Today's electricity supply falls short of current demands, leading to the utilization of gas turbines in both ground based and avionic infrastructures. Nevertheless, these often rely on carbon-based fuels, resulting in escalating CO2 emissions. However, adopting hydrogen as a fuel eliminates carbon emissions. Aside from zero carbon emissions, hydrogen has a higher energy density by weight compared to conventional fuels. This makes it an distinct option for applications requiring efficient energy storage and delivery. Due to its wide flammability range and low ignition energy, hydrogen can combust in scenarios where traditional fuels might not. This unique characteristic, while advantageous in certain contexts, requires detailed study to ensure safe and efficient combustion in gas turbines. However, the combustion of hydrogen inherently results in elevated flame temperatures, thereby generating increased NOx levels. Furthermore, hydrogen's high mass diffusivity translates to a reduced Lewis number. Consequently, it becomes vital to grasp the local dynamic characteristics of the flames, particularly at stable and flashback points. Understanding the thermo-physical behavior of hydrogen flames, especially at stable and flashback points, therefore, requires experimental studies to reflect real life complexities. These tools can offer insights into turbulence-flame interactions, flame stabilization, and emission formation mechanisms.


In this research, an exploration was undertaken to understand the local kinematics and dynamics of Hydrogen and DNG flames, with emphasis on stable states and those approaching flashback conditions. Particle Image Velocimetry (PIV) experiments were employed on a Bunsen burner setup, facilitating the capture of the jet flames' velocity fields. Both low-speed and high-speed recordings were captured by high-speed camera, providing distinct insights into flame dynamics. Flame front detection was achieved using Mie-scattering, capitalizing on the differential seeding particle densities between the unburnt and burnt regions. Intensity differences between these regions were meticulously captured with a bilateral filter, leading to the successful extraction of the flame front. This extracted front was subsequently distinguished via segmentation and superimposed onto the velocity field. Low-speed recordings offered a generalized perspective on flame turbulence characteristics through cold flow validation, while high-speed recordings unveiled specific dynamics, inclusive of flame curvature, local flame and displacement speeds, and both normal and tangential velocities and stretches. It consistently holds the 1-D unstretched flame speed, even as the Reynolds number increases, aligning with the respective flashback points and stable conditions of the flames. As a result, in-depth comparison of DNG and $H_2$ fuels in terms of flame dynamics and kinematics were discerned.

With ongoing research towards clean combustion, hydrogen has been identified as a potential alternative to natural gas fuel, for example in power generation sectors utilizing gas turbines because of their inherent nature of being a carbon-free energy carrier. However, it is crucial to clarify that the ultimate goal is not just carbon-free combustion but clean combustion, which entails addressing other post-combustion emissions, such as NOx (nitric oxide) emissions. To meet stringent NOx emissions regulations, gas turbine fuels are combusted under premixed conditions. But in these premixed conditions, it has been observed that flames produced as a result of combustion have a tendency to flashback, which means that the flame travels back into the premixing chamber causing severe structural damages. This is particularly concerning when using hydrogen, which due to its high reactivity and flame speed is more prone to flashback than natural gas. To understand this high propensity of hydrogen to flashback, there is a strong requirement to examine the topology or structure of flame which can be obtained using optical combustion diagnostics technique.

Considering a flame has a three-dimensional structure, in this thesis, an optical combustion diagnostics technique of Computed Tomography of chemiluminescence (CTC) was applied to Bunsen burner flames using six CCD cameras. The cameras were arranged around the flame and a tomographic algorithm was used to reconstruct the three-dimensional structure of the flame. The technique was applied to turbulent 100% by volume fraction Dutch Natural Gas (DNG) flames at various Reynolds numbers representing stable, close to flashback, and flashback case. The reconstructed DNG flame results highlighted the capability of the CTC technique to offer valuable insights into the intricate features of the flame. Furthermore, these results not only indicated the possible location of the origin of the flashback within the structure of flame but also revealed specific features associated with events prior to the flashback. recognizing the potential of this technique, it was subsequently applied to a turbulent flame consisting of 50% hydrogen blended with 50% DNG. The reconstruction results offered insights into the fundamental structural differences between a flame consisting of 50% hydrogen and 50% DNG and a pure 100% DNG flame. The conclusions drawn from the visual assessments of the reconstruction results were further supported by the subsequent statistical analysis and the resulting cone angle values.

A 1:5 scaled rowing boat had been designed by a previous research group to determine the performance of rowing blades with different sizes and blade angles. In this research modifications were done to allow more control of the rowing motion. The aim of this study is to assess whether an angled oar blade can reach faster speeds with the same input power. This is done by collecting data with force and position sensors. With the collected data, the input power and speed of the rowing boat can be compared between several modified oar blades. The results of the tests in the towing tank show that it is very likely that the rowing boat can go faster with the same input power by adjusting the oar blade angle.

The growing energy demand and climate change poses a need for alternative energy generation in terms of renewable resources. Renewable energy resources are characterised by their intermittent behaviour. A backup power supply is required that can deliver electricity when the supply from the renewables is not sufficient. Gas turbines operating with hydrogen is an attractive option, since hydrogen combustion has zero carbon emissions and can be used as an energy storage when the supply of renewable energy sources are abundant. However, hydrogen combustion poses several challenges. A hydrogen flame has a higher flame temperature than natural gas, leading to more NOe production. To
reduce these NOe emissions, gas turbines operate in lean premixed conditions. This creates a risk of flame flashback, in particular boundary layer flashback, which can lead to severe damage to the gas turbine.
Recent research performed on boundary layer flashback revealed that two flame configurations, i.e. unconfined and confined, showed fundamentally different flashback phenomena. Unconfined flame flashback refers to the situation where an initially stable flame anchored at the burner rim eventually moves into burner tube. When a flame is partially of completely surrounded by walls and propagates and then starts propagation along the wall, it is called confined flame flashback. Previous studies have focused on one of the two flashback processes at a time. However, the transient flashback process between unconfined and confined flame flashback is not well understood. Research has shown that hydrogen is much more prone to flashback compared to natural gas. This has been attributed to the difference in flame speed between natural gas and hydrogen, but the exact reason for the difference in flashback behaviour between natural gas flames and hydrogen flames has yet to be found.
In this study, a quartz Bunsen burner is used to investigate the flashback phenomena of turbulent premixed hydrogen-natural gas-air flames. To gain more insight in the flashback phenomena, three experiments have been performed. First, flashback maps are obtained to determine the flashback limits of the quartz Bunsen burner. Secondly, the influence of a flame on the flow was investigated using turbulent statistics. Finally, both unconfined and confined flashback are visualised, thereby capturing the transient flashback process between these two configurations. This has been done for a stoichiometric natural gas flame and a lean hydrogen flame. Laser diagnostics like Particle Image Velocimetry (PIV) and Miescattering are used to obtain the turbulent flow statistics and to visualize the instantaneous flashback process. The results show that regions with negative velocity fluctuations in the unburned mixture are the predominant physical mechanism for the start of unconfined flashback and for the transient flashback process of the flame propagating into the burner. However, the start of a flashback event depends on the combination of several parameters: the bulk velocity, the position of the flame front before it interacts with a region with negative velocity fluctuations, the magnitude of the negative velocity fluctuations and whether a region with positive velocity fluctuations is absent after interaction of the flame with the region consisting of negative fluctuations. So, unconfined flashback is rather a statistical phenomena, where the chance of the occurrence of a flashback event is increasing for a decreasing bulk flow velocity. Experiments showed that the transient process between unconfined and confined flashback is very short and fast. After a distance of approximately 5 mm upstream of the burner rim, a backflow region starts to develop in front of the flame, which denotes the start of confined flashback. The time needed for a natural gas flame propagating upstream from the burner rim to reach a confined configuration is approximately 17 ms and only 5.6 ms for the hydrogen flame. The suggested physical mechanisms leading from unconfined flashback to confined flashback are the convex shape of the flame towards the reactants during upstream flame propagation and the reduced cross-sectional flow area of the burned gases at the flame tip. Due to the created backflow in front of the flame and the abovementioned mechanisms, the upstream flame propagation is strongly enhanced, which explains why the flashback propensity for confined flames is much higher than for unconfined flames. The experiments showed that the hydrogen flame propagates closer to the wall than the natural gas flame, indicating higher backpressure effects. The hydrogen flame is thermaldiffusive unstable and the convex shape of the flame tip during flashback strongly enhances the local flame speed and thus the upstream propagation velocity. In contrast, the hydrodynamic instability encountered in the natural gas flame only retards the flow in front of the flame tip, but does not affect the flame speed. This might explain the difference in flashback behaviour between natural gas and hydrogen flames.

Due to the growing energy demand and global warming renewable energy is currently of interest. To deal with the production intermittency of renewable energy, energy might be stored in hydrogen. Hydrogen can be converted into electricity with a fuel cell or by combusting hydrogen in a gas turbine. Gas turbines are more suitable for large power outputs. Large power outputs require large volumetric flow rate through the gas turbine. The large volumetric flow rate causes high gas velocities and results in a turbulent flow. Using hydrogen instead of natural gas as energy resource changes the stable operating regime and increases flame flashback propensity. Current studies done at the Technical University of Munich and the Technical University of Delft suggest flame confinement, burner wall temperature and velocity fluctuations are vital to the flashback process. Des-pite the recent studies flashback is yet not completely understood. \\ This research focuses on premixed turbulent hydrogen and hydrocarbon combustion to further investigate the possibilities of using hydrogen in a gas turbine. The aim of this research is to gain more insight in the flashback process by varying velocity fluctuations while keeping the bulk velocity constant. The first objective is to gain insight in the relation between bulk velocity at flashback and velocity fluctuations. The second objective is to gain insight in the relation between bulk velocity at flashback and low velocity streaks upstream the flame front. The third research objective is to gain insight in the relation between flame front, boundary layer obstruction and bulk velocity at flashback. \\ All experiments were conducted with a Bunsen burner. A screw, acting as boundary layer obstruction, was placed below the burner rim to locally disturb the flow. The influence of the local disturbance on bulk velocity at flashback was investigated by mapping flashback propensity and by examining non-reacting and reacting pipe flow with planar Particle Image Velocimetry (PIV). The fuel compositions studied in this research are DNG, hydrogen and mixtures containing 20, 40, 60 or 80 volumetric percentage $H_2$. \\ The flashback maps show that bulk velocity at flashback increases with obstruction height for all gas mixtures. The obstruction height is the height from burner wall to screw tip. The results obtained through the non-reacting field experiment confirmed that the flow was fully-developed and turbulent. Furthermore, the results showed that obstruction height locally increased velocity fluctuations. The average velocity downstream the obstruction slightly decreased with increasing obstruction height. The reacting field was examined during flashback and during stable operation to relate flashback propensity and time-averaged flow characteristics with statistical analysis. The results showed flame front axial distance from the burner rim decreases with increasing obstruction height, indicating that a flame produced by a burner with boundary layer obstruction is more prone to flashback. The flow characteristics just upstream the flame front were analyzed because once the conditions upstream the flame front are suitable, flashback occurs. \\ A flashback prediction model developed at the TU Munich was modified to account for the obstruction height. The modelled results indicate that, in an unconfined set-up, flashback might be described with time-averaged flow characteristics and that the turbulent burning velocity is strongly correlated to the bulk velocity at flashback. Histograms of the instantaneous velocity showed that obstruction height has no influence on low velocity streaks and thus it seems low velocity streaks are not dominant in the flashback process of premixed unconfined flames. The absence of low velocity streaks for different obstruction heights indicates that flashback of unconfined flames is a different process than flashback of confined flames.

Hydrogen combustion in gas turbines could play an important role in the future energy transition. However, the design of flexible gas turbine combustors able to operate with a wide range of Dutch Natural Gas (DNG) and hydrogen fuels is accompanied by new complex challenges. Small quenching distance, high burning velocity and propensity to develop instabilities at leaner conditions makes hydrogen-rich fuels particularly prone to boundary layer flashback (BLF). Both academia and industry are currently involved in developing a better understanding of the BLF phenomenon, so that new generation combustors can be designed. From recent experimental investigations carried out by the Technical University of Munich (TUM) research group, it has been noted that the configuration of the flame plays a crucial role in flashback propensity. In confined geometry, where the flame is partially or completely surrounded by walls, the boundary layer flashback propensity is much higher. Indeed, in the confined configuration the flame-flow interaction effects are very strong. However, even in the unconfined configuration it is not completely clear how the flame-flow interaction affects the boundary layer flashback onset.
In the present work, tube burners have been used to investigate different hydrogen/DNG turbulent flames. Particular attention was paid to the lean hydrogen premixed flames relevant for gas turbines combustors. In the first part of the investigation, flame regime maps have been used to characterize flashback propensity of different hydrogen/DNG mixtures at different equivalence ratios. Furthermore, the effect of tip temperature has been investigated by comparing flashback onset in both cooled and uncooled conditions. Particle image velocimetry (PIV) and Mie-scattering measurements have been used, both to obtain useful statistical data and to visualize the flashback transient phenomenon. Indeed, the lean hydrogen flame and DNG flame behaviour during flashback have been visualized and compared.
The results highlight that, the flame-flow interaction plays an important role. The interaction is related to the hydrodynamic or Darrius Landau instability which causes the presence of an adverse pressure gradient just downstream of the flame front. The adverse pressure gradient leads to a slowdown of the flow, which allows the flame front to propagate upstream. The coupling of this interaction with the velocity fluctuations of the turbulent field leads to flashback. The flashback location is outside of the viscus sublayer, but still in the proximity of the wall, where the velocity fluctuations are stronger and the distance between the average flame front and the burner exit is small. The distance between the burner tube outlet and the flame front directly affects the strength of the adverse pressure gradient on the approaching flow. The closer the flame front is to the burner exit, the more the incoming flow is deflected and retarded by the burner walls. In these experiments the most significant difference between the DNG and hydrogen flame is not qualitative but quantitative. The mechanism itself is qualitatively the same, however, hydrogen flame flashback occurs much more abruptly. This is related to the response of hydrogen flame speed to stretch. Indeed, since the Lewis number is lower than one, the lean hydrogen flame speed increases with stretch leading to thermal-diffusive instability.

Heat pipes are typically used in the semiconductor industry. This means that the scale of these heat pipes is typically in the order of centimeters. Zijm suggests that heat pipes could be used for geothermal applications, but literature is lacking. To further investigate the geothermal application of heat pipes, a large scale heat pipe is built. This thesis gives an insight in the typical design challenges that one faces when constructing a heat pipe of this scale. The heat pipe that is constructed can support a heat flow of 10 kW. The heat pipe is constructed from mainly 54 millimeter copper and glass pipes. The evaporator section is 1.5 meters long and facilitates controlled electric heating. Then follows a 4 meter long adiabatic section. The condenser section is 2.5 meters long. The total length of the heat pipe is 9 meters. The individual sections are held together by EPDM connectors. The dimensions of the heat pipe are compared to the operational limits posed by the Engineering Sciences Data Unit. Then the thermal resistances of the heat pipe sections are calculated and afterwards validated. The interfacial thermal resistance between the electric heaters and evaporator wall was reduced by applying thermal conduction paste to the band heaters. It was found that at coolant flows of 1000 l/h and higher, the vapour temperature in the heat pipe drops significantly. The drop in temperature facilitates a higher heat flow through the heat pipe. Also, the resistance across the evaporator and the condenser section gets smaller for higher coolant flows. The results found are supported by theory and formulae from the Engineering Sciences Data Unit.

The turbulent lifted Dutch natural gas jet diffusion flame in cold co-flowing air is widely used in gas turbine operation and subject of the current investigation. In striving to clean and sustainable combustion, adding hydrogen to the Dutch natural gas (DNG) fuel and CO2 in the coflow is promising. Objective of this investigation is to study the effect of hydrogen fuel-addition and CO2 coflow-dilution on the stability of a turbulent lifted DNG jet diffusion flame in cold co-flowing air. OH* chemiluminescence and particle image velocimetry (PIV) are applied simultaneously at a low sampling frequency of 50 Hz to determine time-averaged statistics of the lift-off height and stabilization point location of the flame. PIV measurements at a sampling frequency of 2.5 kHz are performed to determine the transient behaviour of the stabilization point, the flow conditions at the stabilization point and the burning velocity. Hydrogen fuel-addition leads to increased flame stability. With more hydrogen in the fuel, the lift-off height reduces and the stabilization point location shifts upstream and radially inward. The burning velocity at the stabilization point increases with increasing hydrogen fuel-concentration. A significant difference is observed in the burning velocity conditioned on upstream or downstream stabilization point motion. Hydrogen enables the flame to stabilize in a region of the flow that is characterized by high gas velocity and high vorticity. CO2 dilution of the coflow leads to decreased flame stability. With increasing CO2 coflow-dilution, the lift-off height increases and the stabilization point position shifts downstream and radially outward. The burning velocity at the stabilization point increases with increasing CO2 coflow-dilution and the flame stabilizes in a flow region with significant lower vorticity.
Additionally, this report provides useful statistics of the investigated quantities and presents a description for the transient behaviour of the stabilization point.

Rowing is a competitive sport where victory is determined by fine margins. In the past, the focus has been on optimizing the shell. Due to the increasing regulations placed on the shell design, the scope for manoeuvring in this area is greatly limited. This increases the necessity to focus on the less explored option of propulsion in rowing. The propulsion is caused by the momentum transfer from the rowers to the water with the help of oar blades. The design of these oar blades has also been extensively studied. However, knowledge on the effect of the air-water interface on the drag force on the oar blades is still lacking. Therefore, visualizing the flow around the oar blades will lead to better understanding of the flow structures, which would help in optimizing the drag force acting on the oar blades.

In the present study, a simplified scenario of a rectangular flat plate moving normally to its plane along a straight line and parallel to the free surface has been studied using hydrogen bubble visualization and Particle Image Velocimetry (PIV). Using a 4-axes industrial robot and a force/torque transducer, the effect of the air-water interface on the drag force on the plate has been studied at a Reynolds number, based on the longest edge of the plate, of 6×10⁴. The analysis of the drag force profiles at different plate depths led to the identification of a high drag case, which occurred at a plate depth h of 20 mm. The hydrogen bubble visualization indicated that in this specific case, the high drag was related to the formation of a compact wake behind the plate.

Hydrogen bubble visualizations were also performed at two other plate depths, h = 0 and h = 100 mm, to determine the effect of the air-water interface on the flow structures and ultimately on the drag force acting on the plate. The visualization of the deep water case (h = 100 mm) captured the formation of a vortex ring, which was found to have unique effects on the drag force. The presence of the air-water interface was found to significantly influence the drag force on the plate. Additionally, an extensive decomposition of the drag force profile showed a time-dependent added mass force and a decaying force acting on the plate.

Particle Image Velocimetry (PIV) measurements were carried out in the horizontal mid-plane of the plate. The tracking of the starting vortex and its disintegration has been identified using the swirling strength analysis. Finally, combining the hydrogen bubble visualization and PIV results, the formation and the development of the starting vortex was identified.

Impinging jets are relevant ow configurations in many technological developments. For example, on some short take-off and landing aircraft the high speed exhaust from the jet engine is deflected by direct impingement on the aps to create extra lift during take-off. Fatigue due to excessive dynamic loading on the aps and high levels of noise radiation are among the problems encountered in such designs. Additionally, such flow-structure interaction is a good model for cooling of turbine blades, annealing of plastic and metal sheets, deicing of aircraft systems etc.. Jets are easy to simulate and contain all the constituents necessary for the study of shear flows. The shear-layer instability at the nozzle edge develops into axisymmetric toroidal vortices which magnifies in size and strength downstream of the nozzle. The interaction of these vortices with the solid structures induces pressure fluctuations that manifests in the form of noise in the far-field region. Hence, it is also a benchmark case for studying vortex-structure interaction noise.