the start of combustion. As a consequence, the Gequation model is not jus< tified and does indeed provide inaccurate results for the heat release during combustion. Even though the SAGE detailed chemistry solver in conjunction with the GriMech 3.0 mechanism is assumed to have a larger computational cost, it was found that the opposite is in fact true, and the latter should be preferred. <br/>The aforementioned conclusions are used to calibrate the simulation model against measurement data. The recommended settings are the RANS turbulence model, SAGE detailed chemistry and a turbulent Schmidt number of 0.6, while considering results from a second engine cycle. The Lower Heating Value (LHV) of the fuel was decreased in order to account for an unforeseen inconsistency with the measurement. This ad hoc solution showed that the predictivity of the RANS simulations can substantially be improved, and a good agreement between simulation and measurement was observed. The calibration is however not consistent with other operating conditions, as the part load analysis showed. Finally, the comparison between the peak power and part load operating point have revealed that the engine performance can potentially be improved by altering the piston bowl shape or the position of the injector and the spark plug.ACFD; CNG; Automotive; Engine; Methodology; Turbulence; Combustion
20240214)uuid:a53b78eed5c341dc85cf1b9da36109f8Dhttp://resolver.tudelft.nl/uuid:a53b78eed5c341dc85cf1b9da36109f8LfastFoam  An aeroservoelastic wind turbine simulation method based on CFD]Becker, Max (TU Delft Aerospace Engineering; TU Delft Aerodynamics, Wind Energy & Propulsion)Simao Ferreira, Carlos (mentor); Baptista, Carlos (mentor); Daniele, Elia (mentor); Dwight, Richard (graduation committee); van Zuijlen, Alexander (graduation committee); Delft University of Technology (degree granting institution)fThe current wind turbine design tools are based on aerodynamic simulations using blade element momentum (BEM) codes to calculate loads and power, see for instance the tools HAWC2, Bladed or FAST. This engineering method is computationally efficient, but theoretically valid only for steady two dimensional flow in nonyawed conditions. <br/><br/>To overcome these limitations engineering addons are used based on measurements or advanced methods. However, these include approximations which may result in introduced errors, especially in <br/>extreme operational conditions such as yaw. Moreover, future turbines may utilize flaps or slats and experience tip speeds higher than 110 m/s leading to Mach and Reynolds number effects for which the current tools are not validated yet.<br/><br/>Therefore, the objective of this Master thesis is the development of an aeroservoelastic simulation method based on an aerodynamic method with increased fidelity compared to BEM such as computational fluid dynamics (CFD). This was achieved by replacing the BEM module of NREL FAST by an OpenFOAM CFD code. Therefore, FAST and OpenFOAM were coupled by utilizing the neutral interface MpCCI. <br/><br/>Finally, it was investigated how such a method can be justified when compared to the stateoftheart tools, which use BEM. The implemented coupled method was thereby validated against experimental data from the NREL phase VI experiment. The Master thesis project, which was carried out externally at the CFD department of Fraunhofer IWES, showed that there is a need for more accurate methods especially in extreme conditions such as heavy yaw.,CFD; FSI; OpenFOAM; WInd turbine; Windenergy)uuid:2c964dfc8b3d44ba9e3dcd22704aa862Dhttp://resolver.tudelft.nl/uuid:2c964dfc8b3d44ba9e3dcd22704aa862UAerodynamics of a Rotating Wheel in a Wheelhouse: A Numerical Investigation using LESTViswanathan, Veeraraghavan (TU Delft Mechanical, Maritime and Materials Engineering)Elsinga, Gerrit (mentor); Pourquie, MAthieu (mentor); Westerweel, Jerry (graduation committee); van Zuijlen, Alexander (graduation committee); Delft University of Technology (degree granting institution)
The aerodynamic drag force is a crucial factor for passenger cars and trucks, where fuel economy is of< utmost importance. While a lot of research has been performed to optimize the design of the upperbody of the vehicle, the flow over a rotating wheel that is enclosed by a wheelhouse is not understood to the same level. Estimates show that the addition of wheel and wheelhouse to a vehicle body, can increase its drag and lift coefficient by as much as 40 \%. With the major automobile manufacturers striving to improve the fuel economy of the vehicle, understanding the flow over a rotating wheel in a wheelhouse, offers a new dimension for the improvement of the drag coefficient of the vehicle. In the present research, the flow over a simplified car body with a wheel and a wheelhouse was studied using Large Eddy Simulation (LES). When performing an LES on a given numerical grid, it can be expected that the contributions from the subgrid viscosity model reduces on decreasing the Reynolds number of the flow. A study focusing on the dependence of the flow field on the Reynolds number was thus performed. Three Reynolds numbers based on the wheel diameter  9 x $10^3$, 75 x $10^3$, 150 x $10^3$ were chosen for the dependence study. A comparison of the relative pressure distribution and the flow topology between the different Reynolds numbers showed that the flow over a rotating wheel in a wheelhouse was qualitatively independent of the Reynolds numbers considered in this work. Experimental results from Particle Image Velocimetry (PIV), performed on a similar vehicle model, were used to validate the numerical solution at a Reynolds number of 9 x $10^3$. The behaviour of the flow in the wheelhouse region was investigated and a new description of the flow field inside a wheelhouse, consisting of 11 vortex structures, was presented in this work. To determine the qualitative dependence between the flow field and the drag coefficient, a comparison of the flow between two geometric configurations, with different wheelhouse radius, was performed at a Reynolds number of 9 x $10^3$. It was found that the configuration with the larger wheelhouse radius had a higher wheelhouse drag coefficient and a lower wheel drag coefficient when compared to the configuration with lower wheelhouse radius. Based on a detailed flow field, four out of the eleven vortices were found to be crucial in determining differences in the drag coefficient of the wheel and wheelhouse. Based on this analysis, an alternate wheelhouse geometry, combining the advantages of the two geometric configurations considered in this work, was proposed, that could potentially lead to a reduction in the drag coefficient of the vehicle.5CFD; LES; Aerodynamics; Wheel; Wheelhouse; Automobile)uuid:09dc7b81d2684e93be17b6f0c683c361Dhttp://resolver.tudelft.nl/uuid:09dc7b81d2684e93be17b6f0c683c361\Aerodynamic Investigation of an Exit Guide Vane Followed by a Curved Duct: A Numerical StudyYPfeifle, Ole (TU Delft Aerospace Engineering; TU Delft Flight Performance and Propulsion)Gangoli Rao, Arvind (mentor); Ramm, Guenter (mentor); Pini, Matteo (graduation committee); Schrijer, Ferdinand (graduation committee); Vitale, Salvo (graduation committee); Delft University of Technology (degree granting institution)
Curved exhaust ducts are used in aero engine applications for different purposes, including thrust vectoring, shielding of parts from the exhaust or improving stealth properties. Their integration, however, regularly represents a design problem due to flow separation and high aerodynamic losses occurring in the bend. Curved duct flows for both compressible and incompressible conditions have been studied extensively in the past. However, no experimental results of a high Reynolds number flow through a turbine exit guide vane (EGV) followed by a curved duct have yet been published. An indepth CFD analysis of the aerodynamic effects is therefore carried out, using the RANS solvers TRACE (at MTU Aero Engines) and SU2, to analyze the flow field and geometric sensitivities of a high Reynolds number flow through an EGV followed by a 90 degree bent duct. The geometry of interest is investigated at < a Reynolds number of Re=10^6 and a ratio of bend radius to duct diameter of one. The inflow conditions are prescribed to closely resemble typical flow conditions at the low pressure turbine exit plane of a turbojet engine. After validating the solver with experimental data using the test case of a 90 degree bent duct, an initial CFD analysis of the combined exit guide vane geometry with curved duct is carried out to identify the dominant flow phenomena and mutual effects of the EGV and the bend. Three zones of flow separations are found, each at the convex and concave sides of the bend and one at the lower side of the plug. Secondary flows caused by the bend are found to have an effect on the flow upstream of the EGV, leading to a nonuniform flow inlet angle and aerodynamic loading of the individual blades. Separation downstream of the EGV is influenced by the presence of low velocity wakes from the EGV. Subsequently, a sensitivity study is carried out to find the effect of different geometrical parameters on the flow field. Main investigated parameters are the ratio of bend radius to diameter (R_c/D) and the distance between EGV and bend (l/D). Additionally, the aerodynamic effects of the circumferential EGV positioning, swirl and the plug shape are investigated. It is found that an increase in both bend radius and distance between bend and EGV improves aerodynamic efficiency, while swirl can decrease pressure losses in the duct for small bend radii of R_c/D<0.8. Improving the plug shape and rotating the EGV allows to further increase the aerodynamic efficiency without weight increase. For further optimization of the geometry, it is recommended to include duct geometries with a nonconstant bend radius and outer duct diameter to increase the design space.XCFD; turbulent flow; 90 elbow bend; flow separation; sensitivity study; Exit Guide Vane)uuid:85b601cc84544686afcb2344ec752ac1Dhttp://resolver.tudelft.nl/uuid:85b601cc84544686afcb2344ec752ac1+Hydrodynamic behavior of inline structuresRodermans, M.J.Metrikine, A. (mentor) Heerema Marine Contractors (HMC) is, amongst other activities, involved in the installation of subsea pipelines and subsea structures. In order to increase efficiency during production and installation these subsea structures are welded into the pipeline instead of installed separately. The presence of subsea structures in the pipeline increases the stresses in pipelines during installation. The stresses in pipelines during installation are analyzed beforehand. These analyses show that with subsea structures becoming bigger, the workability of inline subsea structure installation becomes unacceptably low. In order to model the hydrodynamic forces acting on subsea structures simplified models are used. The suspicion is that these simplified models are conservative because of the lack of knowledge on the hydrodynamics around complex shaped structures. A first research on the behavior of inline structure installation and the effects of alternative hydrodynamic loading models has been performed at HMC. On the basis of this research, questions remain. The most important ones are: " Is Morison s equation applicable for sharp edged and asymmetric structures? " Can forced motion experiments be used to model the dynamic behavior of subsea structures? " Is the motion of sharp edged and asymmetric structures decoupled? This thesis is performed to investigate the hydrodynamics around subsea structures and answer these questions.. With the use of Dynamic Fluid Body Interaction (DFBI) simulations the dynamic behavior of sharp edged and asymmetric structures and the description of this behavior have been investigated. Subsea structures are simplified as thin flat plates and simulations in one and two degrees of freedom have been performed. Over a range of frequencies the behavior in regular oscillating flow has been investigated. From the research it can be concluded that over the range of KeuleganCarpenter (KC) numbers associated with inline subsea structure installation Morison s equation can be used to desc< ribe forces accurately and these forces are decoupled. Further research should be performed on the hydrodynamic coefficients for subsea structures as the DNV GL prescribed coefficients cause large differences with the simulated behavior where KC dependent coefficients perform better. Hydrodynamic moments are observed which are not taken into account during installation analysis. The impact of these moments on the dynamic behavior of subsea structures should be further investigated.?Morison's Equation; CFD; DFBI; Hydrodynamics; Subsea structures)uuid:56d1ccc2b50a44a69908d5f495c6951cDhttp://resolver.tudelft.nl/uuid:56d1ccc2b50a44a69908d5f495c6951cZTsunami induced failure of bridges: Determining failure modes with the use of SPHmodeling9Salet, Jesse (TU Delft Civil Engineering and Geosciences)Bricker, Jeremy (mentor); Antonini, Alessandro (graduation committee); Yang, Yuguang (graduation committee); Suzuki, Tomohiro (graduation committee); Kostense, N.W. (graduation committee); Delft University of Technology (degree granting institution)Recent major tsunami events generated by earthquakes inundated coastal cities and caused extreme destruction and loss of human lives. The collapse of coastal bridges due to tsunami wave impact represents a huge obstacle for rescue works. The need to understand tsunami effects and develop tsunamiresilient bridges became apparent in the aftermath of extreme tsunami events in the Indian Ocean (2004), Chile (2010) and Japan (2011). <br/><br/>Different coastal topographies affect tsunami propagation near shore. Varying wave characteristics lead to various failure mechanisms of bridge decks. Together with the wave characteristics, the bridge properties and the settings around the bridge play a major role in this failure, think for example of shear keys, seawalls or inclination of the bridge.<br/><br/>To find out more about these failure mechanism and what role all these measures have in the failure, a laboratory experiment is executed and a numerical SPH model is set up to investigate the impacts of various wave characteristics, a seawall, shear key and inclination of the bridge deck. The numerical SPH model is validated with the help of wave gauge data and tracked bridge deck movement from the executed physical tests. <br/><br/>In this thesis the focus is on the movement of the bridge deck, what kind of effect do the different interventions have on the movement of the deck. Since the movement is highly dependent on the forcing on the bridge deck, the forces are analyzed thoroughly. From the force time series countermeasures are proposed and modeled in the SPH model. <br/><br/>Wave forces from different type of waves are simulated with the SPH model. The overall behavior of the hydrodynamics and the deck movement are validated and suited for qualitative analysis.<br/>Some disadvantages of the model are the lack of bottom friction and air bubbles in turbulent regions.<br/>The 3D model represented the movement on the deck in a very good way, runtimes and storage capacity formed an obstacle. A 2D model was used to do qualitative analysis of the changes of wave characteristics and the effects of the structural measures.<br/>The limiting factor in the commercial use of SPH is the computation time. In future models this could be accelerated by the use of GPU processors instead of CPU processors which are able to solve many parallel processes at the same time.<br/><br/>Apart from wave heights and inundation heights, the wave phase appeared to be a major decisive factor in the failure method of the bridge deck. If the wave breaks near the shore and reaches the bridge structure as a propagating wave front, the hydrodynamic situation results in high horizontal forces and a sliding failure mode is apparent. When a wave is still in a surging phase and the fluid particles still have their rotational movement, the dominant forcing on the bridge is in vertical direction. Since the vertical force applied to the bridge deck moves from seaside to shore side, the sea side of the bridge deck has a higher vertical vel< ocity which initiates rotation. <br/>A seawall causes the water to confine underneath the bridge deck. Which will result in higher vertical forces, thus a rotational failure mode follows. Inclination of the bridge deck has significant effect on the vertical forcing. Positive inclination lead to a decrease of upward forcing and negative inclination lead to an increase of upward forces. The introduction of shear keys resulted in higher moments, since the point of rotation is set at the point the deck interacts with the shear key, which creates a larger distance around the point of rotation.<br/><br/>Possible countermeasures that are introduced are a sacrificial beam and a different geometry of the deck. A sacrificial beam was effective in lowering the total horizontal forces on the combined structures. The deck itself was not exposed to a high horizontal impact force. Different geometries are tested to see how the forces on the structure would chance. A wing shaped geometry has positive effects in mitigating the horizontal forces on the bridge deck.qBridge; SPH; Smooth particle hydrodynamics; LSDYNA; Laboratory experiment; CFD; Tsunami; wave; Flume experiments)uuid:4e54cfb33024420a820548c7eb767ce9Dhttp://resolver.tudelft.nl/uuid:4e54cfb33024420a820548c7eb767ce9XParametric Modeling and Optimization of Advanced Propellers for NextGeneration Aircraft^Klein, Peter (TU Delft Aerospace Engineering; TU Delft Aerodynamics, Wind Energy & Propulsion)qSinnige, Tomas (mentor); Vitale, Salvatore (mentor); Delft University of Technology (degree granting institution)As traditional fossil fuels become scarcer and more attention is given to environmental impact of the combustion of fossil fuels, for nextgeneration aircraft, the focus of development will mainly be on reducing fuel consumption. Open rotor engines have the advantage over conventional turbofans that they are expected to perform 25% to 30% better in terms of fuel consumption. The topic of this thesis focuses on one of the current challenges related to open rotor configurations; the unwanted unsteady effects imposed on the aerodynamic characteristics of a propeller, imposed by an upstream pylon. There is a potential for reducing undesired propellerpylon installation effects by taking these effects into account in aerodynamic design process of the propeller. A propeller design optimization routine is implemented that includes a parametric modeling tool, a low and highfidelity performance analysis method and an existing installation effects performance model.gPropeller; CFD; Installed pusher configuration; Optimization; Parametric design tool; Noise; Pylon wake)uuid:57f7c2b3918f433da804d82e04c639ddDhttp://resolver.tudelft.nl/uuid:57f7c2b3918f433da804d82e04c639ddEvaluation of the implementation of a Flettner rotor on a Exploration cruise vessel with respect to the roll motion and rotorsuperstructure interactionKSteijger, Miranti (TU Delft Mechanical, Maritime and Materials Engineering)van 't Veer, Riaan (mentor); van Terwisga, Thomas (mentor); Eggers, R. (mentor); Groefsema, A. (mentor); Delft University of Technology (degree granting institution)+Flettner rotor; CFD; URANS; Kutta Joukowski
20221121)uuid:86621aa390a64929953497459ef11f87Dhttp://resolver.tudelft.nl/uuid:86621aa390a64929953497459ef11f870LES of a novel wing/body junction : AntifairingDRaghunathan Srikumar, Sampath Kumar (TU Delft Aerospace Engineering)Dwight, Richard (mentor); Belligoli, Zeno (mentor); Eitelberg, Georg (graduation committee); Hickel, Stefan (graduation committee); Delft University of Technology (degree granting institution)[ Large Eddy Simulations (LES) of a novel type of wing/body junction called the antifairing are performed in the current thesis to study the complex turbulent flow physics involved in the junction area and also to obtain a clear understanding of the drag reduction capabilities of the antifairing. In regards to that, two separate LES are performed: one for the baseline case with a Rood wing/flat plate combination and another with the Rood wing/antifa< iring combination. A detailed comparative study is performed between the two cases to observe important differences in junction flow characteristics. Both the simulations are performed on a 25 million immersed boundary Cartesian mesh by solving the incompressible NavierStokes equations using the inhouse finite volume LES solver called INCA. Results from the LES study confirms the existence of the propulsive pressure mechanism of drag reduction for the antifairing case, previously proposed by Belligoli et al. However, the results also show that there exists a secondary drag reduction mechanism caused by a combination of increase in approach boundary layer momentum thickness and dampening of the turbulence associated with the horseshoe vortex (HSV) upstream of the wing. This secondary mechanism has been found to be caused by the convex dent present at the start of the antifairing geometry. The total drag reduction for the antifairing case comes out to be 1.8%. A new parameter called junction drag is defined which accounts for the drag only due to the presence of a junction. The reduction in junction drag obtained for the antifairing case is about 6.8%. Apart from the LES analysis, a RANS analysis has also been performed to further investigate the drag reduction capabilities of antifairing for different approach boundary layer thicknesses and antifairing depths. All the RANS analysis have been performed on a 5 million bodyfitted mesh by solving the incompressible NavierStokes using the open source finitevolume solver OpenFOAM. Results from the RANS analysis indicate that there exists an optimum depth for the antifairing which corresponds to the least drag. Furthermore, it is found that the effect of approach boundary layer thickness is mostly on changing the base drag of the case where no antifairing is present, rather than actually affecting the performance of the antifairing at different depths.,Antifairing; LES; INCA; Drag Reduction; CFD)uuid:89e3697b6d234cf28f31cfc2448b286aDhttp://resolver.tudelft.nl/uuid:89e3697b6d234cf28f31cfc2448b286aVMoonpool in Waves: CFD Verification and Validation of Wave Elevation Inside a MoonpoolLMarelli, Giancarlo (TU Delft Mechanical, Maritime and Materials Engineering)van 't Veer, Riaan (mentor); Schreier, Sebastian (mentor); Jaoun, Frdrick (mentor); Delft University of Technology (degree granting institution)Large water motion inside the moonpool of vessels operating in waves can be excited by pressure fluctuations produced by external waves and vessel motion. Extreme consequences of this effect may include injuries for crew members, and damages to deck equipment resulting in downtime for the vessel. The accepted method to predict this nonlinear phenomenon is a combination of model tests and potential solver. Nevertheless, model tests are generally conducted at the end of the design phase leading to serious problems if the moonpool performance is not sufficient. CFD solvers proved their capability of modelling complex flow phenomena and their use as a design tool for the moonpool is growing. However, a complete verification & validation is still missing. In the present work the water motion inside a moonpool and the forces on the hull, for a vessel in head waves without forward speed, are estimated using the MARIN software ReFRESCO. In doing so, the goal is to define the accuracy of the code for a rectangular moonpool with sharp edges without additional damping devices. For a deeper comprehension of the physics involved, a stepwise approach was followed. The work starts with an empty domain in which only regular waves were generated to assess the propagation and absorption of waves in ReFRESCO. Secondly, fixed vessel simulations were performed to study the influence of grid dimension, mesh refinement and boundary conditions. Then forced heave oscillations were simulated to estimate the damping and added mass. Verification studies were carried out for all the presented to this point. Finally, results from freefloating simulations were validated against experimental res< ults.*Moonpool; Validation and Verification; CFD)uuid:4ef56c0f1dbd4edb8b89f672dd3597ecDhttp://resolver.tudelft.nl/uuid:4ef56c0f1dbd4edb8b89f672dd3597ec`Design and Analysis of Swirl Recovery Vanes for an Isolated and a Wing Mounted Tractor PropellerStokkermans, T.C.A.cIn light of the energy crisis of the early 1970's, NASA and industry gained a renewed interest in highspeed propellers for improved propulsive efficiency and explored the idea of swirl recovery vanes (SRV) to generate a net thrust from the residual swirl in the propeller slipstream. After this first effort on the aerial application of SRV, only recently research is resumed. When a wing is introduced in the slipstream of a propeller, for instance for a wingmounted tractorpropeller, conclusions drawn on SRV in isolated condition may not hold. The objective of this research is to gain an improved understanding of the aerodynamic interaction between the propeller and swirl recovery vanes in an isolated configuration and wingmounted tractor arrangement in the cruise condition and in a highthrust condition. This study is realized by performing a series of transient Reynoldsaveraged NavierStokes CFD simulations of a propeller with and without SRV in an isolated and installed configuration. Throughout this research the 6bladed propeller of the European APIAN project is used. Available experimental propeller performance, blade pressure and slipstream measurements are used to validate the isolated propeller CFD model. Within the limitations of fully turbulent modelling of the boundary layer by means of automatic wall functions, good agreement is found with the experimental data, including the existence of a conical separation vortex at low advance ratios. Simulated performance and slipstream results are presented of the APIAN propeller with SRV designed for the APIANINF test program in the DNWLLF. PIV measurements in a plane spanned by the radial and rotation axis provide a comparison of the slipstream velocity components and vorticity. This simulation combined with the PIV measurements enables an extensive description of the structure of root and tip vortices induced by the propeller blades and swirl recovery vanes. It is found that the propulsive efficiency increase by the addition of SRV is only 0.57% which is much lower than the design prediction of 1.8%. Therefore this design is not used in the remainder of the research and new SRV designs are proposed. An SRV analysis tool based on liftingline theory modified for nonuniform inflow is presented. In combination with an optimisation routine, this tool allows for the design of SRV for an isolated propeller. From a simplified analysis of an elliptical vane in a uniform swirl flow, it is concluded that optimisation for maximum SRV thrust is preferred over complete swirl recovery to reach the highest gain in propulsive efficiency. Four designs are presented: Design 1 is optimised for the cruise condition with a constraint on stall for the highthrust condition. Design 2 is optimised for the highthrust condition with a constraint on the cruise condition for zero or positive efficiency benefit. These are designs where the SRV have a fixed pitch in flight. Also two variable pitch designs are proposed. The effect of cropping and the number of vanes on the propulsive efficiency is investigated as well for the objective of design 1. Design 1 and 2 are used in CFD simulations behind the isolated propeller to validate the predictions from the SRV analysis tool. In general the simulation results show that SRV lead to an increase in propulsive efficiency by increasing the system thrust over a wide range of advance ratios, with minor effect on the system power. Gains in propulsive efficiency of 0.39% and 0.20% are found in the cruise condition and 2.62% and 3.07% in the highthrust condition for design 1 and 2 respectively. For high advance ratios the prediction is very accurate, while towards lower advance ratios the tool overpredicts the propulsive efficiency gain. The difference is within the limits that can be expla< ined by the set assumptions. Design 1 proves that it is possible to increase the propulsive efficiency of an operating point close to the point of maximum propeller propulsive efficiency. Design 2 shows that if a larger increase in propulsive efficiency at low advance ratios is desired, the design can be changed at the cost of propulsive efficiency benefit at higher advance ratios, for a fixed SRV pitch design. Downstream of the SRV, somewhat less than half of the swirl is recovered on average. An expansion of the slipstream boundary is present, which is the result of the interaction of propeller blade and vane tip vortices. In the last part the wing of a Fokker 50 is introduced behind the propeller and SRV design 1. The loading on the wing induces an upwash upstream of the wing, resulting in a deviation from the SRV design inflow that is different for each vane by such a degree that flow separation degrades the SRV performance to a large extent. Therefore a change in the SRV design is made by turning each vane over an angle to obtain the time and radialaverage design inflow in the cruise condition. For future research it is recommended to find a different design for each vane. Since the effect of the wing upwash on the SRV inflow field varies with advance ratio and with wing loading and thus varies in flight, a variable pitch SRV design is recommended where the pitch of each vane is adjusted individually. For the cruise condition the increase in propulsive efficiency by the addition of SRV without considering differences in wing drag is found to be 0.93%, which is considerably higher than without wing, mainly due to the increased propeller propulsive efficiency, but partly by increased SRV thrust as well. 2.14% for a mediumthrust condition, which is very similar to the value without wing. For a wingmounted tractorpropeller conclusions on SRV performance can only be drawn from the complete force balance of thrust and lift of the propeller, SRV, wing and nacelle. Considering the drag of all components, the net increase in propulsive efficiency by the addition of SRV is found to be 0.14% for the cruise and 1.00% for the mediumthrust condition with a net increase in lift of 0.35% and net decrease in lift of 0.55% respectively. Careful optimisation of SRV taking the wing into account as well as the lift as a constraint will most likely result in a performance benefit, since already with this nonoptimised design an increase in thrust or lift can be found depending on the advance ratio. The propeller slipstream greatly affects the wing lift and drag distribution by its increased axial velocity and introduced swirl. It is concluded that SRV reduce some of the effects of the propeller on the wing lift and drag distribution by a reduction of the swirl, resulting in a smaller deviation from the wing loading without propeller. A design procedure for SRV should include the wing for instance by an additional lifting line and optimise for combined SRV and wing maximum thrust with a constraint on the net lift. This may lead to SRV designs more focussed on providing the optimal inflow for the wing in order to reduce the wing drag.dSwirl Recovery Vanes; SRV; propeller; APIAN; wing; Fokker 50; CFD; propulsive efficiency; slipstream)uuid:2d40af8cfcdf41a98aad05b16ad4f9baDhttp://resolver.tudelft.nl/uuid:2d40af8cfcdf41a98aad05b16ad4f9ba;Ambergy Industrial: A heating control system to save energy
Muris, V.L.C.`Nijsse, R. (mentor); Schipper, H.R. (mentor); Van Drimmelen, R. (mentor); Bokel, R.M.J. (mentor)
Industrial buildings, used as warehouses or distribution centres, are characterized by large doors which are opened temporarily or for longer periods. A new technique  called Ambergy  is investigated which prevents unnecessary energy loss through open overhead doors in heated industrial halls. This technique consists of a smart coupling between the heating system and the overhead doors. By temporarily switching off gasfired heaters near open doors, the heat loss through the door should be minimized. It is expected that the Ambergy syst< em can contribute to energy savings, also when the indoor thermal comfort is taken into account. However, at the start of the research, the amount of energy which can be saved and the effect of the Ambergy system on the thermal comfort, was not known yet. If energy can be saved by using the Ambergy system, and the current thermal comfort level can be retained, many industrial buildings can benefit from this. This thesis aimed to get an insight in the energy saving effects of the Ambergy system and to determine its potential  technical  feasibility. To fulfil this aim, a literature study is performed to predict important physical aspects affecting the heat balance, to gain insight in air transport phenomena and to derive criteria to compare thermal comfort levels. By using the software packages Matlab and Simulink, these physical aspects and air flow phenomena are implemented in a thermal buildingdynamics simulation. This simulation predicts effect of the Ambergy system on the air temperature across the hall and the fuel savings for different circumstances during a whole winter season. Assumptions made in the thermal buildingdynamics simulation  regarding the air flow direction  are verified with a computational fluid dynamics (CFD) model. Furthermore, measurements are performed in one representative industrial hall (Alphatron, Rotterdam), to gain input data for both models and to validate the outcome. To make the system as optimal as possible and accepted by the employees, also the effect on the indoor thermal comfort is taken into account. A comparison between the current thermal comfort level and the expected thermal comfort level, when applying Ambergy, is performed by calculating the required insulation value of the clothing of the employees (IREQvalue). As part of this thesis, also requirements for pilot projects at business facilities of DHL and Alphatron  in s Hertogenbosch and Rotterdam respectively  are defined and these pilot projects were carried out during this thesis. Due to the confidential nature of this research and the embargo set by the TU Delft and BreedofBuilds B.V., no information can publicly be given regarding the results, conclusions and recommendations done in this research until August 2017.ambergy; heating system; energy; industrial; hall; control; heat loss; building physics; sustainable; sustainability; finite element modeling; heat; CFD; civil engineering; building engineering; gasfiredheaters; thermal comfort
20170801Structural EngineeringBuilding Technology and Physics)uuid:0bef658bca1c4edbbaf998263b88fd7bDhttp://resolver.tudelft.nl/uuid:0bef658bca1c4edbbaf998263b88fd7b,Interface Investigation of Coreannular Flow`Jia, Kangjun (TU Delft Mechanical, Maritime and Materials Engineering; TU Delft Fluid Mechanics)SHenkes, Ruud (mentor); Delft University of Technology (degree granting institution)JThe transport of highlyviscous oil in the coreannular flow regime in a horizontal pipe is investigated with a numerical simulation study. In this flow type the viscous oil is lubricated by a water annulus along the pipe wall. The LaunderSharma lowReynolds number k turbulence model is used with the VolumeofFluid solver in interFOAM. Three types of simulations were carried out: 3D multiphase flow in a pipe section, 2D multiphase flow in an axisymmetric pipe section (wedgeshaped section), in which gravity is ignored, and 2D single phase water flow in the annulus, using an imposed wavy boundary. This enabled to study the effect of the viscous oil core on the annulus behaviour, such as the turbulence structures and possible dispersion of oil into the water annulus. In particular the rationale of a 'solidcore' assumption is discussed through the comparison between simulations for the annulus that use the 'interfacebounded' flow and 'wallbounded' flow. <br/><br/>A main finding of this study is that the use of the Compressive Volume of Fluid (CVOF) method in the multiphase simulations gives spurious dispersion of oil in the water annulus. This is the primary cause of the overpredic< tion of the pressure gradient simulations carried out in a previous study. Therefore, the Coupled LevelSet Volume of Fluid (CLSVOF) interface capturing method has been implemented in the 3D multiphase model. The new predictions show a good agreement with the experimental data at 20, 30, 40! (corresponding to a descrease in the oil viscosity). The oil dispersion in the annulus region close to the wall is diminished by using the new method while the sharpness of the interface is not improved. The CLSVOF interface capturing method is concluded to be an efficient and reliable interface capturing method for the numerical prediction of coreannular flow.<brUInterface Capturing; Coupled LevelSet Volume of Fluid Method; Coreannular flow; CFDEnergy & Process Technology)uuid:fc21b131b75844e3842db854a935f5c6Dhttp://resolver.tudelft.nl/uuid:fc21b131b75844e3842db854a935f5c6yHybrid EulerianLagrangian Vortex Particle Method: A fast and accurate numerical method for 2D VerticalAxis Wind TurbineManickathan, L.1Palha, A. (mentor); Simao Ferreira, C.J. (mentor)The wake geometry of a Vertical Axis Wind Turbine (VAWT) is unlike the standard Horizontal Axis Wind Turbine (HAWT). The blades of the turbine continuously passes through its own wake, creating complex wakebody interactions such as flow separation and dynamic stall, and convectional gridbased numerical method which is capable of describing such nearbody phenomena fails at efficiently resolving the wake geometry. However, as these phenomena have a direct impact on the performance of the VAWT, it is paramount that there exists a numerical method that is not only capable of accurately resolving the smallscale nearbody phenomena but also excels at efficiently resolving the unsteady largescale wake geometry. This was the goal of the research and the numerical method that satisfied these requirements was the domain decomposition method known as the Hybrid EulerianLagrangian Vortex Particle Method (HELVPM), based on the doctoral thesis of Daeninck [24] and the additional study performed by Stock [61]. In the present study, we coupled an Eulerian Finite Element Method, which only resolves the nearbody domain, with a Lagrangian Vortex Particle Method, which resolves the entire wake. The advantage of such fluid domain segregation was that the Eulerian method could focus on accurately describing the nearbody features whereas the Lagrangian method could focus on efficiently evolving the wake using simulation acceleration methods such as Fast Multipole Method (FMM) and parallel computation in Graphical Processing Units (GPU). The present study initially developed, verified and validated the Finite Element method and the Vortex Particle method separately ensuring it performs according to the theory. These methods were then coupled using the algorithm of Daeninck [24] and Lagrangian correction strategy developed by Stock [61]. However, during the study we determined that additional modifications to the coupling strategy is required to ensure conservation of circulation. Furthermore, it was determined that the spatial resolution of numerical method at the overlap region, where coupling takes place, plays a crucial role in the accuracy of the coupling. Even though the hybrid method of the present study sacrificed some efficiency to ensure an accurately coupled scheme, we must not that it is still at its infancy. With the help ofadvanced techniques such as varying particle core size, higher order time marching scheme in the Eulerian method, and boundary element method acceleration techniques such as FMM and/or GPU calculation, the hybrid method has the potential to substantially outperform standard grid based methods. In conclusion, the hybrid method that has been developed here has the potential to accurately describing the nearwake phenomena and efficiently evolving the wake of a VAWT.RVortex Method; VAWT; Hybrid; Finite Element; Python; Aerodynamics; CFD; Windenergy
20141204Aerodynamics & Wind Energy)uuid:0314e592ffa24f1dbc315da7011805b7Dhttp://resolver.tudelft.nl/uuid:0314e< 592ffa24f1dbc315da7011805b7KImplementation of a FluidStructure Interaction Solver for a Spinnaker Sail/Ramolini, Anna (TU Delft Aerospace Engineering)van Zuijlen, Alexander (mentor); van Oudheusden, Bas (graduation committee); Schmehl, Roland (graduation committee); Folkersma, Mikko (graduation committee); Delft University of Technology (degree granting institution)7The design of sails has always been done experimentally, and only recently simulations are starting to be used in the design process. This thesis is a first attempt in creating a solver that couples CFD and FEM in order to compute the deformed sail shape (flying shape) and the thrust it can provide. Such solvers already exist but are not available to the public, or if they are they come with a very high license price. The complexity of the problem is both in the flow, which is fully turbulent and detached, and in the structure, which is deformable and free to move in all directions. Moreover, the coupling of the solvers has to be performed in a way that minimizes loss of information and accuracy.<br/>First the CFD simulations have been run and validated with two softwares, OpenFOAM and FINE/Open. The results were very satisfying for FINE/Open, while quite poor for OpenFOAM. Consequently, the FEM solver has been successfully validated for some cases of which the analytical solution is known, due to lack of reference data for this specific case. <br/>Finally, the interpolation techniques have been implemented in Matlab and the fluid structure interaction solver has been run. The solver has been validated on a given testcase with satisfying results; however there is still room for improvement in terms of run times and automatization of the solver. From the results it can be argued that the design and flying shape of the sail are quite different and provide different thrusts. That is an indication of the significance of this type of analysis in the sail design process.9Fluid Structure Interaction; CFD; FEM; Sailing; Spinnaker)uuid:2d9caa9462e348a8ab52b74d8c38816eDhttp://resolver.tudelft.nl/uuid:2d9caa9462e348a8ab52b74d8c38816eKDesign and development of a small scale jet head for multilateral drillingShreedharan, V.A.,Pecnik, R. (mentor); Reinicke, A.B. (mentor)5Extraction of hydrocarbons from the subsurface is becoming harder with the reservoirs getting smaller and the formations increasingly tighter. Considering the current circumstances, production enhancement otherwise known as stimulation techniques, play a vital role in ensuring economic returns from a hydrocarbon reservoir. The use of water jetting for enhancing the connectivity of oil & gas wells to reservoirs is gaining attention these days over hydraulic fracturing in the oil & gas industry due to the enormous costs and rising concerns over environmental impacts associated with fracturing. Water jetting systems operate at high pressures and flow rates in order to effectively drill through tight sandstone formations. Computational fluid dynamics provides the right platform to help design and improve a jet drilling tool before progressing to field trials. Such an optimization process requires insight into the intergranular interaction between fluid & rock surface at downhole conditions and the mechanism of rock erosion due to water jetting. Initially, the rock surface is geometrically modeled in order to provide the fluid flow domain to analyse the fluidrock interaction. Subsequently, empirical rough wall function is utilized to emulate impinging jet flows over rock surfaces. Finally, the most viable wall treatment is employed for simulations of the smallscale rotating jet head in a borehole environment. The developed jet head has the potential to lower water usage & power consumption, improve water recyclability and reduce logistics cost when compared to hydraulic fracturing. Additionally, controlled placement of the needle wells allows operations in reservoirs which are in close proximity to underground aquifers, reservoirs with naturally occurring faults and fractures, reservoirs having co< mpatibility issues with fluids and additives used in hydraulic fracturing. The needle stimulation technology is expected to perform comparable to hydraulic fracturing in these challenging reservoirs by offering advantages in health, safety and environment (HSE) aspects.CFD; jet drilling; impinging jet; rough walls
20180831)uuid:90c8d527721447ac8a0b0f03c40625ccDhttp://resolver.tudelft.nl/uuid:90c8d527721447ac8a0b0f03c40625cc>Regenerative cooling analysis of oxygen/methane rocket engines
Denies, L.Zandbergen, B.T.C. (mentor)Methane is a promising propellant for future liquid rocket engines. In the cooling channels of a regeneratively cooled engine, it would be close to the critical point. This results in drastic changes in the fluid properties, which makes cooling analysis a challenge. This thesis describes a twopronged approach to tackle this problem. Simple and fast engineering tools allow for the development of insight in the design space using rapid iterations and parametric analyses. However, they are often rather inaccurate. In contrast, detailed multidimensional tools for numerical analysis are more accurate, but they require more computation time. Both approaches are developed for the analysis of regenerative cooling channels of oxygen/methane engines. Each approach uses complex but accurate models for the thermodynamic and transport properties of methane. OMECA (short for Onedimensional Methane Engine Cooling Analysis) is a onedimensional tool that was developed in Python from scratch. This tool divides a nozzle into stations and analyses the onedimensional thermal equilibrium at each station. It makes extensive use of semiempirical equations to calculate the heat transfer at both the hot gas side and the coolant side. The tool is compared to a coupled multiphysics analysis tool, showing that the accuracy of the wall temperature is rather poor, with discrepancies of up to 150~K. Both at the hot gas and coolant side, large deviations are present. However, if the input heat flux is correct, OMECA predicts the coolant pressure drop and temperature rise with a 10% accuracy. To obtain a higher accuracy at the coolant side, the opensource CFD package OpenFOAM is adapted for analysis of supercritical methane. Of particular note is the custom library that interpolates the fluid property tables at runtime. The selected solver is applicable to steadystate compressible flows. The software is then systematically validated using three validation cases. With experimental validation data obtained through cooperation with CIRA, an accuracy of 15 K for the wall temperature prediction is demonstrated. The pressure drop is predicted within 10%. Traditionally, the launcher industry uses copper alloys as wall material in regeneratively cooled combustion chambers. They offer a high allowable temperature and high thermal conductivity, but are also heavy and expensive. Recently, several companies have demonstrated aluminium combustion chambers. Aluminium alloys have weight and cost advantages, but have lower allowable temperature and thermal conductivity. The developed tools for cooling analysis are therefore employed to compare aluminium and copper for a generic 10~kN combustion chamber. It is discovered that a thermal barrier coating must be employed to protect the hot gas side of an aluminium combustion chamber, otherwise regenerative cooling is not feasible. Even with such a coating, the pressure drop required to cool the coated aluminium chamber is three times higher than the pressure drop required for a copper chamber. A difference in pressure drop has effects on the vehicle level. A larger pressure drop in the cooling channel of a rocket engine necessitates a higher feed pressure. For a pressure fed engine, this means the tank must be stronger and heavier. It is found that even at modest fuel mass, the increase in tank mass is eight times as large as the decrease in engine mass offered by aluminium. This shows that using aluminium for the chamber wall is not advantageous with respect to copper for a pressure fed, regene< ratively cooled, oxygen/methane rocket engine.Nrocket engine; regenerative cooling; methane; CFD; OpenFOAM; aluminium; copperSpace EngineeringSpace Systems Engineering)uuid:ccb561540b704a41822324b0f8d145c5Dhttp://resolver.tudelft.nl/uuid:ccb561540b704a41822324b0f8d145c5]An Investigation of the NonLinear 3D Flow Effects Relevant for Leading Edge Inflatable KitesDeaves, M.E.BSchmehl, R. (mentor); Gaunaa, M. (mentor); Gillebaart, T. (mentor)9 The kite power group at TU Delft is currently researching the use of leading edge inflatable (LEI) kites for use in power generation. A thorough understanding of the aeroelastics of these kites is paramount to the development of system simulation models and optimum kite and system designs. The current lack of understanding is therefore seen as a roadblock to the development of a commercially viable kite power system. The aeroelastics of LEI kites are complicated by three main challenges. There is a high degree of coupling between the flexible kite and the aerodynamic loading. This means that a fluidstructure interaction approach is typically needed to produce accurate simulation results. The low aspect ratio and large anhedral of the kite means that 3D effects are significant. During normal power production it is desirable to fly the kite at high angles of attack where significant nonlinear viscous phenomena (e.g. flow seperation) are known to occur. In order to model correctly the 3D viscous aerodynamic phenomena present in LEI kite flight a computational approach utilizing a steadystate ReynoldsAverageNavierStokes (RANS) solver has been suggested. This work presents a review of relevant literature, outlines the computational approach taken, and discusses the limitations and computational costs of the approach. It was found that the RANS approach is able to model the kite s flow environment up to angles of attack of 24?. At angles larger than this significant flow separation from the suction surface of the kite precludes the use of a steadystate solver. At angles as low as 18? significant nonlinear effects begin to take effect, decreasing lift and increasing drag. It was also found that at lower angles of attack separation from behind the leading edge tube serves to decrease effective camber and therefore lift. The computational cost of the approach is heavily influenced by the quality of the mesh generated, in particular the presence of nonorthogonal cells. It is concluded that the RANS approach is capable of quantifying well the nonlinear flow effects of LEI kites at moderately high angles of attack. The challenge of this method in the future will be to decrease it s significant computational costs so that it may be used in the context of systems modeling, optimization, or fluidstructure interaction.<Windenergy; kite power; aerodynamics; wind energy; RANS; CFDWind EnergyEuropean Wind Energy Master)uuid:d9daf737a58342c3b8e9590fd7d66acbDhttp://resolver.tudelft.nl/uuid:d9daf737a58342c3b8e9590fd7d66acbEForce characterization for a submerged velocity cap in unsteady flows[Doni, Giovanni (TU Delft Civil Engineering and Geosciences; TU Delft Hydraulic Engineering)Bricker, Jeremy (mentor); Hofland, Bas (graduation committee); Uijttewaal, Wim (graduation committee); Jacobsen, Niels G. (mentor); de Fockert, Anton (graduation committee); Delft University of Technology (degree granting institution)
Coastal facilities such as desalination plant or a nuclear power plant need a continuous discharge of salty water to carry out their functions. Hereby an intake structure, such as a velocity cap, can be used to take in water from the sea. These intakes are open seafloor founded constructions mounted at a water depth ranging between 10 and 20 meters. In order to design such an intake cap in an offshore environment, the action of waves and in general of unsteady flows is the most important design load to be accounted for. This thesis has the aim to investigate the nature of the forces and turning moment acting on the structure in presence of waves and to provide tools, < in the form of hydrodynamic coefficients, that can be used to compute the design loads. The analysis is based at first on experimental records collected during a previous campaign including force measurements and PIV recordings. The measurements are then used to validate a CFD model in OpenFOAM. Structureinduced turbulence is shown to be fundamental to define the total loads on the structure in the numerical model. The use of a turbulence closure is in fact observed to be needed in order to come to a validation of the CFD model. Even if the flow separation around the cap is not always accurately predicted, the peak of the inline force is estimated by the model with an error of 8%. In the case of the vertical load the error observed reaches up to 15% but part of the mismatch is attributed to a bias in the experimental records. The CFD tool is then used to generate additional test cases on solitary waves and regular waves in order to expand the scope of this research. The hydrodynamic force coefficients are defined fitting both the experimental records and the load estimates of the numerical model by means of the weighted least squares method. The inline force characterization follows the theory of the Morison equation which is shown to provide a good fit in all analyzed cases. Good agreement is found between numerical and experimental results with regards to the estimate of the inertia coefficient while the numerical estimate of the drag coefficient is up to 19 % lower than the experimental estimate. Vertical force and overturning moment signals are originally fitted with the most common formulas used in literature which however produced poor fits to the force and moment signals. New equations are suggested therefore to come to a better characterization of these loads. The best fits for the vertical force are found with an equation that includes the effect of the horizontal drag, the vertical drag and of the vertical inertia, while in the case of the turning moment the best fits are obtained with a combination of horizontal drag, vertical drag, horizontal inertia and vertical inertia.VIntake structure; Velocity cap; CFD; OpenFOAM; Morison's equation; hydrodynamic forces)uuid:5dbf825133a841f8a4542b1b55cbf1c7Dhttp://resolver.tudelft.nl/uuid:5dbf825133a841f8a4542b1b55cbf1c7WPrediction of the performance of ducted propellers with BEM and hybrid RANSBEM methodsNegrato, C.c
A ducted propeller configuration consists of a fixed annular nozzle surrounding the propeller. The nozzle has an airfoil shape which depends on the required performance of the system as well as the operating conditions. The flowaccelerating ducts provide a positive contribution to the thrust of the propulsor and they are used to increase the performance in heavy loading conditions, such as in the cases of tug boats or often for azimuthal thrusters. The performance of the propeller can be assessed with model tests or numerical simulations. As concerns the numerical simulations, boundary element methods (BEM) are used daily in the design stage for open propeller configurations, but the use for ducted propellers is still under development: viscous flow effects become important at the duct surface and the accuracy of the BEM method decreases. Alternatively, Reynolds averaged NavierStokes (RANS) simulations are possible but they require large CPU time so they cannot be used at the design stage routinely. In addition, a hybrid RANSBEM method was developed at the Maritime Research Institute of the Netherlands (MARIN). The hybrid approach couples the viscous flow solution with the boundary element method: the propeller is not physically present in the RANS simulations, where it is substituted with a body force distribution whose strength is given by the BEM. This research has the objective to determine the accuracy and limitations of the BEM and the hybrid RANSBEM approaches to predict the performance of ducted propellers in open water condition. Two test cases are considered. The propeller is the same for both cases (Ka470 propeller) whereas the duct is differen< t (namely duct 19A and duct 37). The BEM method requires a modification of the duct trailing edge geometry to enforce the modelling of the wake sheet attached to the duct. An new iterative scheme that automatically computes the change in geometry based on a pressureequality condition is developed; as a result, the boundary element method computations provide 2% to 7% accuracy in the prediction of the open water efficiency for the design condition and the high loading conditions. However, for the light loading conditions the BEM method is limited by the occurrence of flow separation at the duct outer surface. The RANSBEM method improves the prediction at the large advance ratios for the first test case. However, for the second test case there is a constant overprediction of the propeller thrust and torque. The reason for this overprediction is likely related to the extent of flow separation occurring at the duct inner surface for all operating conditions, even though the lack of validation data for the second geometry tested does not allow to confirm this hypothesis. Finally, another objective of this study is to provide insight on the flow behaviour at the gap between the propeller and the nozzle. The ultimate goal is to give guidelines on the modelling of the flow in that region in a potential flow context. Former CFD calculations are analyzed in detail with focus on the gap flow. It is observed that the gap flow is dominated by the detachment of a vortex at small chordwise positions from the blade pressure side to the blade suction side. This is the socalled tip leakage vortex. The vortex is seen to obstruct the gap at the midchord positions and it strongly interacts with the tip trailing edge vortex. RANSBEM; Ducted Propellers; CFD
20181208)uuid:11de9007cb024540bb423b537d86ef7eDhttp://resolver.tudelft.nl/uuid:11de9007cb024540bb423b537d86ef7eFlap SideEdge Noise Reduction
Crepain, T.P.+Aeroacoustics; CFD; Beamforming; Flap noise)uuid:af624f59e4774dbea5b739902a0b1b99Dhttp://resolver.tudelft.nl/uuid:af624f59e4774dbea5b739902a0b1b99PCavitation: CFD Analysis of Cavitation Dynamics in a ConvergingDiverging NozzleHCointe, Benoit (TU Delft Mechanical, Maritime and Materials Engineering)van Terwisga, Thomas (mentor); Schenke, Sren (graduation committee); Pourquie, MAthieu (graduation committee); Melissaris, Themis (graduation committee); Jahangir, Saad (graduation committee); Delft University of Technology (degree granting institution)eThe main goal of this master thesis is to reproduce by CFD computations the characteristic cavitation dynamics in a venturi which were experimentally discovered at TU Delft by Jahangir and al. The different cavitation regimes were obtained by modifying the global static pressure and flow velocity in the flow loop. Based on this, three different cloud cavitation shedding regimes were identified. One of the mechanisms appears at high cavitation number and can be attributed to the presence of a reentrant liquid jet. Another dominant shedding mechanism was found at lower cavitation number, with the presence of propagating bubbly shock waves. A transition regime was also observed where both mechanisms seem to coexist. A two phases flow model is implemented in the opensource software OpenFOAM. The transition regime from the bubbly shock to the reentrant jet dominated regime will be identified and compared to experimental findings. The three cavitation shedding regimes will be investigated by using different cavitation numbers. The simulations will mostly be conducted using an inviscid and incompressible solver.CFD; Cavitation; Fluid dynamics
20190319)uuid:4a40e302bdf1431981dea6aad9376d65Dhttp://resolver.tudelft.nl/uuid:4a40e302bdf1431981dea6aad9376d65dMultipoint aerodynamic shape optimization for airfoils and wings at supersonic and subsonic regimes/Mangano, Marco (TU Delft Aerospace Engineering)la Rocca, Gianfranco (mentor); Martins, Joaquim (mentor); Veldhuis, Leo (graduation committee); Dwight, Richard (graduation committee); Delft University of Technology (< degree granting institution)rThe secondgeneration of supersonic civil transport has to match ambitious targets in terms of noise reduction and efficiency to become economically and environmentally viable. Highfidelity numerical optimization offers a powerful approach to address the complex tradeoffs intrinsic to this novel configuration. Past and current research however, despite proving the potential of such design strategy, lacks in deeper insight on final layouts and optimization workflow challenges. Stemming from the necessity to quantify and exploit the potential of modern design tools applied to supersonic aircraft design, this work partially fills the gap in previous research by investigating RANSbased aerodynamic<br/>optimization for both supersonic, transonic and subsonic conditions. The investigation is carried out with the stateoftheart, gradientbased MDO framework \textit{MACH}, developed at University of Michigan's MDO Lab  which hosted the author for the 14month research stint. Details of the tool and a brief overview of supersonic aircraft design and modern aerodynamic optimization strategies are reported in the first part of this manuscript.<br/>After circumscribing the research niche, I perform single and multipoint optimization to minimize the drag over an ideal supersonic aircraft flight envelope and assess the influence of physical and numerical parameters on optimization accuracy and reliability. Leading and trailing edge morphing capabilities are introduced to improve the efficiency at transonic and subsonic flight speed by relaxing the tradeoffs on clean shape optimization. Benefits in terms of drag reduction are quantified and benchmarked with fixededges results. It is observed how the optimized airfoils outperform baseline reference shapes from a minimum of 4\% up to 86\% for different design cases and flight<br/>conditions. The study is then extended to the optimization of a planar, lowaspectratio, and lowsweep wing, using the same schematic approach of 2D analysis. I investigate the influence of wing twist alone and twist and shape on cruise performance, obtaining a drag reduction of 6\% and 25\% respectively as the optimizer copes with both viscosity and compressibility effects over the wing. Results for 3D multipoint optimization suggest that the proposed strategy enables a fast and effective design of highlyefficient wings, with drag reduction ranging from a minimum of 24\% up to 74\% for cruise at different speeds and altitudes, including edge deflection. Ultimately, this work provides an extensive and, to the best of author knowledge, unprecedented insight on the optimal design solutions for this specific aircraft configuration and the challenges of the optimization framework. The benefits of RANSbased aerodynamic shape optimization to capture nonintuitive design tradeoffs and offer deeper physical insight are ultimately discussed and quantified. Given the promising results in terms of performance improvements and design efficiency, it is hoped that this work will foster the implementation of this method for more comprehensive fullconfiguration, multidisciplinary supersonic aircraft optimization studies.mOptimization; Aerodynamics; MDO; CFD; Supersonic; wing design; Airfoil; morphing; Gradientbased Optimization)uuid:82c6b30c768c46d7bbddccc086e8c28bDhttp://resolver.tudelft.nl/uuid:82c6b30c768c46d7bbddccc086e8c28bMAerodynamic optimisation and thermal loss modelling of a radial micro turbine
Subramani, K.Micro turbines are miniaturised gas turbines and have the potential to be effectively used in combined heat and power (CHP) applications due to their advantages over reciprocating engines. Micro Turbine Technology B.V. is developing a 3kW recuperated micro turbine for a microCHP application and is currently looking for ways to improve the efficiency of the system. At present, offtheshelf turbocharger components are used to reduce costs and there is a scope for improvement by shifting to a bespoke design. With the reduction in the size of the turbine, the volumetosurfa< ce ratio decreases and hence the heat transfer effects become important. The decrease in performance due to thermal losses needs to be evaluated in order to make an accurate prediction of the efficiency of the turbine. In this thesis, a numerical model is developed for estimating the heat lost to the environment and to the bearing housing using a lumped analysis method. This is coupled to a CFD simulation of the volute and the rotor to provide a diabatic boundary condition for the solver. The results from the thermal loss model are compared with existing data from MTT and it was found that the model can predict the thermal losses with good accuracy. To improve the efficiency of the rotor, a 3D shape optimisation is performed using the tools in ANSYS suite. To ensure that the new design can be directly used in the system, systemlevel parameters like shaft speed and mass flow rate are constrained. The important geometric parameters are parametrised and a response surface optimisation is setup. A lowerorder model of the design space is built and due to the large number of input parameters, a significant number of refinement points are added to ensure good accuracy of the response surface. In the next stage, a twostep optimisation is executed. Gradient based optimisers are computationally less expensive, but are prone to converging at local optimum values. This drawback can be overcome using evolutionary algorithms, but they require more computational time to converge. Therefore, an evolutionary algorithm with a relatively relaxed constraint is used to identify the location of the global optimum and a gradientbased search is performed in this confined design space to accurately identify the optimum design point. The influence of thermal losses in determining the optimised design is also evaluated. From the results, it was seen that the optimum point does not change significantly when the thermal losses are included and hence it can be inferred that the thermal loss evaluation can be decoupled from the optimisation process. An improvement of 2.17 percentage points in efficiency is achieved through the optimisation while respecting the mass flow rate constraint and also imposing a limit on the leading edge hub radius to ensure the stresses do not exceed permissible levels. However, when the limit on the leading edge radius is removed, the efficiency increased by 5.36 percentage points. Based on a detailed study of the optimised geometries, it was seen that the blade angle distribution and the leading edge hub radius are the most important parameters influencing the efficiency of the rotor. The blade widths at the inlet and exit can be used to control the mass flow rate in the system and compensate for any change in mass flow rate that is caused due to the variation of the other parameters. The influence of the trailing edge sweep on the efficiency is also studied and it was seen that effect of the sweep angle is strongly influenced by the shroud blade angle distribution. The future work based on this thesis can be directed towards including a structural solver inside the optimisation loop to ensure that the optimiser also consider the stress constraints that need to be satisfied for a feasible design. Also, it was clear that the inflow angles are not at the optimum value and hence a redesign of the volute is suggested to reduce the incidence losses.ACFD; micro turbine; thermal losses; optimisation; lumped analysis
20211221)uuid:fea305b0209b4c06bd85e780b8309b27Dhttp://resolver.tudelft.nl/uuid:fea305b0209b4c06bd85e780b8309b27Performance investigation of VentiFoil ship propulsion: Research into the propulsive performance of VentiFoils using CFD simulationszLagendijk, LaurensJan (TU Delft Mechanical, Maritime and Materials Engineering; TU Delft Marine and Transport Technology)#van Terwisga, Thomas (mentor); Gerritsma, Marc (graduation committee); Vrijdag, Arthur (graduation committee); van der Bles, G (graduation committee); Eggers, Rogier (graduation committee); Garenaux, Maxime (graduation committee); Delft University of < Technology (degree granting institution)1In recent years the awareness and the effort to reduce air pollution and global warming have increased. Many new ideas are being developed to reduce harmful emissions. Despite this, the shipping industry is still a large contributor to air pollution worldwide. To reduce the environmental impact of shipping, the use of<br/>sustainable energy sources such as wind energy onboard ships is being explored. Wind energy is widely available at sea, the challenge is to harness this energy. The relatively unknown wind propulsion device called the Turbosail is a vertical wing shaped device which uses wind energy to provide thrust. This propulsion technology was invented in the 1980 s by the Frenchman JacquesYves Cousteau. This report describes the investigation into the similar wind propulsion device called the VentiFoil. Two retractable VentiFoils are fitted inside a 40 foot container, this ship propulsion device is called the eConowind unit. Multiple eConowind units can be installed on the hatch covers of general cargo vessels. If successful, this wind propulsion device can be applied on many different ships. The VentiFoil concept will be investigated and improved using Computational Fluid Dynamics (CFD) research tools. These tools are used to simulate the flow around different VentiFoil geometries. The result of this project will be a better understanding of the working principles of the VentiFoil, sensitivity information for the variation of different characteristic design parameters and an evaluation of the generated forces and performance of VentiFoil propulsion.>VentiFoil; Turbosail; Wind assisted ship propulsion; WASP; CFD
2024010151.971114 5.654773)uuid:a64199f2d56740d4a536beee349adebcDhttp://resolver.tudelft.nl/uuid:a64199f2d56740d4a536beee349adebcD3D Numerical Modelling of Sediment Transport under Current and WavesSaud Afzal, M.7Asger Arntsen, O. (mentor); Sebastian Bihs, H. (mentor)
The sediment transport module of REEF3D is used to calculate the scour and the deposition pattern for abutment, pier and contraction under constant discharge and for pier under waves. The time development of scour for all these cases is also observed. To this effect, the ReynoldsAveraged Navier Stokes (RANS) equations are solved in all three dimensions, making it fully three dimensional. The location of the free surface is represented using level set method. The eddy viscosity in the RANS equation is determined by the use of the twoequation k? model and k? model. Using the conservative finitedifference framework on a structuredstaggered grid, convective terms of RANS equation is discretized with the fifthorder WENO scheme. The pressure gradient term in the RANS equation is modelled using Chorins projection method on a staggered grid. For the implementation of waves, the CFD code is used as a numerical wave tank. For the representation of the moveable sediment bed, the level set method is used. Kovacs and Parker and Dey formulations of bed shear stress reduction due to the sloping bed is implemented along with the sand slide algorithm to take care of erosion of individual bed cells. In the first case, the numerical model results are validated against the experimental findings of the abutment scour study done at Politecnico di Milano, Dept. I.I.A.R., Milan, Italy.. This test case is used as a benchmark for validation of the sediment transport module of REEF3D. The effect of the grid size, the turbulence model, the time discretization scheme, the formulations of critical shear stress for the sloping bed and porosity was observed and compared against the experimental results.. In the second configuration, the numerical model is used to predict the scour pattern around a circular pier. The scour pattern around the pier is compared against the experimental data of the hydraulic laboratory of the Technical University Darmstadt. The performance of the turbulence model and the formulations of critical shear stress for the sloping bed was observed. The numerical model is then tested for predicting a < general contraction scour. The result is compared against the physical experiments of the contraction case conducted at the laboratories of the BAW (Federal Waterways Engineering and Research Institute), Karlsruhe, Germany. For the above mentioned three cases, a constant discharge at the boundary is used. The numerical model in the above configurations predict the general evolution (geometry, location and maximum scour depth) and time development of the scour hole accurately. In the final configuration, 3D local scour around a vertical pile under waves is modelled. The numerical result is first compared with theoretical observations and then validated against the pier scour experiments conducted in the Department of Hydrodynamics and Water Resources (ISVA), Technical University of Denmark by Sumer and Fredsoe. The effect of the variation of the sediment time stepping with the decoupling of the hydrodynamic and the morphodynamic time step is tested. The numerical model shows good agreement with the experiment and theoretical erosion and deposition pattern. The decoupled approach for the simulation of hydrodynamic and sediment transport processes is found to be a reasonable assumption.QSediment Transport; CFD; REEF3D; Numerical Wave Tank; Abutment; Pier; ContractionCOMEM)uuid:e82f94f0f56e4a14a5d97c4c27d2d8d0Dhttp://resolver.tudelft.nl/uuid:e82f94f0f56e4a14a5d97c4c27d2d8d0=DNS study of scalar transport in a compressible turbulent jetRNayak, Pratik (Mechanical, Maritime and Materials Engineering; Process and Energy)[Boersma, Bendiks Jan (mentor); Delft University of Technology (degree granting institution)Direct Numerical simulations (DNS) belong to the class of simulations that strive to emulate the real physical flow by trying to simulate the complete range of scales involved in the flow. Though the computational power has steadily increased in the past few decades, resolving the complete range of scales from the largest scales up to the Kolmogorov scales can be very complicated and time consuming for moderately high Reynolds numbers as well. Hence, DNS is still expected to be a research tool to study relatively simple flows and geometries for the forseeable future unlike its counterparts such as Large Eddy Simulations (LES) and Reynolds Averaged NavierStokes (RANS) methods. The main objective of DNS studies is to study the flows so that turbulence models which are parametrized can be improved upon with the physical insight from the results of DNS. It also helps in studying the effect of numerical methods and techniques applicable to real flows and their standalone effects without any type of modeling as done for LES or RANS.<br/><br/>This thesis is concerned with the DNS study of a compressible turbulent jet. A turbulent jet belongs to the class of free turbulent flows, in the sense that, it is not bounded physically. In this thesis, the transport of a passive scalar through the jet is studied. At higher Reynolds numbers, for convection dominated equations such as the scalar transport equation, discontinuities can occur due to the hyperbolic nature of the equations. When central methods are used, oscillations are observed in the regions of the discontinuities. These oscillations can make the solution unphysical. Therefore, in this thesis the scalar transport equation has been modeled with an additional Weighted Essentially NonOscillatory (WENO) interpolation to accurately capture the discontinuities without oscillations. The WENO interpolation is combined with central compact finite difference methods to reduce the numerical dissipation while maintaining the order of accuracy in the smooth regions which is essential in high Reynolds number flows. <br/><br/>It was observed that a high dissipation setting for the WENO interpolation removed the oscillations but introduced artificial numerical viscosity. Therefore a relatively large domain was used with a low dissipation which while removing the oscillations and reducing the numerical dissipation. The WENO method for the compressible turbulent jet was first validated with< experimental results to ascertain the accuracy of the grid resolution. The same grid resolution was used to study some properties of the jet at two different Schmidt numbers at a slightly lower Reynolds number. It was observed that the results obtained with the WENO interpolation matched well with the experimental results while being physical valid solutions because they had no oscillations. The results also showed different modes of instabilities for different Schmidt numbers and that the decay rate is a good characterization of the properties of the turbulent jet.<br/><br/>In conclusion, the WENO methods can be a very helpful method to accurately capture the discontinuities in an efficient manner and is also suitable for methods such as LES and RANS to model flows that have some hyperbolic nature of terms in the governing equations.CFD; DNS; Turbulence)uuid:215ed6ef5fcb4366b39eb2706050d580Dhttp://resolver.tudelft.nl/uuid:215ed6ef5fcb4366b39eb2706050d580\Heat transfer enhancement of a permanent magnet synchronous machine used in vehicle tractionSrisankar, V.B.1Polinder, H. (mentor); Van Der Geest, M. (mentor)DOne of the major restrictions that hinder the versatility of electrical machines in any sector is their thermal limitations. This thesis deals with the study of the heat transfer occurring in an outer rotor permanent magnet synchronous machine used in the electric vehicle industry with the aim of exploring possibilities to reduce the hotspot temperature. The project is in collaboration with eTraction, an electric vehicle systems designer. A 3D numerical model of the machine is designed using the Finite Element Method (FEM) with heat transfer coefficients obtained from a computational fluid dynamics (CFD) simulation. A closer look at the secondary flow with vortices in curved tubes is also presented as part of the pursuit to estimate boundary conditions required for the FEM model. The equivalent thermal conductivity of the slot region was estimated by a 2D FEM model and the external heat transfer coefficients were represented using empirical relations. The model was validated with experimental tests performed at the eTraction's test bench facility and was found to be sufficiently accurate to predict the influence of material and mechanical design changes on the heat transfer. Results show that by using certain potting materials and making slight changes in the cooling region, hotspot temperatures can be significantly reduced.>PMSM; FEM; CFD; Heat Transfer machines; Dean Vortices; Potting8Electrical Engineering, Mathematics and Computer Science+DC Systems , Energy Conversions and Storage)uuid:9f1f56126f1547f89fb72f0239f5ead2Dhttp://resolver.tudelft.nl/uuid:9f1f56126f1547f89fb72f0239f5ead2iInternal flow simulation of an ESP air purifier using RANS flow field and Lagrangian tracking of aerosols0Goyal, Jatinder (TU Delft Aerospace Engineering)Gerritsma, Marc (mentor); Vons, Vincent (graduation committee); Delft University of Technology (degree granting institution)The gradually increasing air pollution calls for more and more effective solutions. This master thesis is focused on the investigation of aerosol trajectories inside the agricultural purifier named `ASPRA Agro', developed at VFA Solutions. Various CFD simulations have been carried out using an opensource C++ toolbox named `OpenFOAM'. The flow field was solved by carrying out axisymmetric simulations using a pressure jump boundary condition for the perforated wall. The comparison with the experimental results reveals that the axisymmetric simulations can capture the radial velocity profile gradients closely, but there is a considerable difference in velocity values. This difference has been ascribed to various factors such as wall effect due to mounting the purifier on the stand during the experiment, suction pockets due to the joints between the perforated wall. For electrostatic simulations, a very fine mesh is required, due to large gradients near corona needles. An axisymmetric simplification has been made to reduce the required computation< al effort, which increases the current to twice the expected value. The increase in current directly increases the particle charge. However, particle trajectories are unaffected because of negligible contribution from the electrostatics, assuming particles to be oneway coupled with the flow and electrostatics. OpenFOAM tools developed during the thesis will help in future simulations and the results obtained will aid in the further development of the purifier.\CFD; OpenFOAM; Air purifier; Electrostatic Precipitators; RANS; Lagrangian particle tracking
20241130)uuid:3c1c9416aa2647cdb77bd8ba00824bb1Dhttp://resolver.tudelft.nl/uuid:3c1c9416aa2647cdb77bd8ba00824bb15CFD modeling for moving jet penetrating cohesive soilEWang, Boyao (TU Delft Mechanical, Maritime and Materials Engineering)Keetels, Geert (mentor); Jarquin Laguna, Antonio (graduation committee); van Zuijlen, Alexander (graduation committee); Delft University of Technology (degree granting institution)Tclass="MsoNormal">In dredging operation, the highpressure water jet is widely used for the excavation of soil. To study the jetting process and optimize the dredging devices, the moving vertical water jet penetrating cohesive soil experiments were carried out by Nobel (2013). However, in terms of the design optimization for the dredging devices, it is not easy to change the jet scale and soil properties during the experiment due to time and economic constraints. Some detailed physics during the jetting process, e.g. pressure on the soil surface and shear plane inside the soil during jetting, were also not monitored during the experiment. Therefore, numerical simulation is chosen to optimize the design of dredging devices and study the physics of the jetting process. A CFD (computational fluid dynamics) numerical model is built to simulate the moving jet penetrating cohesive soil. The soil is modeled as a Bingham plastic. The sediment transport is modeled by using driftflux model. The moving jet modeling is achieved by using dynamic mesh algorithms AMI (arbitrary mesh interface) and A/R (cell layer addition removal). The CFD numerical model has been validated with the experiment of Nobel. After the validation, an analysis of the jetting process based on this CFD model is accomplished proving that the CFD model can reveal the details of the soil failure process during jetting. This thesis work reveals that it is possible to describe the hydraulic excavation of cohesive soil with reasonable accuracy using CFD numerical model. The CFD model can also reveal the details of the soil failure process that could not be retrieved from the experiment. Since the model is generic, it can be applied for a jet bar with multiple nozzles. This is helpful to improve the design of dredging equipment, optimize the operational settings and estimate the production.>CFD; cohesive soil; Moving jet; dynamic mesh; Driftflux model)uuid:5e5cd914cc094e06aa1f37f5aa7611cbDhttp://resolver.tudelft.nl/uuid:5e5cd914cc094e06aa1f37f5aa7611cbSNumerical Simulation of DeepSea Minings Plume Using Arbitrary, NonOrthogonal MeshfAgung Christy Rado Togaraja Simarmata, Agung (TU Delft Mechanical, Maritime and Materials Engineering)Keetels, Geert (mentor); van Rhee, Cees (graduation committee); van Grunsven, Frans (graduation committee); Delft University of Technology (degree granting institution)In recent years, the demand for minerals and rareearth elements are escalating due to rapid technological advancements and developments. This condition raises the importance of DeepSea Mining (DSM) as an option to fulfill the global demands, for the sake of future ambitious projects. On the other hand, DSM still faces some drawbacks and obstacles in its operations, e.g. environmental impact of its tailings discharge. Thus, the presence of tools for predicting the behavior and environmental impact of DSM tailings becomes crucial for the sake of conducting sustainable DSM operations. Researches, both numerical and laboratory experiments, are then done to achieve this goal. In the study of DSM tailings < behavior through numerical simulation, the challenge lies in the ability to implement the complex physics phenomena around DSM plumes to a numerical model. This research is thus aimed to observe one of the parts of the socalled physical phenomena, and numerical constraints on the simulation of DSM plumes: the effects of implementing arbitrary nonorthogonal mesh. Arbitrary nonorthogonal mesh would give users the freedom to refine the mesh based on the required resolution on a certain area in the simulation domain. In this research, an arbitrary nonorthogonal mesh is constructed forming a domain geometry of 3D tank with round pipe as a source of tailings discharge, adapting the laboratory setup of Byishimo (2018). Mesh refinement is done around the pipe discharge area, where the jet mainstream expected to occur, and near the bottom, where settling of solids fraction and high velocitygradients are expected to occur. Tophat approximation theory is then used for defining the inlet boundary condition of the simulation domain, and smooth solid wall is used for the bottom boundary. In observing the effect of arbitrary mesh, two parameters are varied: solids settling rate, and differentiation scheme. Following this, six simulation cases are prepared, containing three solidssettling conditions (minimum, realistic, and extreme solids fraction settling) and two differentiation schemes (Gauss Gamma and Gauss Linear). These cases are then simulated using the CFD driftflux model in OpenFOAM for two incompressible fluids, with ambient fluid and tailings mixture as the two incompressible fluids. Turbulence is modeled using LES method, with WallAdapting Local Eddyviscosity (WALE) LES model for modeling the subgridscale of the turbulence. Two simulated field variables are picked to be observed and compared with the laboratory measurement data: flow velocities and local Suspended Solids Concentration (SSC). Simulations show that the constructed domain able to generate stable results using not only Gauss Gamma differencing scheme but also Gauss Linear, which originally expected to give unbounded results. From simulations with various solids settling conditions, it is analyzed that the implementation of solids settling using the mentioned function leads to constant settling rate with inability to resuspend the settled solids. Thus, simulation of cases with extreme solids settling leads to hyperconcentration of solids fraction in the cells on the bottom. The tophat profile of the inlet boundary condition is constantly sustained throughout the simulation, resulting in uniform jet vertical velocity profile. The simulations also show that momentumdriven jetlike flow can be observed around the impingement point, while gravity current generated further from the impingement point. The simulation results show that the mesh refinement is enabling simulating and validating the flow with the required resolutions. Moreover, the constructed domain also shows that Gauss Linear can be used for simulating DSM plumes, furthermore using full tank domain. The simulated SSC also turns out not only affected by the simulation domain and mesh, but also the differencing scheme, and the presence of settling and pickup functions.FCFD; Deepsea mining; Nonorthogonal mesh; tailings plume; buoyantjet)uuid:1b6bdccb52754d2cbaebd7a5d423b1e0Dhttp://resolver.tudelft.nl/uuid:1b6bdccb52754d2cbaebd7a5d423b1e0dDetached Eddy Simulation applied to threedimensional aerodynamic full car simulations in motorsport2de Jong, Danille (TU Delft Aerospace Engineering)xvan Zuijlen, Alexander (mentor); Dutzler, Gerhard (mentor); Delft University of Technology (degree granting institution)? Detached Eddy Simulation is investigated as a potential method to improve the design process in the Aerodynamic Prototype Development Department of Porsche Motorsport. Detached Eddy Simulation is a CFD method that combines the scale resolving capabilities of Large Eddy Simulation with the low cost modelling of the ReynoldsAveraged NavierStokes (RANS) equations. The method promises to captu<re unsteady flow phenomena better than RANS with only a limited increase in costs. Various DES variants exist, which use different methods to switch between the RANS and the LES regions in the flow domain. The main method used is the Improved Delayed Detached Eddy Simulation (IDDES) method. This is a combination of the Delayed Detached Eddy Simulation (DDES) method and a WallModelled LES (WMLES) method. The DDES method shields the full attached boundary layer from LES and models it with RANS. The WMLES method does the opposite, allowing the use of LES within the boundary layer. IDDES chooses the most suitable one of these methods locally, to ensure a high accuracy and a robust method. The performance of the method when simulating limited span airfoil models and complex full car geometries is assessed and improved. A number of parameter variations are done on the airfoil models to improve understanding of the method through less costly simulations. The main application domain however, is motorsport. Three different aerodynamically complex car geometries are simulated with the IDDES method. The performance of the method is assessed by comparison with various RANS results and wind tunnel data in the form of axial force data, pressure taps and PIV data. Some parameter variations on the car are also tested to improve the results and obtain a reliable design process. The IDDES method is able to visualize more flow phenomena and characteristics. Additionally, drag and the impact of ride height changes are predicted well. Downforce is unfortunately generally overpredicted. There are indications that the results might be improved by varying the inflow conditions and by testing with finer time steps and/or meshes. Because of the costs involved and the desire to investigate the method as part of an extensive design process, this was deemed outside the scope to the master thesis. It is however, a recommended step for further exploration of the topic.KCFD; Detached Eddy Simulation; Motorsport; Full Car Simulations; DES; IDDES
20241115
*+&ffffff?'ffffff?(?)?"dXX333333?333333?U}}}}}}}}}} }
}}}
}}}}}}}}}}}}
@
!
"
#
$
%
&
'
(@
)
*
+
,
!

.
/
0
1
2
3x@
4
5
+
6
7
8
9
:
;
<
=@
>
?
+
@
A
B
C
D
E
F
G@
H
I
+
J
K
L
M
N
O
P@
Q
R
+
S
T
U
V
W@
X
Y
+
Z
[
\
]
^
_
`@
a
b
+
c
@
B
d
e
f
g
h
i @
j
k
+
7
l
m
n
o
p
q
r
@
s
t
+
u
7
v
w
x
y
z
{@

}
+
~
@
+
@
+
@
+
@
+
@
+
@
+
@
+
@
+
7
@
+
@
B
B
@
+
7
@
+
@
@
+
@
+
h@
+
7
@
+
@
+
@
@
+
@
+
@
+
x@
+
@
@
!
"
+
7
#
K
!$
!%
!&
!'
!(!@
!)
!*
!
! +
!@
!+
",
"
".
"/
"0"@
"1
"2
"
" +
"m
#3
#4
#5
#6
#7#@
#8
#9
#
# +
#@
#+
$:
$;
$<
$=
$>$@
$?
$@
$
$ +
$7
$A
$B
%C
%D
%E
%F
%G%@
%H
%I
%
% +
%7
%A
%B
&J
&K
&L
&M
&N&p@
&O
&P
&
& +
&7
&l
&Q
'R
'S
'T
'U
'V'@
'W
'X
'
' +
'Y
(Z
([
(\
(]
(^(@
(_
(`
(
( +
(a
(7
(b
(
)c
)d
)e
)f
)g)@
)h
)i
)
) +
)j
*k
*l
*m
*n
*o*@
*p
*q
*
* +
*r
*K
*s
+t
+u
+v
+w
++@
+x
+y
+
+ +
,z
,{
,
,}
,~,@
,
,
,
, +
,




@



 +

.
.
.
.
..@
.
.
.
. +
.@
.B
.
.
.
/
/
/
/
//@
/
/
/
/ +
/
/
0
0
0
0
00@
0
0
0
0 +
0
0
1
1
1
1
11@
1
1
1
1 +
1
17
1
1
2
2
2
2
22l@
2
2
2
2 +
2
2
2
2
3
3
3
3
33@
3
3
3
3 +
3B
4
4
4
4
44@
4
4
4
4 +
4B
5
5
5
5
55@
5
5
5
5 +
5
5
5
5
6
6
6
6
66@
6
6
6
6 +
6@
6
7
7
7
7
77@
7
7
7
7 +
7
8
8
8
8
88@
8
8
8
8 +
9
9
9
9
99@
9
9
9
9 +
9
:
:
:
:
::@
:
:
:
: +
:@
:B
;
;
;
;
;;@
;
;
;
; +
;
;@
<
<
<
<
<<@
<
<
<
< +
<
=
=
=
=
==x@
=
=
=
= +
=
=
=
=
>
>
>
>
>>@
>
>
>
> +
?
?
?
?
??X@
?
?
?
? +
?@
?
?
@!
@"
@#
@$
@%@@
@&
@'
@
@ +
@(
A)
A*
A+
A,
AA@
A.
A
A +
B/
B0
B1
B2
B3B@
B4
B5
B
B +
B
C6
C7
C8
C9
C:C@
C;
C
C +
C<
C@
C
D=
D>
D?
D@
DAD@
DB
DC
D
D +
ED
EE
EF
EG
EHE@
EI
EJ
E
E +
EK
FL
FM
FN
FO
FPF@
FQ
FR
F
F +
F7
FS
GT
GU
GV
GW
GXG@
GY
GZ
G
G +
G
G@
G
H[
H\
H]
H^
H_Hx@
H`
Ha
H
H +
H
H
H
H
Ib
Ic
Id
Ie
IfI@
Ig
Ih
I
I +
IB
Ji
Jj
Jk
Jl
JmJ@
Jn
Jo
J
J +
J
Kp
Kq
Kr
Ks
KtKl@
Ku
Kv
K
K +
K
Kw
Kx
Ly
Lz
L{
L
L}L@
L~
L
L
L +
L
L
M
M
M
M
MM@
M
M
M
M +
M
M
M
M
N
N
N
N
NN@
N
N
N
N +
N
O
O
O
O
O^O@
O
O
O
O +
O7
O
P
P
P
P
PP@
P
P
P
P +
P
Q
Q
Q
QQx@
Q
Q
Q
Q +
Q
Q
Q
R
R
R
R
RR@
R
R
R
R +
R
S
S
S
S
SS@
S
S
S
S +
S
S7
S
S
T
T
T
T
TT@
T
T
T +
T
T7
T
T
U
U
U
U
UU@
U
U
U
U +
U
U7
U
UK
V
V
V
V
VV@
V
V
V
V +
V7
V
V
W
W
W
W
WWx@
W
W
W
W +
W
W
W
W
X
X
X
X
XX@
X
X
X
X +
X
X7
X
X
Y
Y
Y
Y
YYl@
Y
Y
Y
Y +
Y7
Yl
Y
Z
Z
Z
Z
ZZp@
Z
Z
Z
Z +
Z7
Zl
Z
[
[
[
[
[[@
[
[
[
[ +
[
\
\
\
\
\\@
\
\
\
\ +
\
]
]
]
]
]]@
]
]
]
] +
]@
]B
^
^
^
^
^^@
^
^
^
^ +
^@
_
_
_
_
__@
_
_
_
_ +
_
`
`
`
`
` `@
`!
`"
`
` +
`#
a$
a%
a&
a'
aa@
a(
a)
a
a +
a*
a@
a
b+
b,
b
b.
b/b@
b0
b1
b
b +
c2
c3
c4
c5
c6c@
c7
c8
c
c +
c9
c7
c
c:
d;
d<
d=
d>
d?d@
d@
dA
d
d +
dB
d@
eC
eD
eE
eF
eGe@
eH
eI
e
e +
e
fJ
fK
fL
fM
fNf@
fO
fP
f
f +
gQ
gR
gS
gT
gUg@
gV
gW
g
g +
hX
hY
hZ
h[
h\h@
h]
h^
h
h +
h7
h
i_
i`
ia
ib
ici@
id
ie
i
i +
i
jf
jg
jh
ji
jjj@
jk
jl
j
j +
jB
km
kn
ko
kp
kqk@
kr
k
k +
ks
lt
lu
lv
lw
lxl@
ly
lz
l
l +
m{
m
m}
m~
mm@
m
m
m
m +
n
n
n
n
n7n@
n
n
n
n +
n@
nA
nB
o
o
o
o
oot@
o
o
o
o +
o
o
o
o
p
p
p
p
pp@
p
p
p
p +
p
q
q
q
q
qqx@
q
q
q
q +
q
q@
q
r
r
r
r
rr@
r
r
r
r +
r@
s
s
s
s
ss@
s
s
s
s +
s
s7
sl
s
t
t
t
t
tt@
t
t
t
t +
t@
t
t
u
u
u
u
uu@
u
u
u
u +
u@
u
u
v
v
v
v
vv@
v
v
v
v +
v
w
w
w
w
w^w@
w
w
w
w +
w
w7
wS
w
x
x
x
x
x7x@
x
x
x +
x@
xB
y
y
y
y
yy@
y
y
y
y +
y
z
z
z
z
zz@
z
z
z
z +
z@
{
{
{
{
{{@
{
{
{
{ +
{
{@




@



 +



}
}
}
}
}}t@
}
}
}
} +
}
}
}
~
~
~
~
~~@
~
~
~
~ +
@
+
@
+
@
+
K
!@
"
#
+
K
$
%
&
'
(@
)
*
+
+
@
!"#$%&'()*+,./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{}~
!"#$%&'()*+,./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{}~
!"#$%&'()*+,./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{}~
!"#$%&'()*+,./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{}~
!"#$%&'()*+,./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{}~
!"#$%&'()*+,./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{}~
!"#$%&'()*+,./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{}~>@ddyKyKhttp://resolver.tudelft.nl/uuid:b7ccbf2752cf4d50ba649282aec248e4yKyKhttp://resolver.tudelft.nl/uuid:0da53df4df774afd9bc089ccca593719yKyKhttp://resolver.tudelft.nl/uuid:bfb6a4c71c60472c858ebc519fc46521yKyKhttp://resolver.tudelft.nl/uuid:267beaa58b624e58b1adff94355c68f9yKyKhttp://resolver.tudelft.nl/uuid:de63784a11d04cb39e27b32f92415c32yKyKhttp://resolver.tudelft.nl/uuid:f9a658fe9d0b4a1d808a908698c8e8c2yKyKhttp://resolver.tudelft.nl/uuid:21c1e156c8df4ab6863025c6170d3bc6yKyKhttp://resolver.tudelft.nl/uuid:b403c8ca64634e8a9e14d6a76fddd297 yKyKhttp://resolver.tudelft.nl/uuid:4e756c0704c8464f9c602c4efcab753c
yKyKhttp://resolver.tudelft.nl/uuid:dbd143ca790844479cf0193011e81984yKyKhttp://resolver.tudelft.nl/uuid:8b4d06026045437d816709203f8ac6aayKyKhttp://resolver.tudelft.nl/uuid:b7a76cb18dcc47e18e6f0b46b78f3f50
yKyKhttp://resolver.tudelft.nl/uuid:7a0b85902b9744818b0d0b832189ad88yKyKhttp://resolver.tudelft.nl/uuid:db908fbf66ca4b0286883d5b35b8deb2yKyKhttp://resolver.tudelft.nl/uuid:4c9d0463b31841ffa7ae94a139464b4eyKyKhttp://resolver.tudelft.nl/uuid:6c590444c6824f1ca271a053c0dc3ba7yKyKhttp://resolver.tudelft.nl/uuid:3b257aa35168441f9b168fad291edc1eyKyKhttp://resolver.tudelft.nl/uuid:75f3e25ab9914e85ab8a736a32383bb5yKyKhttp://resolver.tudelft.nl/uuid:3dea23d8f101495e8050bf8302f2e609yKyKhttp://resolver.tudelft.nl/uuid:5f8a549a443b40aab9283c2e205526b0yKyKhttp://resolver.tudelft.nl/uuid:0fd3498571434110965d8f6f1af59c01yKyKhttp://resolver.tudelft.nl/uuid:94162987534c4912ac6855b26df20585yKyKhttp://resolver.tudelft.nl/uuid:54552d42d0b64356b3b4049ad305a505yKyKhttp://resolver.tudelft.nl/uuid:178e2496cf4a440ba877ed34b2206284yKyKhttp://resolver.tudelft.nl/uuid:9db0dc728ec24d75b1f936dea1f40918yKyKhttp://resolver.tudelft.nl/uuid:3d1ff1ee3dc24a98861d37453354867fyKyKhttp://resolver.tudelft.nl/uuid:88a16dcdf4044d299badd3ffd0cc2023yKyKhttp://resolver.tudelft.nl/uuid:a47f9e446f684b39b9a989b16b99cf9cyKyKhttp://resolver.tudelft.nl/uuid:44818c8328984e34be3a75efd1a71e2dyKyKhttp://resolver.tudelft.nl/uuid:73c8079a17604fcaa05f166f836163a0yKyKhttp://resolver.tudelft.nl/uuid:2e3c46fa175d4580a1768f6c8c5425e9 yKyKhttp://resolver.tudelft.nl/uuid:d4c717e8809041eea6a06c7a243ea836!!yKyKhttp://resolver.tudelft.nl/uuid:d1c58f2e40dc43c68383f2041593d1c4""yKyKhttp://resolver.tudelft.nl/uuid:7985085a7aad4fce885369eaa5289b6c##yKyKhttp://resolver.tudelft.nl/uuid:ee2db063a1c04f0cb8a679f6f3acb749$$yKyKhttp://resolver.tudelft.nl/uuid:de6e82b528f14a7da073032620a88b60%%yKyKhttp://resolver.tudelft.nl/uuid:8656b6a70c73499e91bea617acf6c28c&&yKyKhttp://resolver.tudelft.nl/uuid:b3a53d4fa28a486da09a51ce441edc6f''yKyKhttp://resolver.tudelft.nl/uuid:9629143d68d444099f69a274690485e5((yKyKhttp://resolver.tudelft.nl/uuid:b8276e19acf74c298203d09d6922e1ad))yKyKhttp://resolver.tudelft.nl/uuid:e5465aeb59334b75b06889b7ad906296**yKyKhttp://resolver.tudelft.nl/uuid:731efb5525dc4cfdb972368cea65a2e4++yKyKhttp://resolver.tudelft.nl/uuid:87ad4bb361b54a55a9561db361f133c1,,yKyKhttp://resolver.tudelft.nl/uuid:c2edbffbb2244699984e545ade1b8815yKyKhttp://resolver.tudelft.nl/uuid:acf41200c4f440048d24dcc5f8d6dd7c..yKyKhttp://resolver.tudelft.nl/uuid:324b6057aeab4ad8ab245aa1a62819a0//yKyKhttp://resolver.tudelft.nl/uuid:9778a8246d724960a8124615a53c2c4100yKyKhttp://resolver.tudelft.nl/uuid:b1c5e142e5244cf594d1bdfeca28189711yKyKhttp://resolver.tudelft.nl/uuid:2c7f8595243f4c429c26187bf4976bce22yKyKhttp://resolver.tudelft.nl/uuid:d795958e2b8d41039723bf55db1b0cb733yKyKhttp://resolver.tudelft.nl/uuid:eda271970f434b83a2acd0bb564a7ef444yKyKhttp://resolver.tudelft.nl/uuid:54c220fbfcfb4e2c9dba40afc70e3b7f55yKyKhttp://resolver.tudelft.nl/uuid:fe21dbbe56ae40a8a4860490dcce913666yKyKhttp://resolver.tudelft.nl/uuid:6286f9e2c24a430ca4fa9fb67b9558b477yKyKhttp://resolver.tudelft.nl/uuid:700fdbda04074aaa9a047e1f7a64b03e88yKyKhttp://resolver.tudelft.nl/uuid:1843361f52ce4f80b651eec1e4e5633b99yKyKhttp://resolver.tudelft.nl/uuid:5a19d53360e4414ca713f4738236d669::yKyKhttp://resolver.tudelft.nl/uuid:e429748960e2403ea2461b1ea4c4ea63;;yKyKhttp://resolver.tudelft.nl/uuid:c5ddec13e10349b48b84a8ad013c753c<<yKyKhttp://resolver.tudelft.nl/uuid:0caecfdb85ec4f5fb5a9cc98dcb9722a==yKyKhttp://resolver.tudelft.nl/uuid:4792857c1c2e4c05bc0f9ca1ece4499c>>yKyKhttp://resolver.tudelft.nl/uuid:20e84104201e42b5a67ea91b6f9c5557??yKyKhttp://resolver.tudelft.nl/uuid:2099392381e744eaa2961f02893412f1@@yKyKhttp://resolver.tudelft.nl/uuid:ab209a8a079b44b1afb8f2159361a906AAyKyKhttp://resolver.tudelft.nl/uuid:7fce49ea3da0451494de15515f616190BByKyKhttp://resolver.tudelft.nl/uuid:e675fadcfd8d4cedb59e45b6faba6fddCCyKyKhttp://resolver.tudelft.nl/uuid:5ffe749f81aa42d1aed52f07580e8078DDyKyKhttp://resolver.tudelft.nl/uuid:3dd54665f48c4e489f57dc285cece612EEyKyKhttp://resolver.tudelft.nl/uuid:fdcf842311f04b33956e3e761635ac41FFyKyKhttp://resolver.tudelft.nl/uuid:957b762614584c84a7ce80ab64a7a121GGyKyKhttp://resolver.tudelft.nl/uuid:09dd8ba2303b470c8a8e4eeac2c44200HHyKyKhttp://resolver.tudelft.nl/uuid:57693312431545ffbf5682551871dc52IIyKyKhttp://resolver.tudelft.nl/uuid:244669d3f4be4df8baa5dca9ba9fc323JJyKyKhttp://resolver.tudelft.nl/uuid:ef4b9c51111f445fae272966a31e9d44KKyKyKhttp://resolver.tudelft.nl/uuid:6640f075f0b94c0f9b4a756a8ea085fcLLyKyKhttp://resolver.tudelft.nl/uuid:bd04b1ffc245403a9149b027d03a9a95MMyKyKhttp://resolver.tudelft.nl/uuid:5e2ab613335c427784a68b55c7ee5bbcNNyKyKhttp://resolver.tudelft.nl/uuid:df3b1b0294f6427cad844c4342dddf57OOyKyKhttp://resolver.tudelft.nl/uuid:2f18b7fe598146689777851513ca3acfPPyKyKhttp://resolver.tudelft.nl/uuid:bfcf5b18428848b48587918cc7899b86QQyKyKhttp://resolver.tudelft.nl/uuid:1a0b82791283470d87b2107e3ed90ec2RRyKyKhttp://resolver.tudelft.nl/uuid:fbf93d6e211f4b92a057956d694db315SSyKyKhttp://resolver.tudelft.nl/uuid:61b58a943a0141aebf8d2d105840bef4TTyKyKhttp://resolver.tudelft.nl/uuid:9bc88456e5374965a874666e5edae0c0UUyKyKhttp://resolver.tudelft.nl/uuid:9ef11baf576c4eb0a84db507590b6000VVyKyKhttp://resolver.tudelft.nl/uuid:632afd2e81de42e3a25e3e2d4eb1aad4WWyKyKhttp://resolver.tudelft.nl/uuid:71f03228175e40b09fd25be4480dcfecXXyKyKhttp://resolver.tudelft.nl/uuid:f8b6c47cb08e46d79358488602c50973YYyKyKhttp://resolver.tudelft.nl/uuid:ed6cc506622d49be9b3c0ab396b9c3d5ZZyKyKhttp://resolver.tudelft.nl/uuid:581b8923d82c470d9b1e8cd6b8a3b44b[[yKyKhttp://resolver.tudelft.nl/uuid:43faf68c7e5f459c9d1b88ec7c4c71b9\\yKyKhttp://resolver.tudelft.nl/uuid:ae6fc82c152d442899d5d531cc3ae6ba]]yKyKhttp://resolver.tudelft.nl/uuid:ec68055b990a45b5af269dd67372f6d1^^yKyKhttp://resolver.tudelft.nl/uuid:57fecdc34987439ca6fcf2cf6a1d70c3__yKyKhttp://resolver.tudelft.nl/uuid:1b6037cf6cb94ebfb9c225e7fa6cf575``yKyKhttp://resolver.tudelft.nl/uuid:e55916dbe9464bc099f3df3b768760cfaayKyKhttp://resolver.tudelft.nl/uuid:1ff626fb177d44b18095091923eb8f3abbyKyKhttp://resolver.tudelft.nl/uuid:0ab1c5b34a674b21b572a0384abdc576ccyKyKhttp://resolver.tudelft.nl/uuid:d42b40e023c14e49abb696a05a3dcd06ddyKyKhttp://resolver.tudelft.nl/uuid:aa09e42ab148468cbd7e0ec7cbfcdd97eeyKyKhttp://resolver.tudelft.nl/uuid:a53b78eed5c341dc85cf1b9da36109f8ffyKyKhttp://resolver.tudelft.nl/uuid:2c964dfc8b3d44ba9e3dcd22704aa862ggyKyKhttp://resolver.tudelft.nl/uuid:09dc7b81d2684e93be17b6f0c683c361hhyKyKhttp://resolver.tudelft.nl/uuid:85b601cc84544686afcb2344ec752ac1iiyKyKhttp://resolver.tudelft.nl/uuid:56d1ccc2b50a44a69908d5f495c6951cjjyKyKhttp://resolver.tudelft.nl/uuid:4e54cfb33024420a820548c7eb767ce9kkyKyKhttp://resolver.tudelft.nl/uuid:57f7c2b3918f433da804d82e04c639ddllyKyKhttp://resolver.tudelft.nl/uuid:86621aa390a64929953497459ef11f87mmyKyKhttp://resolver.tudelft.nl/uuid:89e3697b6d234cf28f31cfc2448b286annyKyKhttp://resolver.tudelft.nl/uuid:4ef56c0f1dbd4edb8b89f672dd3597ecooyKyKhttp://resolver.tudelft.nl/uuid:2d40af8cfcdf41a98aad05b16ad4f9bappyKyKhttp://resolver.tudelft.nl/uuid:0bef658bca1c4edbbaf998263b88fd7bqqyKyKhttp://resolver.tudelft.nl/uuid:fc21b131b75844e3842db854a935f5c6rryKyKhttp://resolver.tudelft.nl/uuid:0314e592ffa24f1dbc315da7011805b7ssyKyKhttp://resolver.tudelft.nl/uuid:2d9caa9462e348a8ab52b74d8c38816ettyKyKhttp://resolver.tudelft.nl/uuid:90c8d527721447ac8a0b0f03c40625ccuuyKyKhttp://resolver.tudelft.nl/uuid:ccb561540b704a41822324b0f8d145c5vvyKyKhttp://resolver.tudelft.nl/uuid:d9daf737a58342c3b8e9590fd7d66acbwwyKyKhttp://resolver.tudelft.nl/uuid:5dbf825133a841f8a4542b1b55cbf1c7xxyKyKhttp://resolver.tudelft.nl/uuid:11de9007cb024540bb423b537d86ef7eyyyKyKhttp://resolver.tudelft.nl/uuid:af624f59e4774dbea5b739902a0b1b99zzyKyKhttp://resolver.tudelft.nl/uuid:4a40e302bdf1431981dea6aad9376d65{{yKyKhttp://resolver.tudelft.nl/uuid:82c6b30c768c46d7bbddccc086e8c28byKyKhttp://resolver.tudelft.nl/uuid:fea305b0209b4c06bd85e780b8309b27}}yKyKhttp://resolver.tudelft.nl/uuid:a64199f2d56740d4a536beee349adebc~~yKyKhttp://resolver.tudelft.nl/uuid:e82f94f0f56e4a14a5d97c4c27d2d8d0yKyKhttp://resolver.tudelft.nl/uuid:215ed6ef5fcb4366b39eb2706050d580yKyKhttp://resolver.tudelft.nl/uuid:9f1f56126f1547f89fb72f0239f5ead2yKyKhttp://resolver.tudelft.nl/uuid:3c1c9416aa2647cdb77bd8ba00824bb1yKyKhttp://resolver.tudelft.nl/uuid:5e5cd914cc094e06aa1f37f5aa7611cbyKyKhttp://resolver.tudelft.nl/uuid:1b6bdccb52754d2cbaebd7a5d423b1e0gg
Root Entry F<<<<@SummaryInformation( F<Workbook FDocumentSummaryInformation8 F
!"#$%&'()*+,./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{}~
!"#$%&'()*+,./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{}~
!"#$%&'()*+,./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{}~
!"#$%&'()*+,./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{}~
!"#$%&'()*+,./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{}~
!"#$%&'()*+,./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{}~
!"#$%&'()*+,./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvw