Suspension pipe flows can exhibit a behaviour called core-peaking where the particles accumulate in the centre of the pipe. This is due to shear-induced migration, where particles migrate towards areas of the flow with lower shear rate. While this concept is well documented, the exact causes are still unknown. Experimental research can uncover how this behaviour is impacted by different flow properties. This knowledge can be used to predict whether a given system will display core peaking behaviour. Knowing this a priori is convenient, as core peaking can impact the pressure drop in the pipe flow significantly. This thesis investigates the applicability of an experimental method that uses light attenuation to measure volume fraction distributions in a suspension pipe flow. Investigating this method is worthwhile as it is relatively quick and affordable compared to other methods like MRI. The theoretical relationship between the concentration of a substance and the attenuation of light is given by the Beer-Lambert law. However, this linear law does not hold for dense suspensions. To account for this, a set of calibration experiments was done in a setup where the path length was varied consistently. The results give a relationship between the attenuation and the amount of particles, expressed as the product of the volume fraction and the path length. This relationship is initially linear before it transitions to a cube root function for higher particle loadings. This change is thought to be due to multiple scattering becoming more prominent when more particles are present. The found calibration curve was then applied to attenuation measurements that were done in a pipe flow setup. However, the resulting volume fractions deviate significantly from the values expected based on the known amount of particles in the flows. This deviation suggests that there are significant differences between these pipe flow experiments and the calibration experiments that cause a difference in the measured attenuation for the same particle loadings. The volume fraction distributions that were found are thus not quantitatively correct, but by comparing them, the accuracy of this method can still be defined. Because the behaviour in the pipe flow is axisymmetric, the radial volume fraction distributions can be found from a single measured projection with the inverse Abel transform. However, the measured attenuation profiles were not symmetric. This means that the resulting radial volume fraction profiles are not actual representations of the real volume fraction distributions. This also means that the current data cannot be used to study particle migration in detail. Nevertheless, the accuracy of the method can be determined by looking a the measured attenuation profiles directly. Even at small path lengths, a difference of 1% in volume fraction was measured successfully. This proves that the proposed experimental method is in theory accurate enough to be used to measure volume fraction distributions in suspension pipe flows. To apply this method successfully, the identified improvements to the experimental setup and processing will need to be implemented. Additional research will be necessary to verify if these improvements are sufficient.

The presence of living organisms in a fluid can affect a fluid’s properties. Organisms propel themselves through the fluid using specialised organelles that generate propulsion. This propulsion creates disturbances in the fluid, leading to variations in its properties. The overarching research question is: ”Can the properties of a fluid be modified with biologically active suspensions?”. Organisms respond to their environment for survival and thereby induce intriguing patterns which enhance flows. By using their natural instincts to react to external stimuli, fluid systems could generate flows or intrinsically modify their properties, which can be controlled by a user’s own will. It is thought that the local orientation of the organism influences the viscosity in a shear flow, because of the added stress induced by the activity of the organisms. This research investigates the influence of the orientation of the stress induced by the green alga Chlamydomonas reinhardtii on the viscosity using light-controlled rheology. This is done through two research questions: ”Is it possible to modulate the swimming orientation of the green algae using a light source?”, followed by the question: ”Can we thereby control the viscosity of the algal sample?”. Phototaxis is the natural response of the algae to swim in the direction of light and was the main asset used to control the swimming orientation. The first research question was explored through three-dimensional microscopy. Images were acquired from algae controlled by light entering at a specific angle of 45°. The second research question com bined light-control with a rheometer that imposed an external shear flow. The rheometer measures the viscosity of light-controlled algae using an on-and-off light cycle. Through the three-dimensional imaging, the position and velocity vectors of each par ticle was found and converted into the relevant spherical coordinates which in turn could present the trajectories of the organisms. These were found to be swimming at the tar geted angle at sufficient light intensity. The second experiment however showed ambiguous viscosity responses that had to be quantified using power analysis. The quantification of the data showed possibility of a viscosity change as a result of the reorientation towards light. The first experiment showed the possibility to control the orientation of the algae using phototaxis and thus positively answering the first research question. The second research question cannot be answered taking strong conclusions. The quantification of the data showed possibility of a viscosity change as a result of the reorientation towards light.

Drag reduction in ships is a key approach to decreasing their energy consumption. Besides saving fuel costs, it also decreases greenhouse gas emissions. One of the most promising techniques for drag reduction in ships is air lubrication, which reduces drag due to friction. Despite its great potential, application of air lubrication in the maritime industry is limited, owing to a lack of understanding of its underlying physical phenomena and stability issues.

The present work aims to experimentally study drag reduction on a flat plate resulting from air lubrication at a representative free-stream velocity of 2.5 m/s over a set of increasing air injection rates covering all the three air layer regimes (BDR, TALDR, ALDR) and assess its dependence on air morphology in terms of the plate’s non-wetted area. The experiments were conducted at the Multi-Phase Flow Tunnel located at the Ship Hydromechanics group of TU Delft.

To achieve this, first, a custom force balance comprising a spring system was designed to facilitate the measurement of the total drag acting on the plate by means of a load cell. Drag measurements in single-phase flow were conducted on a conventional flat plate to obtain a reference dataset and validate the drag measurement system. An uncertainty analysis was performed to quantify the measurement accuracy. Then, single-phase and dual-phase experiments were conducted on another plate of similar dimensions fitted with additional parts necessary for the generation of air lubrication. For the quantification of the plate’s non-wetted area in dual phase flow, these experiments were accompanied by image capture with cameras positioned vertically below the plate. The recorded images were processed using a binary approach to distinguish regions under the plate covered with water and air. Finally, the non-wetted area ratio of the plate was computed based on these processed images. Uncertainty in both drag reduction and the non-wetted area ratio was also determined before
analysing the results.

Results from the present work indicate a positive linear correlation between drag reduction and the associated non-wetted area ratio achieved in BDR and ALDR. However, this correlation varies in slope per air layer regime, possibly due to the physical phenomena governing the regime. Thus, further investigation into the underlying physical phenomena governing each regime could shed light on the reason(s) for different slopes in the
correlation. To improve the current work, estimation of the non-wetted area ratio in the TALDR regime may be carried out to increase the resolution of the correlation found. Moreover, the non-wetted area resulting from the thickness of the air layer may also be studied to define the non-wetted area ratio of the plate more accurately.

In the face of escalating global emissions and rising awareness of the dangers of climate change, there is a need for sustainability initiatives across all industry sectors. Within the shipping industry, this is leading to a resurgence of interest in wind-assisted propulsion systems. This research investigates the hydrodynamic design of a wind-assisted cruise ship, aiming to redefine the image of the historically very polluting cruise industry by designing a vessel aimed at achieving 50% propulsive fuel savings through wind-assisted propulsion. The literature review provides the broad theoretical and computational background necessary for the design evaluation. The description of the historical evolution of wind propulsion, as well as state-of-the- art wind-assisted vessels and emerging concepts, provides the inspiration for the design investigation. A detailed hydrodynamic analysis was conducted, focusing on variations in hull design (B/T ratio, dead- rise angle) and appendage design (skegs, passive and active anti-drift fins). The efficacy of the mod- ifications was evaluated on an operational level using power prediction programs, allowing for perfor- mance assessments in varying conditions. A semi-empirical approach for the prediction of hydrodynamic forces and moments under drift using the maneuvering tool SURSIM is hypothesized as a computationally light alternative to numerical methods. However, SURSIM was found not to have the required fidelity to accurately predict the lift coefficient of the hull at small drift angles. Therefore, a computational fluid dynamics (CFD) approach was adopted for the determination of the hydrodynamic loads. The optimal unappended cruise ship design resulted in fuel savings of 34.5%. Optimal hydrodynamic efficiency favors a hull optimized for minimal resistance, and a retractable, active angle-of-attack fin for improved side force and yaw moment balance. The introduction of the fin led to an increase in fuel savings to 37.3%. Notable improvements in ship maneuverability can also be expected based on significantly reduced rudder actuation in sailing conditions. Operational variations have a large impact on fuel savings. Using the wind statistics of the N. Atlantic Holland America Line increased fuel savings to 48%, compared to the MEPC.1/Circ.896 standard. Con- siderable further gains can be achieved through the efficient handling of surplus wind power. Allowing a variable operational speed between 10.6kn and 20kn instead of a fixed speed of 12kn increased the fuel savings on the N. Atlantic route to 56.7%. Wind-assisted propulsion can be a viable strategy for significant fuel savings beyond 50%, and efficient appendage design can meaningfully improve the ship’s performance. Accurate prediction of fuel sav- ings requires clear knowledge of the operational conditions and control strategies. Further investigation of the detailed appendage design, and comparison of variable speed operation to the employment of a regenerative propeller mode is recommended. This thesis contributes to the evolving discourse on sustainable maritime transport, specifically within the cruise ship sector, by providing a comprehensive analysis of wind-assisted propulsion’s potential benefits.

The development of biofouling on ship hulls is known to be deeply damaging for the overall performance of the vessel. The real effects of soft fouling, especially due to biofilms and diatoms, are however not yet understood and precisely defined. The current work numerically assesses the influence of an increase in surface roughness due to soft biofouling over the hydrodynamic performance of a containership. The simulations are carried out through MARIN’s in-house software ReFRESCO. Unsteady free-surface computations are used to evaluate both the resulting frictional drag, and the wave profile modification for different input surface roughness values. The results show that frictional resistance at 14 knots is increased of up to 45% for a limit equivalent sand-grain roughness height ks = 300 μm, correspondent to biofilms. Residuary resistance at the same speed suffers decreases of maximum 2.5% in the same range. These consist in hull pressure variations in the aft and transom areas of the vessel, and in small height differences in the wave pattern. A dataset obtained from monitoring of the containership for five years is used to study the added frictional resistance evolution due to roughness. Holtrop & Mennen
procedure integrated with environmental variables show a maximum increase of ΔCF = 1.4 · 10−3. The methodology proves useful to identify potential drydocking periods, and to provide an overview
of the added resistance trend. It should however be implemented with Machine Learning models to improve accuracy, in case the precise influence needs to be assessed. A novel comprehensive soft
fouling growth model is created to assess the surface roughness ks and ΔCF time evolution. This constitutes one of the few attempts in the literature to create a model that simultaneously considers the
effect of multiple environmental and operational variables based on the GPS position. Light intensity, sea surface temperature, draft and concentration of nutrients. Increases of up to 100 kN in resistance
and of ΔCF = 0.001 are observed if only diatoms are considered are integrated. Temperature and nutrients result to be more relevant for growth, in comparison to light intensity and draft. A sensitivity
study is also carried out to prove the small dependence on the choice of the analytical approach. A comparison of the growth model + CFD results with data processing show that the analytical procedure
provides a good indication on the growth trend, while added resistance intensity is underestimated in the order of ΔCF = 4.65 · 10−4.

The Landing Platform Dock vessel is mainly designed to support amphibious operations in which landing operations play an important role. For operating landing craft the vessel has a well dock at the stern, designed to station several landing craft. Landing operations are restricted by the motions of the Landing Platform Dock vessel and by the motions of the water in the flooded well dock. Due to turbulent flow in the dock entrance and nonlinear wave motions inside the dock, the (dis)embarking procedure can be very hazardous for landing craft. Therefore it is necessary to execute this operation as safe as possible, meaning that the water motions in the well dock should be investigated thoroughly.
The primary objective of this thesis is to investigate how various physical mechanisms, when combined with wave characteristics, influence the flow within the well dock. The wave profile within the dock arises from the interaction of two physical mechanisms: radiation, caused by ship motions, and diffraction, which is observed when the ship remains stationary, causing incoming waves to diffract around its hull. Specifically, this study wants to determine which of the two physical mechanisms radiation or diffraction has a more pronounced influence. Also, it investigates how different wavelengths impact these mechanisms and the results combination of independently analysed mechanisms with a model that integrates both mechanisms from the start of the simulation.
The second objective centers on the development of an accurate and efficient model to capture the physical mechanisms of the flow within the well dock of an LPD. The validation for this model derives from model tests conducted at TU Delft[20].
The third objective focuses on simplifying the model in order to facilitate the design process. Upon validating the model from the second research question, wave kinematics are captured using a wave probe located at the dock entrance. These kinematics are then used as inputs for a 2D model. Notably, this model specifically represents the well dock and the ramp region, excluding the complete ship structure. For validation, the same procedure employed in the second objective is followed, referencing experiments conducted at TU Delft[20].
All Computational Fluid Dynamics (CFD) calculations were executed using the ComFLOW program.

In this research project, multiphase behaviour in zero gap alkaline water electrolyzers is investigated, focusing on the influence of various parameters including current density, electrode-wall distance and electrode height. This involves a cell containing a vertical electrode pair separated by a membrane submerged in an electrolyte that has a liquid-air interface near the top of the electrodes. The study aims to identify and understand turbulent phenomena through different analysis methods. Additionally, the outcomes derived from the conducted experiments serve as validation data for concurrent modeling research.
The analysis begins by examining vortex formation on the cathodic side, where hydrogen bubbles originate, using Particle Image Velocimetry (PIV). The results indicate the presence of bubble vortices, especially near the top of the channel. At electrode-wall distances of 1.0 cm or more, the bubbles that fail to reach the liquid-air interface tend to circulate extensively within the channel without exhibiting any noticeable discrepancies, such as small vortices. It should be noted that some low-velocity vortices, typically on the order of millimeters per second, can be observed within the bulk of the channel. As a general rule, shorter electrode-wall distances lead to increased vortex formation throughout the entire channel. Additionally, higher current densities induce a greater amount of these vortices or chaotic and irregular motion of the bubbles in the bulk region. In modern configurations, elastic elements, such as materials of thin wires with high porosity and low solid volume fraction, are strategically incorporated into the electrolyzer design to enhance mechanical resilience and accommodate variations in pressure and temperature. These elements serve to ensure the stability and reliable performance of the electrolyzer system under diverse operational conditions. Much of the chaotic behaviour of the bubbles seems to be counteracted when the electrode-wall distance is large (1.5 cm) while an elastic element is placed in between the cathode and opposing wall. When the whole channel is filled by the elastic element (0.6 cm electrode-wall distance), this results in more small-velocity vortices than without the element.
Another examined phenomenon is the presence of a bubble mist. This mist arises when the bubbles that originate at the electrode do not escape the channel at the liquid-air interface, but recirculate towards the opposing wall and downwards from there. Light intensity coming from the bubble clouds is used to gain and calibrate results. Increasing the current density and decreasing the electrode-wall distance lead to deeper bubble mist plumes. Surprisingly, elevated electrolyte height directly influences mist depth, with larger electrolyte heights showing the highest mist depths, even when dividing these by the exposed electrode height. For these measurements, diamond shaped apertures on the electrodes are used. For another series of experiments where higher current densities are examined, slotted apertures are used and the effect of lowering the electrode-wall distance on the bubble mist depth is diminished. Whether this is due to the type of electrodes or the different measurement circumstances (for example lighting) can not be concluded from the results. This emphasizes the need for consistent experimental conditions.
Thirdly, the continuous release and upward movement of gas bubbles originating at the cathode, i.e. the bubble plumes, are investigated. The effect of changing the current density is investigated along different heights of the electrodes. The analysis performed employs three different strategies to assess turbulent behaviour. This includes the PIV method used initially to depict vortices, the light intensity coming from the bubble clouds and visual analysis to find discrepancies like bubble bursts and big bubble trajectories. The PIV method shows capabilities in tracking the plume width and depicting turbulent behaviour. The light intensity analysis however proves challenging due to disparities in lighting. Improved lighting and higher-resolution cameras could yield more reliable results, enabling further investigation into correlations between parameters and turbulence, such as bubble bursts.

A water electrolyzer is an electrochemical device that produces hydrogen and oxygen from water via electricity. A sustainable way of producing hydrogen with renewable electricity, also known as green hydrogen, could enable its use in decarbonisation of sectors that depend on fossil fuels. However, to scale up the production of green hydrogen, larger water electrolyzers are required, which presents challenges. Water electrolyzers produce hydrogen in the form of bubbles, which are transported away through outlet channels. The geometry of the outlet channel is important for performance and safety. A channel that is too wide can result in unwanted ionic current leakage, reducing efficiency and increasing the risk of explosion. To reduce the leakage current, the electrical resistance in the channel can be increased to make it difficult for ions to flow through. However, a narrow channel can increase the hydraulic resistance, leading to clogging due to bubble build-up. This study focuses on the characterization of the slug flow regimes that eventually lead to clogging of the outlet channel in a novel electrolyzer design. To characterize the flow when blockage develops, electrochemical tests with industrially relevant 6M KOH were carried out at room temperature and ambient pressure in conjunction with high-speed video recording. The examined channel was 2 mm high, 3 mm wide, and 40 mm long. The experiments showed that clogging occurs for certain flow regimes. The hydraulic configuration caused clogging at the hydrogen side, which caused the liquid flow prefer to travel via the anode rather than the cathode, which was reflected in the temperature plots. Clogging did not occur when the superficial gas velocity was greater than 0.3 cm/s, which corresponds to a current density of 140 mA/cm2. The clogging was also not visible at a superficial liquid velocity greater than 0.25 cm/s. Below these values, clogging is expected to take place for the investigated channel dimensions. The bubble diameters in the bubble layer in front of the channel were measured with a microscopic camera. The mean bubble diameter was 0.4 mm, which is 80% smaller than the height of the channel through which they flow. However, exceptional bubble sizes that were 3 to 4 times the channel height, which ultimately led to clogging of the channel, were also present. The bubble layer thickness and velocity at increasing liquid flow rates and current densities were measured using a high-speed camera. At low current densities, the velocity and thickness of the bubble layer were not affected by the liquid flow rate. At higher current densities, a higher liquid flow rate appeared to reduce the thickness of the bubble layer, but not the velocity of the bubble layer. The bubble layer thickness as well as the velocity strongly depend on the applied current density (and thus the gas flow velocity). This knowledge might be, when the flow regimes are scaled to larger electrolyzer dimensions accordingly, essential for future electrolyzer manufacturers and operators for the operation and design of larger-scale electrolyzers. It will give insight into which flow regimes should be avoided in order to avoid clogging and thus explosion risks during operation.

The electricity demand increases globally and requires a shift toward renewable sources to prevent the exhaustion of the planet. The shipping industry is responsible for 2-3 % of the global Greenhouse Gas emissions and Heerema Marine Contractors (HMC) identified floating solar as a promising solution to reduce the emissions, of their crane vessel Sleipnir, during operations. This study is the starting point of a technical feasibility study as a temporary energy supply for Sleipnir. The design is strongly focused on the temporary deployment and limited occupied deckspace during transit. Current floating solar systems are commonly designed for permanent deployment as supporting structure for rigid glass photovoltaic (PV) panels. Furthermore, modular constructions are used to limit the transportation costs however they are not designed for temporary deployment. A new PV innovation is flexible lightweight films which allow a more flexible supporting structure. A flexible thin sheet can be spooled on a drum to make temporary deployment possible. The influence of the wave loading on the coupled hydrodynamic behaviour is evaluated since wind and current loading are predictable based on previous research.

The structural design parameters of the thin sheet and drum are designed to mimic the excitation motion since wave structure interaction has been minimized to reduce the mooring force. Therefore, the draft must be low and the characteristic length related to the bending stiffness of the sheet should be smaller than the excitation wavelength. The draft of the drum should be low to have a natural heave frequency higher than the excitation frequency.

The coupled hydrodynamic response for head loading is evaluated with model tests in a towing tank. The concept is scaled according to Froude to ensure the surface waves, which are gravity-driven, are properly scaled. Regular waves are chosen based on the workability wave spectrum of Sleipnir. The roll and heave response over the frequency domain is indicated by analyzing the stable response at certain frequencies. The motions of the drum are obtained with the use of object tracking based on video recordings. The force within the connection of the system was measured with a force transducer whereas the mooring force was measured with a newly developed 3D-sensor.

It turns out that the heave motion of the system mimics the excitation motion over the wavelengths resulting in small drift forces. Significant rotations of the drum were observed for the longer wavelengths leading to water pumping over the sheet. The overturning moment is driven by the dynamic pressure over the drum diameter and the measured force in the connection generates a counteracting moment. The connection force is proportional to the buoyancy required to submerge the sheet and the acceleration of the free-floating sheet.

The feasibility of an OFPV concept for Sleipnir is demonstrated but the rotations have to be reduced by lowering the natural roll frequency. The drum dominates the coupled hydrodynamic behaviour compared to the sheet. Either the dimensions of the drum should be lowered or the thickness of the sheet must be increased. Decreasing the drum diameter is favourable over a thicker sheet since that would increase the characteristic length. Another option is to adjust the geometry of the drum to a shape where increased water displacement is required for the roll motion.

A fairly recent development in the maritime industry is the rising interest in composites, as they have great potential to outperform conventional metallics. They offer good corrosion resistance, fatigue resistance, a low magnetic signature, and a high strength to stiffness ratio. In case of a propeller, the latter may be utilized by adapting the geometry passively to suit the loading condition more optimal. A possible is to mitigate cavitation by utilizing the relatively large deformations when subjected to loads. The hydrodynamic response of a flexible propeller in a flow field can be predicted by the use of Fluid-Structure Interaction (FSI) software, which is currently being developed at MARIN. The project is called ComPropApp, and combines existing fluid and structural solvers. As the ComPropApp is still under development it needed to be verified and validated, which is the main objective of this thesis. As an initial step in the verificatio, a falsification study was applied on the procedures followed in the ComPropApp. This led to the discovery of several errors, to which corrections have been applied to improve the program. With an updated version, computations with a number of different materials were performed to finalize the verification. Then, a model size polyurethane propeller has been manufactured at MARIN to be used in the experimental validation. Experiments were performed in the cavitation tunnel testing facility at MARIN. Here the propeller was tested in several operating conditions, which was then compared with ComPropApp simulations. Lastly, it was investigated whether the application of a composite can reduce cavitation. This was for theoretical research only, since the testing propeller would fail far before reaching cavitating conditions. With the fluid solver, a requirement for twist deformation was set up. Based on these requirements, a range of composite materials was defined, and with it, ComPropApp computations were performed. The resulting displacements and pressure distributions were then compared for rigid and composite propellers. With the presented verification study, it can be concluded that the FSI software is capable of providing realistic computation results. The validation study has led to conclude that the unsteady FSI module is capable of qualitatively predicting the bend deformations in open water conditions. However, due to the large uncertainties arising from the testing material properties and questionable machining quality, the measurements cannot be utilized to define the accuracy of the FSI software. In wakefield conditions the additional uncertainty of the wake velocity distribution meant that these measurements are inconclusive, hence the validation was only performed for open water conditions. The material study with the purpose of mitigating cavitation has shown potential in the application of anisotropic materials. Composites with a specific ply orientation sequence have the possibility of realizing bend twist coupling motions, such that the propeller would unload itself in the vicinity of the ship hull, with reduced cavitation as an expected result.

Convective heat transfer finds applications in several domains of industry like heat exchangers, gas turbine blades, IC engine surfaces etc. The surfaces of these heat transfer applications are either naturally rough owing to manufacturing techniques or become rough over a period of time during operation. These rough surfaces usually contain multiple length scales and exhibit random and heterogenous properties. Fractal roughness, in principal, is characterized by self-similar detail on smaller and smaller length scales and hence fractal dimension which is independent of any length scale can become a very viable option to characterize these multi-scale random surfaces. Therefore in the current thesis, rough surfaces characterized by fractal dimensions were designed and their role in the heat transfer enhancement relative to the pressure drop was studied using DNS.

The fractal surfaces were designed by randomly placing five generations of self-similar cuboids with decreasing sizes from higher to lower generations. The quantity of cuboids sprinkled randomly for each generation followed a fractal dimension. Two principal fractal dimensions were selected, i.e 퐷 = 1 and 퐷 = 2 and eight random realizations of each were generated in order to study the averaged effect of the heat transfer performance. Since the cuboids were randomly placed, different realizations within
the same fractal dimension experienced varied sheltering effects by larger generation cuboids. This ultimately produced a fluctuation in the ”sheltered” solidity of the fractal surface for different realizations whose effect was also studied. The cuboids were resolved in the simulation using an immersed boundary method.

To quantify the heat transfer performance of the rough surfaces, two performance factors namely the aero-thermal efficiency and Reynolds analogy factor were defined. It was found that the two types of fractal surfaces performed approximately similarly with a slight higher mean for 퐷 = 2 based on the above heat performance parameters even though the two surfaces looked very different. However, the above performance factors showed a very strong correlation with the ”sheltered” solidity of the fluctuating realizations displaying an increasing trend.