The design of high-performing fluid and thermal devices is crucial for many aerospace
applications such as heat exchangers and flow manifolds. These systems often operate
under transient conditions, adding another layer of complexity to their design. Conventional
design principles have limitations as they depend on engineers to propose the
principal structure. In this dissertation, topology optimization (TO) is investigated as a method to design transient flow and thermal devices. To date, the use of TO for transient flow or thermal problems remains limited to experts in the field. Performing successful optimization heavily relies on the tuning of model and optimization parameters. By systematically investigating and improving the algorithms, their parameters and their characteristics, this thesis provides engineers with guidelines for their use.

This paper presents the design of a lab-scale silicon nitride photonic crystal lightsail, demonstrating the LightSailSim software package. LightSailSim, an open-source analysis pipeline, integrates a custom mechanical solver based on a particle system model with optical simulation results. The resulting design balances dynamic and structural stability with propulsive performance. It consists of an optimised photonic crystal that provides a suitable trade-off between stability and propulsive force. The sail is suspended at its edges by a circular ring made of silicon, providing sufficient boundary tension to prevent sail deformation-induced instability. The design is made such that it can be produced from a single silicon wafer and levitated using a single 400 W laser.

Soft robots, characterised by compliant mechanisms (CMs) made from low-stiffness materials, offer improved adaptability compared to traditional, rigid robots. These CMs are often actuated by pressure loads. Moreover, soft robots provide new possibilities in the area of robotics. They can be used in search and rescue and interact safely in collaboration with humans.

The current state-of-the-art in Topology Optimisation (TO) for design-dependent pressure-actuated CMs (PACMs) relies heavily on linear models. The determination of design-dependent pressure loads involves employing the Darcy method, which integrates Darcy's law with the drainage term to obtain the pressure field. Subsequently, the finite element method (FEM) is used to transform the pressure field into consistent nodal forces. However, it is crucial to acknowledge that these linear models are only valid for small displacements.

This thesis introduces a novel approach by incorporating nonlinearities into the solid mechanics of the TO process for PACMs in conjunction with the Darcy method. Additionally, this work incorporates nonlinearities into the solid mechanics of the TO process for PA multi-material compliant mechanisms, presenting another novel method.

Four nonlinearities in the solid mechanics of PA soft robots may occur, two of which are addressed in this thesis: geometric nonlinearities and a hyperelastic material model. Geometric nonlinearities arise from large deformations caused by high applied pressures. The Neo-Hookean material model is implemented to describe the low-stiffness material accurately.

The TO of pressure-actuated (PA) soft robots is simulated using COMSOL, a commercial software program for multi-physics simulation. This research presents a detailed comparison between theoretical predictions and practical outcomes as realised in COMSOL. Furthermore, this thesis includes a case study validating the successful implementation of the new method, covering a PA inverter, a PA compliant gripper, a PA member of the Pneumatic Networks, and a PA multi-material compliant gripper. The obtained results indicate limitations on the allowable applied pressure loads for the mechanisms, specifically in the case of the PA member of the Pneumatic Networks and a PA multi-material-compliant gripper. However, the PA inverter and PA compliant gripper validate the expectation that incorporating a hyperelastic material model yields significantly different results than the linear elastic material model. Moreover, the TO with the hyperelastic material model can predict displacements more accurately than the linear TO, as the differences between the displacements obtained from the TO and the analysis align more closely.

The Wang method is investigated to observe its influence on the range of the applied pressure loads during the TO of PA soft robots. The Wang method employs an interpolation technique that interpolates between linear and nonlinear theories. In this approach, void elements are described using linear theory, while solid elements are characterised by nonlinear theory. This interpolation method is developed to address distorted elements during large displacements. It effectively extended the range of applied loads during the TO of structures. However, it is found that this method does not influence the range of the applied load during the TO of CMs.

Granular materials are all around and have many secrets that still need to be unravelled. This thesis shows how the stochastic behaviour of granular materials can be identified and efficiently included in design of bulk handling equipment. To achieve this, metamodels play a key role as they are able to capture trends of physical models and give fast predictions. In this research it is shown what the opportunities and limitations of metamodels are in a hopper case study based on simulation data from a discrete element model (DEM). Next a procedure is presented in which the stochastic behaviour and metamodels are combined to calibrate a DEM model including the stochastic behaviour of the granular material. The stochastically calibrated DEM model is accurate and used as the basis for a design optimization case study in which the effect of robust optimization was evaluated and validated. These studies combined show that stochastic behaviour of granular material can be included in design of bulk handling equipment and lead to improved and robust designs.

Given the growing number of environmental and societal concerns we confront today, the idea of sustainability has gained importance. At the same time, new strategies for improving the performance of structures and systems have been developed due to developments in engineering and computational design. This research aims to generate a sustainable design using topology optimization by focusing on design for disassembly. One advantage of design for disassembly is that when a product can be disassembled, the parts can be reused, repaired, recycled, and remanufacture. This facilitates other aspects of product sustainability, such as the product's life cycle and end-of-life. A structure is divided into two parts and attached by a connection point, this connection point is called the connector. Due to sustainability, the connection method needs to be a non-destructive method, which in this case is the bolts. Next to the connector, two voids are required to insert, tighten and remove the bolts. Therefore, in this research, a structure is optimized using topology optimization and simultaneously optimizing the position of cut lines and connectors. The approach taken uses level set functions to model the cut of the structure, as well as the connectors and the voids. Then, they are converted into a density field using a smoothed Heaviside function. A Solid Isotropic Material with Penalization (SIMP) motivated method is used to join all the different density fields into an equation for the interpolated elasticity modulus. The optimization aims to minimize compliance with volume and no-overlap constraints. The non-overlap constraint is applied to the connectors.
The structure and the position of the cut line and the connectors are optimized using the Method of Moving Asymptotes (MMA) method. A gradient based sensitivity analysis is used in the MMA. Afterwards, the influence of the cut line, the connector and the voids are observed individually. After optimizing the parts individually, the full optimization was performed, where the structure, the cut lines and the connectors with the voids were optimized. Furthermore, a parameter study was done to observe their influence on the final layout. The optimizer's behaviour was observed by looking at the optimization results and the parameter study. For example, how the optimizer tends to stack some connectors together to create a member of the structure or the influence of the voids.

With the approach presented, the main idea of optimizing a structure using topology optimization and simultaneously dividing it and optimizing the connector's position is obtained. However, the optimization has some limitations, as some assumptions and design considerations are not accurate, further research is needed to get accurate results.

In the engineering industry, all structural parts have to be designed as efficient and lightweight as possible. Traditionally, the design process has been carried out through manual design iterations, which can be time-consuming and require significant engineering expertise. Over the last decades however, several computational design techniques like Topology Optimization and Generative Design have been developed to support engineers in the structural part design process. Even though these techniques can have a positive influence on the design process, they both also have their downsides. Topology Optimization only gives a single result that is often a local optimum, influenced by boundary conditions and numerical settings. Commercial Generative Design tools explore multiple design options in a single run using AI algorithms, but need cloud-based systems to carry out their demanding simulations which still take several hours per run. It is however expected that a combination of the two, a Topology Optimization based Generative Design approach in the form of an auxiliary tool, has potential to improve the early stages of the design process even more. With such a design approach, multiple design solutions are explored quickly to study the effect of boundary conditions or numerical settings. This can help designers by giving direction and insight in trade-offs between multiple objectives, early on in the design process when design decisions still have the highest impact.

The goal for this research was therefore to research the effect of such a Topology Optimization based Generative Design approach on the design performance and experience. In order to do so, a robust and user-friendly TOP-GD tool was created. In this tool, multiple design solutions are explored quickly by implementing a batch-run setup that varies several chosen parameters, without needing to manually run several optimizations consecutively. Calculations are done with a simple TO script using coarse geometries, and without taking into account manufacturing methods yet. This asks for less demanding, detailed and complicated calculations than AI-based Generative Design tools currently offer, while at the same time moving from a single TO result to generating a range of candidate solutions. A lot of effort was put in the user-friendliness of the TOP-GD tool, enabling an easy workflow for the setup of design problems and a clear presentation of the results by means of a simple GUI.

The use of the TOP-GD tool in the design process was evaluated in an experiment, where it was compared with a more simple TO tool and a basic manual design approach using just pen and paper. This was done by giving the participants of the experiments three simple design assignments, that they had to carry out using each of the design approaches one by one. Evaluation of the approaches was done by comparing the design performance, and assessing the design experience with a survey and using Eye-tracking techniques.

The results of this experiment did not show enough evidence to conclude that the different design approaches had an effect on the design performance for the simple assignments executed during the experiment. However, the results of the survey show a clear positive impact of both the TO tools on the design experience, compared to manually designing. Furthermore, the TOP-GD tool has the largest positive impact on the design experience and its use in the design process is considered a big improvement, especially in quickly exploring new design directions and creating overview. This confirms the expectation that a Topology Optimization based Generative Design approach has a positive effect on the early stages of the design process. The differences found with Eye-tracking between the TO tools support this, although a more extensive experiment should be done to convincingly confirm this conclusion.

Offshore wind turbines have seen a significant increase in size over the past decades. A critical consequence of this increase in size is a substantial increase in the torque transferred from the wind turbine rotor. This leads to the necessity of larger and heavier drivetrains to adhere to structural failure requirements and subsequently to larger and heavier tower and support structures increasing the total mass and cost of the turbine. Hence, reducing the mass-to-torque ratio of the drivetrain has become a key design challenge.

The company Delft Offshore Turbines (DOT) proposes to replace the conventional drivetrain in the top of the turbine with a hydraulic drivetrain, which has a lower mass-to-torque ratio. Although this concept shows promise, finding a low mass design that fulfils the infinite fatigue life requirement can be challenging. Using topology optimization to minimize mass while constraining the structural requirements could, therefore, be instrumental in the realization of this concept.

The design case by DOT can be classified as rotating machinery. In general rotating machinery is
commonly subjected to periodic loading that varies non-proportionally in time. This results in fluctuating stresses causing material fatigue. A consequence of the non-proportionality of loading is that the time response needs to be computed to evaluate for fatigue, which adds additional computation cost. However, another common aspect found in rotating machinery is that parts are cyclic symmetric, which allows for a potential reduction in computation cost.

In this thesis a method is presented to implement infinite fatigue life constraints into density based topology optimization for structures subjected to non-proportional loading, while cyclic symmetric properties are exploited to reduce computation cost. It was found that when the load case on a cyclic symmetric part adheres to certain conditions, a single static FE-analysis can provide multiple time steps for a quasi-static analysis. Decreasing the computational burden roughly proportional to the unique number of time steps obtained. The largest local variations in stress are estimated using a smooth min/max function and aggregated into a global constraint.

The method was tested on several numerical problems as well as applied to the DOT design case. The results showed that the method was able to properly constrain a global fatigue constraint while minimizing mass, achieving final designs that might not be trivial to find by hand.

Silicon based power semiconductors have long been used as the standard in ‘semiconductor technology in power conversion applications’. Recent developments replaces the Silicon with Silicon Carbide as it results in superior performance of the power conversion applications. However, due to the increased performance, challenges regarding heat dissipation emerge and the lifetime of the power semiconductor packaging or power module is compromised. Since this leads to an increase power density, the cooling of the power module is becoming of more importance and the heat sink becomes an interesting component to optimize. The best performance of a heat sink can be obtained when the flow through the device is turbulent. Developing turbulent flow heat sinks by using topology optimization methods can significantly improve the cooling performance compared to the current designs. This work is thus aimed towards improving methods for topology optimization of turbulent flow cooling devices. However, this work focuses on turbulent flow topology optimization only and aims to improve the accuracy of current methods. It is important that the flow physics are accurate since the thermal energy transfer is dependent on the flow field. The current state-of-the-art method based on the 𝑘−𝜔 turbulence model developed by Dilgen et al. is investigated. A design domain is subdivided into elements since the finite element method (FEM) is used, such that an optimization algorithm is able to turn every element into either fluid or solid with the goal of finding the best performing structure. This density based approach, models the solid domain as a highly impermeably porous material. To inhibit flow in the solid domain a Darcy penalization is added to the momentum equation. Moreover, in the method by Dilgen et al. boundary conditions in the other turbulent fields are also enforced using a similar penalization approach. Weaknesses and errors in the density based method are investigated by comparing solutions to ones computed on a body fitted mesh. It has been found that the largest errors in the solution, by using the state-of-the-art method, appear at the solid/fluid interface in the design. In these regions the penalizations are not applied correctly for the desired boundary conditions. Therefore, in this work it is improved on by the enforcement of the boundary condition by using the Dilation method. The Dilation method focuses on the solid/fluid region where it shifts the boundary conditions for the specific dissipation rate (𝜔) and ensures it reaches the desired value at the solid/fluid interface. Secondly, severe flow leakage is found in the “porous” solid domains using the state-of-the-art method. Flow leakage is reduced by using an improved formulation of the maximum Darcy penalization in the solid domain. Finally, the improved approach is investigated in several topology optimization cases and compared to the state-of-the-art Dilgen method. It is shown that by using the new approach, different designs with a better accuracy can be obtained. In an extreme test case, the Dilgen method resulted in an infeasible design which disconnects the flow inlets from the outlets while the new and improved method resulted in a feasible design.

The recent success of - and demand for - compliant mechanisms has increased rapidly within the micro-electromechanical systems-, aircraft-, spacecraft-, surgical-, and precision-instrument industries. Yet, even greater success may be achieved by overcoming the computational cost and instability of the mechanisms' design methodology. This involves large-scale topology optimization, geometrically non-linear structural analysis, and particularly integrating the latter into the former. Therefore, a novel framework is proposed that extends the powerful design freedom of topology optimization with most of the geometrically non-linear qualities, without as much of the computational ramifications. Utilizing a Bayesian-enhanced perturbed analysis, the equilibrium curve is locally approximated by an asymptotic expansion, satisfying the curve's higher-order geometric derivatives at the undeformed state. Each of the latter is recursively and efficiently obtained through a linear solve with the same, Cholesky-factorized stiffness matrix. Furthermore, a tensor reformulation and -decomposition of the four-noded bilinear element's Green-Lagrange strain energy model are successfully exploited. A tight error estimator is derived to govern the Bayesian analysis and balance approximation efficiency and accuracy during optimization. Design-sensitivities of this error-estimator and other design-dependent responses are analytically obtained through the adjoint method, and applied in a few classical density-based topology optimizations; While Bayesian-enhanced perturbed analysis required noticeably less computational effort compared to Newton-Raphson analysis under mildly non-linear conditions, it often resulted in practically identical performance improvements compared to linear analysis.

By 2030, the global semiconductor industry is projected to hit the valuation of becoming a trillion dollar industry. Advancements in electronic devices have gave rise to tougher requirements thereby requiring the manufacturers to push the limits of consistency. This has lead to a need to enhance the accuracy of the chip manufacturing process. Wafer flatness is one of the primary prerequisites for attaining high accuracy in integrated circuit manufacturing. The high accuracy is achieved by incorporating optical lithography in the development of circuits with small feature sizes. The current lithography machine requires precise estimation of wafer flatness on a sub-nanometer scale for the elimination of deviations between the exposed region and the plane of focus. In this thesis, the point of focus is on the short stroke positioning system of a lithography machine. The critical aspect of flatness is interpreted by performing scanning operations on a wafer table surface. These operations provide an estimation of the types of deformations the wafer is subjected to. The wafer table acts as a wafer carrier, which undergoes lithography operations. Therefore, it becomes very critical to examine the nature of the wafer table. After performing scans for surface deformations, an optimization problem is formulated which involves placing piezoelectric actuators under the wafer table. The role of the piezoelectric actuators is to correct the wafer table thereby enhancing its flatness property. Through this research, the optimization yields to give a better estimation of flatness by placement of actuators. The problem formulation and the solution proposed are able to correct the flatness by approximately 95% of the initial deformation value. The work done in this research can be extended from an application point of view and can be utilized in estimating and correcting the deformations of any given structure. Since this research involved working in a 2-dimensional domain, the further implications of this research involves extending the proposed flatness correction technique to a 3-dimensional domain.

In a high-precision system that performs measurements or tooling on a workpiece, alignment of the tool and workpiece is of prime importance. To prevent misalignment, which leads to a loss in accuracy and precision, unwanted vibrations in structures must be attenuated. Topology Optimization (TO) is evolving as a mature design tool that provides innovative designs beyond human creativity. This thesis focuses on developing and investigating TO methods for the limitation of response peaks on a flat surface for suspended structures. When optimizing for multiple excitation frequencies at multiple output points, the complexity of the problem increases, and the number of required constraints grows manifold. Thus, for a compact formulation, there is a need for aggregation of peaks in both dimensions, space, and frequency. Furthermore, the application requires the top surface of the suspended structure to remain flat during operation. In such a scenario, where a structure is excited harmonically, the dynamic deformations on the surface become key to quantifying surface flatness. The incorporation of dynamic flatness measures in TO framework is studied and implemented, and the results show that the proposed methods look promising.

The absolute and relative stability are critical aspects of designing Controlled Motion Systems (CMS). The maximum attainable CMS performance depends highly on the design of the structural part and the actuator and sensor configuration, as mechanical vibrations can endanger stability. This thesis presents a method for designing CMS's using optimal design. The set of vibration modes that violates the stability criterion might change during the optimization process, which is undesired from an optimization point of view. An aspect of this method is to ensure stability by imposing a constraint regarding the stability of each vibration mode individually. The stability constraint proposed in this thesis aids in well-performing optimization that converges to feasible designs. This thesis presents a new way of modeling the open-loop response. It uses a standard PID controller often used in industry, and the structural model is a function of the poles and zeros of the system instead of the conventional modal parameters. Only three parameters are required to evaluate the robustness of a single mode. A robustness response surface was obtained via simulations and modeled using NURBS. The robustness response surface is used to evaluate the proposed robustness constraint. The surface model is suited for gradient-based optimization methods. The proposed method was tested on a relatively simple model of a motion system. The optimizer converged to unconventional solutions that outperformed designs obtained using engineering principles.

Reducing greenhouse emissions can be achieved by better insulating buildings. However, wall thickness increases significantly if conventional insulation materials are used. Conventional vacuum insulation panels (VIPs) offer a solution to this problem. VIPs have potential, but can not compete with conventional insulation materials due to their high pricing. A new concept vacuum insulation panel is introduced by the start-up company IQ-Bizz. To what extent this concept panel is a feasible alternative to conventional VIPs is the main research question of this thesis. The concept panel consists of two honeycomb plates, an intermediate foil and an MF2 panel envelope. The panel's geometry is chosen such that heat transfer through the panel is minimal and failure of the panel does not occur. Finite element models, verified by analytical calculations, have been used for evaluating the concept panel's structural and thermal performance. A thermal performance (R-value) of 6.23 m^2K/W is evaluated for a one-centimetre thick square panel with sides of one metre. Conventional VIPs of equal size have an R-value of 2.5 m^2K/W. The global warming potential of the concept is roughly 9 kg compared to 15 kg for conventional VIPs. A conventional VIP's price of roughly 21 euros is compared to the material costs of the concept panel of 8 euros. All performance parameters of the concept panel are higher compared to conventional VIPs. However, the concept panel's thermal performance is expected to be lower, and its environmental impact and price higher. Further research is required for a better evaluation of these performance parameters. Nevertheless, this thesis shows that the vacuum insulation panel concept has promising potential.

High-precision motion systems are crucial for many applications, such as in semiconductor equipment, microscopy, robotics, and medical devices. Next to high operating speeds, high accuracy and precision are required, which makes the design of these systems a challenging task. Dynamics, feedback control, and their interaction all play an important role in the design and its final performance. This thesis shows that topology optimization in combination with additive manufacturing offers new opportunities for the automated design of motion systems with unprecedented performance. The first challenge addressed is the manufacturability of the designs, for which a systematic optimization setup is presented allowing directly producible designs. This is verified by manufacturing and testing an optimized design. Furthermore, the computational time required for full-scale topology optimization is reduced significantly, by using reduced-order models and approximation of design sensitivities. Thirdly, effective optimization formulations are introduced that allow combined optimization of topology and controller for closed-loop performance, such as bandwidth, closed-loop stability, and disturbance rejection properties. Combining all these techniques, this thesis demonstrates that it is possible to perform integrated controller-structure topology optimization of motion systems of industry-relevant complexity.

Current technology of Organ-on-Chips allows investigation of human cells outside the body including a microfluidic medium flow. Control of oxygen in this microfluidic flow is required to mimic the situation as in the human liver (physiological) or to create temporary low oxygen concentrations (hypoxia). In this research, a gas exchanger module is designed for liver-on-a-chip cell experiments at the Erasmus MC. Based on prototype tests and COMSOL simulations, the gas exchanger module is designed and integrated with a commercial Micronit microfluidic channel. The functioning of the fabricated gas exchanger is validated by measuring an oxygen gradient as well as performing oxygen consumption measurements on a liver cell layer.

A compliant mechanism (CM) is a special kind of mechanism which has seen an increase in use in various high-tech applications. A CM is a structure designed with the intention to deform: motion is achieved by deformation of flexible members of its body. CMs are designed by topology optimization (TO) algorithms, which use boundary conditions as an input and give a design as an output. Additive manufacturing (AM) is used to fabricate the designed CMs, as geometric complexity is much less of a problem compared to conventional methods. One issue with AM is that local overheating during printing can cause defects in the design, specifically for metal designs, as the temperature is very high during printing. In this thesis, a CM is designed using TO and fabricated using AM. The design is then printed and analysed for overheating defects to determine if any defects due to overheating are present. Also, a computationally inexpensive AM model is integrated into the TO for CM, which can detect zones prone to local overheating. Next to that, the robust optimization method is used to obtain a design which needs much less post-processing. The obtained designs are compared to existing TO methods and it is found that the added constraint can reduce overheating by a large amount while maintaining a relatively high CM performance.

Applications for thermoelastic metamaterials, in which extremely positive, zero or negative thermal expansion is desired, are plentiful. Nevertheless they are rarely applied, due to their poor manufacturability. To enhance their applicability, a topology optimization framework is proposed and verified, able to generate finite thermoelastic metamaterials with tailored unidirectional thermal expansion, manufacturable through automated multi-material additive manufacturing methods. The superelement method is utilized to optimize for finite thermoelastic unit cell arrays connected to solid strips allowing for easy mounting, instead of non-realizable infinite arrays which are optimized in existing frameworks that use the homogenization method. The robust formulation is combined with a filter domain extension approach, to obtain manufacturing tolerant, black and white solutions and minimum length scale control. Uniform material layers are enforced, to obtain layered designs manufacturable with multi-material additive manufacturing methods. Using the proposed framework, a finite thermoelastic unit cell array is optimized for near zero thermal expansion. It is validated that the superelement method is more suitable for the computation of the thermoelastic response of small unit cell arrays, compared to the commonly used homogenization method. A physical sample is manufactured and experimentally validated. Experimental validation confirms numerical predictions and shows that the proposed design approach is able to generate performant and realizable thermoelastic metamaterials.

Power electronic systems are reaching higher efficiencies as their technology advances, which often results in components of smaller size with higher power densities. Cooling these components becomes increasingly challenging as high power densities require cooling with large heat fluxes. Topology optimization (TO) of thermo-fluids can be used to find cooling interface geometries which achieve high heat transfer with realistic pumping power. However, current methods for thermo-fluid TO show several issues. Firstly, the fluid models used for thermo-fluid TO show weaknesses that cause deviating behavior compared to conventional fluid models, which can result in large under- or overestimation of heat transfer especially when applied to turbulent flow. Secondly, although this deviating behavior is known to result in inferior modeling accuracy, the achievable accuracy of thermo-fluid TO has never been quantified. Lastly, TO currently requires many modeling parameters to be specified manually. Since these parameters largely affect the accuracy of the thermo-fluid solver, tedious parameter tuning is part of the TO design process. This thesis firstly presents a framework which allows quantitative analysis of the modeling accuracy achievable with density-based thermo-fluid models in 2D. The framework reveals several effects causing errors in turbulent flow, as well as a predictability of density-based boundary layer flows. Secondly, a method is tested which minimizes errors of the density-based thermo-fluid models by meta-optimizing the modeling parameters. Applied to a test-case with turbulent flow, this method achieves up to 27% reduction of the modeling error compared to a parameter sweep. When applied to a laminar TO, it achieves similar accuracy as a manually tuned TO without needing any tuning. A second mitigation method which adjusts the thermal conductivity of porous solid material to compensate for erroneous convective heat transfer is infeasible, as it provides less accurate results than the first method. Lastly, a post-processing method which uses the meta-optimized data is tested and found to provide better accuracy than a conventional post-processing method.

In recent developments in the field of multi-material additive manufacturing, differences in material properties are exploited to create printed shape memory structures, which are referred to as 4D-printed structures. New printing techniques allow for the introduction of prestresses in the specimen during manufacturing. This research focuses on bi-polymer 4D-printed structures, where the transformation process is based on a heat-induced stiffness transition in one of the materials and an initial prestress in the other material. Upon the decrease in stiffness, the prestress is released, causing a bending behaviour. A methodology to find the design of 4D-printed structures is developed, where a finite element model is combined with a density-based topology optimization method to describe the material layout. This modeling approach is verified by a convergence analysis and validated by comparing its numerical results to analytical and published data. The use of topology optimization to design 4D-printed structures is explored by applying the methodology to a variety of design problems. Bi-layer designs are printed with acrylonitrile butadiene styrene (ABS) and thermoplastic polyurethane (TPU), of which the latter material is prestressed during the Fused Filament Fabrication (FFF) printing process. Tests are performed with these printed samples, with the goal to find the prestress value and to further validate the modeling approach by comparing the numerical results to the experimental results. Once the prestress value is found, topology optimized samples are printed and tested to evaluate their performance.

In this thesis, a ground structure based thermofluid topology optimization (TO) method is presented which is applicable to problems with creeping flow. Fluid channels with creeping flow are found in a variety of engineering applications, primarily in the field of microfluidics. Current fluid channel TO is typically done with density-based TO and Darcy-Stokes flow. While this approach can improve the performance of cooling devices, certain benefits can be achieved by using a ground structure representation of the fluid channels. By confining the flow to a Poiseuille flow pipe network, a significant reduction (up to a factor 13 in 3-D) in the number of degrees of freedom can be obtained compared to an FEM discretization of the Darcy-Stokes equations, thus reducing computation time. Other benefits of this approach are related to the increased control of explicit geometry features such as channel minimum/maximum width, channel to channel angles, or the number of channel branches at junctions. The presented method, named the Ground Structure Projection (GSP) method, uses a projection step to map the channel geometry and flow onto an Ersatz material model. Performance is subsequently evaluated using an FEM discretization of the advection-diffusion equation. The modeling is validated with comparisons against a trusted commercial FEM package. For Reynolds numbers in the range of Re < 20, the GSP method shows fluid and thermal results within 5% accuracy of an FEM solution. Afterwards, a number of numerical optimization cases are analyzed including a multi-objective (pressure drop + average temperature) optimization of a cooling device. It is shown that the GSP method can achieve improved cooling performance with good objective convergence. In the second part of this thesis, the GSP method is applied to an industrial case concerning the reduction of optical aberration of a Projection Optics Box (POB) mirror in an EUV photolithography machine from ASML. Extreme precision is required in EUV photolithography machines and hence any unwanted optical aberration of the POB mirrors must be minimized. However, the mirrors absorb some EUV radiation resulting in thermomechanical deformation and consequent optical aberration. By optimizing the cooling channel layout to maximize surface temperature uniformity, an attempt is made to reduce optical aberration. A hybrid 2-D/3-D model is created to evaluate the optimization objective. It is shown that for a dipole and a circular heat load the temperature uniformity is improved significantly against a reference fluid channel layout. To check how well the chosen objective translates to reduced thermomechanical deformation, an optimized design is transferred to a commercial FEM package and analyzed. Since the optical systems in the EUV machine can compensate for some optical distortion, a second order surface polynomial is fitted to the FEM deformed mirror surface. The residual root mean square (RMS) error of the FEM deformed surface and the polynomial fit serves as a measure for optical performance. The optimized design performs 20% better than the reference design with a residual RMS of 3.14 µm versus 3.91 µm respectively.