"uuid","repository link","title","author","contributor","publication year","abstract","subject topic","language","publication type","publisher","isbn","issn","patent","patent status","bibliographic note","access restriction","embargo date","faculty","department","research group","programme","project","coordinates"
"uuid:b7ccbf27-52cf-4d50-ba64-9282aec248e4","http://resolver.tudelft.nl/uuid:b7ccbf27-52cf-4d50-ba64-9282aec248e4","Simulation Verification and Optimization of a Vertical Axis Wind Turbine using CFD","Kortleven, M.","Kenjeres, S. (mentor); Heemink, A.W. (mentor)","2016","In this research computational fluid dynamics (CFD) is used to model a vertical axis wind turbine. Well known turbulence models | i.e. the k-epsilon and k-omega 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 k-epsilon simulations approximate the experimental data the closest, but that the difference between the simulations and the experiment in the k-omega 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.","CFD; Simulation; Vertical Axis Wind Turbine; Optimization","en","bachelor thesis","","","","","","","","","Applied Sciences","ChemE/Chemical Engineering","","Transport Phenomena","",""
"uuid:5604201c-a4ba-4ace-8b75-b125c69e5628","http://resolver.tudelft.nl/uuid:5604201c-a4ba-4ace-8b75-b125c69e5628","Designing and optimizing an aerospike micro-nozzle for the sub-mN range: A numerical study","Sousa da Costa, Tiago (TU Delft Aerospace Engineering)","Cervone, A. (mentor); Delft University of Technology (degree granting institution)","2021","class=""MsoNormal"">Micro-propulsion technology is under rapid development and considerably extends the mission capabilities of small spacecraft. However, the flow in conventional nozzles on a microscale is relatively viscous, which, combined with the bounding nozzle walls, leads to low efficiencies. Aerospike nozzles, on the other hand, offer free-flow expansion and mitigate the boundary layer effects. The research objective of this project is to investigate the losses and output efficiency of micro-nozzles by optimizing a 3D axisymmetric aerospike in the μN-mN range. Planar nozzles, typically found in micro- propulsion subsystems, limit the flow radial expansion and considerably suffer from viscous losses, as well as momentum loss. In addition, in lower thrust settings, the boundary layer massively limits the performance of conventional nozzles – conical and bell layouts –, whereas aerospikes do not bound the flow and, subsequently, inherit better gas expansion. The design optimization relies on numerical analyses, with the free and open-source OpenFOAM’s solver, rhoCentralFoam, and self-developed Python scripts that automate the simulations in various machines. This study comprises three design iterations. The first contour parameters derive from Delft University of Technology’s Vaporizing Liquid Micro-thruster (VLM) requirements and the preceding work. However, since this work considers axisymmetric geometries, the aerospike’s throat width is reduced from 45 μm to 15 μm to preserve the original thrust magnitude (< 10 mN). The initial results show that, at the same throat Reynolds number, the aerospike outperforms the bell nozzle, especially towards lower Reynolds, where the specific impulse efficiency variation tops 25%. However, at equal thrust levels, the conventional design surpasses the aerospike thrust and specific impulse efficiencies. The Mach contours reveal that the small throat width and high area ratio ineffectively mitigate the viscous losses and lead to extreme momentum loss. The following iterations with four truncation percentages (20%, 40%, 60%, and 80%) prove the first hypothesis right: when decreasing the area ratio from ~17 to ~3 and raising the throat width to 30 μm, the efficiency of an 80% long aerospike reaches ~94% for the specific impulse and ~89% for the thrust. In addition, the aerospike yields the best performance when it is the least truncated (highest truncation percentage, i.e., 80%). Finally, a ±100 K temperature sensitivity study shows that the aerospike performance oscillates up to 3%, with a maximum thrust efficiency of 91% and specific impulse efficiency close to 98%, rivaling macroscale performance. With a small area ratio and a wide throat, the aerospike nozzle outperforms an equivalent bell nozzle by more than 10% in terms of specific impulse and thrust efficiencies.","Micro-propulsion; Aerospike; Optimization; CFD; 3D","en","master thesis","","","","","","","","","","","","Aerospace Engineering","",""
"uuid:f202b784-4f95-461a-b8cc-f1038643dd8e","http://resolver.tudelft.nl/uuid:f202b784-4f95-461a-b8cc-f1038643dd8e","Thermal Management of a Data Center White Space: A numerical study using computational fluid dynamics, flow & temperature predictions using neural networks, and white space design optimization using a genetic algorithm.","Pal, Somnath (TU Delft Mechanical, Maritime and Materials Engineering)","Pourquie, Mathieu (mentor); Smit, Stephan (mentor); Papadopoulos, Argyrios (mentor); Westerweel, Jerry (graduation committee); Langelaar, Matthijs (mentor); Delft University of Technology (degree granting institution)","2020","class=""MsoNormal"">The total electrical energy consumption by all the operational data centers (located all over the world) is enormous (approx. 1% of the global electricity demand 20000TWh). This electrical energy is required 24x7 to operate and cool all the IT equipment present in the data center. The electrical energy required to cool all the servers present in the white space can range from as low as 10% to as high as 40% of the total data center electrical consumption. The server's inlet temperature has to be within the ASHRAE recommended range (18o C – 27o C), so that they can function correctly. A simplified design of a raised floor white space with hot aisle / cold aisle configuration is considered. The tile flow rate through the floor tiles influences the server's inlet temperature. To control the tile flow rate, 11 design variables of the data center white space are identified. These are the position of the 4 perforated plates, the amount of perforations of each perforated plates, the floor tiles perforation, the raised floor height, and the CRAH distance to the cabinet. 600 design samples of the white space are generated by applying the Latin Hypercube Sampling (LHS) technique on these 11 design variables. A well-validated CFD software 6SigmaRoom by Future Facilities is used to generate the CFD results. The standard k-ε model is used to model the turbulence in the CFD simulations, in a steady-state condition. A database of 600 samples is generated by recording the cabinet's inlet temperature, flow rate of floor tiles from the CFD simulations, and the corresponding changeable design parameters generated by the LHS technique. The error due to CFD simulation is estimated at less than 4% for the tile flow rate and 1.7o C for the server inlet temperature. Four Artificial Neural Networks (ANN) are trained on the data from the database to predict the floor tile's tile flow rate and the cabinet's inlet temperature, respectively. The average R2 prediction (testing) accuracy is 0.97 for the tile flow rate predictions and 0.92 for the cabinet's inlet temperature predictions. Their average prediction error is less than 5% for the tile flow rate and less than 20 C for the cabinet's inlet temperature. The Non-dominated Sorting Genetic Algorithm-II (NSGA-II), a variant of the genetic algorithm, is used to find the optimum values of the 11 design parameters of the white space. These optimum values are going to ensure the server's inlet temperature to be within the ASHRAE recommended range. The genetic algorithm optimizes the design variables based on the predictions made by the neural network predicting the tile flow rate. The values of the optimized design parameters are verified using the 6SigmaRoom software by comparing the server's (mean) inlet temperature of the optimized case with the non-optimized case. More number of servers have their (mean) inlet temperature below 270 C in the optimized case as compared to the non-optimized case. The electrical power required by the CRAHs to cool the white space is reduced by 10% in the optimized case as compared to the power which is required by the CRAHs in the non-optimized case. In this study, a CFD simulation of the white space took 40 minutes. The neural networks took less than a minute to make the predictions, and the NSGA-II algorithm took less than 10 minutes to find the optimized design parameters of the white space. Thus, in this thesis, it is shown that using an artificial neural network and a genetic algorithm, in combination with computational fluid dynamics gives satisfying results in optimizing the white space design, required to keep the server's inlet temperature within the ASHRAE recommended range. The computational time required to find the optimum white space design is also reduced by using a neural network and a genetic algorithm. The prediction by the neural network and the optimization performed by the genetic algorithm can be improved further with the availability of more training data and in-depth knowledge of applying these techniques (the neural network and the genetic algorithm) in predicting and optimizing the solutions respectively. ","CFD; Neural Network; Genetic Algorithm; Turbulence Modelling; Data Center White Space; Prediction; Optimization","en","master thesis","","","","","","","","","","","","","",""
"uuid:f9a658fe-9d0b-4a1d-808a-908698c8e8c2","http://resolver.tudelft.nl/uuid:f9a658fe-9d0b-4a1d-808a-908698c8e8c2","Analysis of an over the wing based distributed propulsion system","Khajehzadeh, Arash (TU Delft Aerospace Engineering)","Veldhuis, L.L.M. (mentor); Hulshoff, S.J. (mentor); Delft University of Technology (degree granting institution)","2018","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 opportunities 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.
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.
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 fifty-nine LES samples tested, with a modest improvement in the mean velocity components. The non-local 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 Co-Kriging LES-HIRANS was not able to outperform the Co-Kriging LES-HIRANS 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 Co-Kriging HIRANS are suggested to be linked to an adequate error estimation integration into the surrogate model.","RANS; LES; HIRANS; Turbulence; CFD; Interpolation; Optimization; Kriging; Co-Kriging; Aerodynamics; Periodic hill; Correction; Prediction","en","master thesis","","","","","","","","2018-12-14","","","","Aerospace Engineering","",""
"uuid:4a40e302-bdf1-4319-81de-a6aad9376d65","http://resolver.tudelft.nl/uuid:4a40e302-bdf1-4319-81de-a6aad9376d65","Multi-point aerodynamic shape optimization for airfoils and wings at supersonic and subsonic regimes","Mangano, Marco (TU Delft Aerospace Engineering)","la Rocca, G. (mentor); Martins, Joaquim (mentor); Veldhuis, L.L.M. (graduation committee); Dwight, R.P. (graduation committee); Delft University of Technology (degree granting institution)","2019","The second-generation of supersonic civil transport has to match ambitious targets in terms of noise reduction and efficiency to become economically and environmentally viable. High-fidelity numerical optimization offers a powerful approach to address the complex trade-offs 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 RANS-based aerodynamic
optimization for both supersonic, transonic and subsonic conditions. The investigation is carried out with the state-of-the-art, gradient-based MDO framework \textit{MACH}, developed at University of Michigan's MDO Lab - which hosted the author for the 14-month 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.
After circumscribing the research niche, I perform single and multi-point 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 trade-offs on clean shape optimization. Benefits in terms of drag reduction are quantified and benchmarked with fixed-edges 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
conditions. The study is then extended to the optimization of a planar, low-aspect-ratio, and low-sweep 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 multi-point optimization suggest that the proposed strategy enables a fast and effective design of highly-efficient 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 RANS-based aerodynamic shape optimization to capture non-intuitive design trade-offs 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 full-configuration, multidisciplinary supersonic aircraft optimization studies.","Optimization; Aerodynamics; MDO; CFD; Supersonic; wing design; Airfoil; morphing; Gradient-based Optimization","en","master thesis","","","","","","","","","","","","Aerospace Engineering","",""