The use of multi-axis additive manufacturing has enabled the possiblity to fabricate parts using curved layer deposition. Compared to the traditional deposition using planar layers of a fixed thickness, multiaxis additive manufacturing significantly increases the number of possible fabrication sequences. With the introduction of a time component for topology optimization in recent work, fabrication-processdependent physics can now be incorporated in the optimization process. The fabrication sequence resulting from such complex optimizations can lead to layers with large variations in thickness. For manufacturability reasons, considering multi-axis additive manufacturing, this variation should be controlled. In this thesis, we present a method to control the variation in thickness of the projected layers to improve the manufacturability. We use the continuous pseudo-time field and its gradients to track the development of the layer boundaries. To demonstrate the effectiveness of the method, we apply the thickness control method on a fabrication sequence optimization for minimizing thermal distortion. The results show that the proposed method is able to reduce the variation of thicknesses compared to the original sequence optimization without thickness control resulting in improved manufacuturability.

In the realm of traditional additive manufacturing, design and fabrication sequence planning have historically followed separate tracks. However, recent strides in the field, particularly in the utilization of robotic arms with multiple degrees of freedom, have brought forth a revolutionary approach known as Space-Time Topology Optimization (STTO). This groundbreaking algorithm breaks down the barriers between design and fabrication by simultaneously optimizing the structure and fabrication sequence. It achieves this feat by employing density and time fields as design variables, allowing for a holistic and integrated approach to the manufacturing process.
However, within the framework of STTO, multiple iterations of finite element computations become necessary. This results in a substantial computational burden throughout the overall process.
My contribution within STTO lies in its adoption of a multi-resolution strategy. This strategy enables the use of different resolutions for the design fields, enhancing computational efficiency. Coarsening, a critical component of this strategy, is implemented through a sophisticated weighted average scheme. This coarsening process facilitates the construction of stiffness matrices with significantly reduced finite element calculations, resulting in substantial time savings during the optimization process.
The impact of coarsening in STTO has been rigorously studied across various levels, yielding remarkable results and advantages. In 2D scenarios, this approach has achieved an impressive 5-fold reduction in computation time, while in the more complex 3D domain, it has led to an astounding 30-fold decrease. Moreover, it's worth noting that compliance, a crucial performance metric, maintains its integrity even with coarsening, with compliance drop remaining below 5% for levels deemed acceptable. This study illuminates the profound implications of coarsening within the STTO framework, emphasizing the significant strides made in computational efficiency while ensuring structural integrity and performance.

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.

Graphene is considered a promising material due to its unique electrical characteristics and excellent mechanical properties. These extraordinary properties make graphene a suitable material for applications like mechanical reinforcements, protective coatings, supercapacitors, sensors, etc. However, when trying to extract these properties experimentally, the challenge of producing pristine graphene prohibits researchers from attaining the required results. The challenge is primarily due to the formation of structural or surface imperfections during the fabrication, transfer or manipulation of the membrane. In addition, graphene possesses a low bending rigidity which inevitably leads to the out-of-plane crumpling of the film. So, determining the effect of these imperfections on the mechanics of graphene is critical to ensure development in the research field and application pertaining to graphene. The main focal point of this work is to experimentally determine the effects of surface corrugations, in the form of out-of-plane crumples, on the mechanical constants of the graphene membrane. In this study, two types of graphene membranes are considered, one is a flat membrane, and the other is a heavily crumpled membrane. The atomic force microscope is used to probe the membranes, and the mechanical constants are extracted from the resulting force deflection characteristics. The mechanical constants of the flat membrane, 2D Young’s modulus (E2D) and pretension (σ0) were found to be 140.42 ± 44.52 N/m and 0.119 ± 0.049 N/m respectively. The values obtained suggest a softening behaviour of the membranes, which is due to the intrinsic crumpling of the membrane in the out-of-plane direction. Surprisingly, in the case of heavily crumpled membranes, E2D and σ0 were found to be 393.34 ± 145.12 N/m and 0.203 ± 0.069 N/m respectively. The values obtained suggests that the heavily crumpled membranes provide greater resistance to deflection than the flat samples. However, it is essential to understand that the values do not depict the effective Young’s modulus or intrinsic strength of the membrane. The presence of folds in the heavily crumpled membranes is the reason for their higher resistance to deflection. The evolution of these folds, when subjected to nanoindentation, is also discussed in this work. This research helps provide insights into how the graphene membrane responds mechanically in the presence of crumples and also provides progress in realising the applications of crumpled graphene membranes.

In recent studies, selection criteria for lattice materials suitable for active shape-morphing have been composed. These criteria have directly resulted in the proposal of a lattice that is able to compete with the Kagome lattice, which in the literature is identified as a lattice material with optimal in-plane stiffness and shape-morphing capabilities. The selection of suitable lattice materials is mostly related to the properties of the lattice before shape-morphing. Naturally the question 'What is the effect of morphing on the macroscopic properties of the lattice material?' arises.

This research investigates the relationship between kinematic properties of the pin-jointed Kagome truss and elastic macroscopic properties of the welded-jointed Kagome lattice during deformation. The kinematic properties of the Kagome truss are explored at all possible collapse configurations. The same collapse configurations of the pin-jointed truss are then used to determine the elastic response of the welded-jointed twisted lattice with the use of the commercial finite element program ABAQUS.

The results show that only the pin-jointed initial configuration of the Kagome truss can support all macroscopic loads without inducing a collapse of the truss, no other pin-jointed configuration shares this property. Indicating different elastic responses of the initial welded-jointed lattice and the twisted lattices. The results of the finite element analysis support this finding, as it can be shown that the type of deformation of the initial configuration of a welded-jointed lattice differs from the deformation of any other twisted configuration. A stark difference between Young's modulus is observed between the initial configuration and the twisted configurations. The deformation induced by single member actuation creates a weak layer in the lattice. Therefore the attractive properties of the Kagome lattice are not preserved during shape-morphing.

Fused deposition modeling (FDM) is one of the fastest-growing additive manufacturing processes in recent years. Nevertheless, the mechanical properties cannot be matched to those of traditional manufacturing methods such as injection molding.The research presented in this report aims to demonstrate the limitations of fusion deposition models and proposes automated fiber placement (AFP) as a way to mitigate them by improving transversal tensile strength and elastic modulus.
The background and main characteristics of the stated processes, as well as the state of the art, are discussed in this paper. It is explained how both methods were integrated for a production line built specifically for the execution of this research. The key parameters are listed, along with their impact on the end part.
Mechanical parameters such as tape-to-print bonding strength, tape-to-tape bonding strength, molding unidirectional fabric tensile strength, and 3D printed part tensile strength were all compared in the experiments. The major challenges to be addressed are surface roughness, print-to-tape contact area, and heat transfer when taping, according to the results. Yet, there is still a large potential for these methods to achieve better performance.

Density-based topology optimization, when used to design structures that show geometrical non-linear behaviour, currently faces computational effort and stability issues. These issues are caused by the iterative method used in geometric non-linear structural analysis. On top of taking much computational effort to complete, this method encounters instabilities when analyzing low-density elements usually present in the design domain.

This study aims to bypass those issues by proposing approximate analysis in the topology optimization routine, which is an analysis based on an approximation of the geometrical non-linear load-deflection curve of a structure, constructed with equilibrium points close to its undeformed configuration.

To study the performance and the influence of the parameters that govern approximate analysis, three numerical examples are considered. These indicate that using approximate analysis in topology optimization leads to designs that perform over a finite range of motion, similar to when a non-linear analysis is used. The computational effort needed for approximate analysis is closer to the effort needed for linear analysis than non-linear analysis. A limitation of approximate analysis is that its results are only similar to non-linear analysis as long as the deflections stay in the mildly non-linear domain.

When concerning the topology optimization of compliant mechanisms that exhibit mildly geometric non-linear behaviour, we conclude that using approximate analysis is a stable and computationally efficient alternative to non-linear analysis.

The urge for more sustainable and cost-efficient repair methods for damaged parts is growing every day. On the other hand, laser metal deposition (LMD) is gaining momentum. This flexible technique, where material is added in a layer-by-layer fashion with a minimal heat input offers solutions for a wide range of production and repair purposes. Especially restoration or replacement of large and complex high-performance parts with long production times could profit from these developments. Industrial gears are certainly such type of parts. Occasionally, unforeseen tooth breakage of gears occurs, often leading to the replacement of the parts. No reliable repair techniques have emerged yet, due to the desired material properties and the uneven distributed hardness seen in carburized gears. This study was initiated to examine the feasibility of LMD for the repair of industrial gears with the goal to reduce both material waste and down time while retaining or exceeding the required mechanical performance. The initial part of the research is focused on finding the best suitable material and deposition parameters to ensure the desired mechanical properties of the core and a good bonding of the structure to the gear. Multiple deposition parameters are varied followed by destructive and non-destructive testing. It was found that the mechanical properties of deposited structures Inconel 718 and its bonding to a carburized substrate can be adjusted by accurately selecting the deposition parameters. Large differences in outcome were obtained by changing the deposition direction between each layer and tuning of the shielding gas flow rate. In further investigation the effects on the mechanical behavior of a deposited tooth structure were studied by altering the thermal cycle during and after deposition. The tests are performed on a specially developed static load testing setup. The results show a negative correlation between the interpass temperature and both the yield and ultimate tensile strength. This effect is attributed to a lower cooling rate during solidification at higher interpass temperatures. Ultimately, the results of a deposited tooth with hard flanks and built on a case hardened base were compared to a case hardened gear tooth that was manufactured according to the regular production method. It is found that the deposited structure exhibits significantly lower yield strength. Nonetheless, the superior ductility of the newly deposited tooth structure in the presented work is considered as a promising method for future gear repair.

The demand for using thin-walled composite structures for instance, in aircraft and ships are rising in the industry. These structures contain material interfaces, and at these interfaces, the structure exhibits non-smooth field behavior which is known as weak discontinuity. It is essential to study the effects of such discontinuities on the performance of the structure. These structures are usually modeled with standard finite element method (FEM) with fitted or geometry conforming meshes to the discontinuities. Shear locking is a key hurdle to overcome in FEM when using first-order shear deformation theory (FSDT) for thin structures. Classical beam (Euler-Bernoulli) or plate (Kirchhoff-Love) theories do not face such issues but ignore shear deformation and set the requirement of shape functions that are C1 continuous. The popular mesh independent method, eXtended/Generalized finite element method (X/GFEM) has also been used to capture the kinematics of discontinuous thin structures that are also devoid of locking. The presented work introduces a high-order discontinuity-enriched finite element formulation for linear and nonlinear (large deformations) discontinuous FSDT beam and plate elements. High-order interpolations are used to mitigate shear locking. The high-order enrichment functions are constructed by using hierarchical shape functions of the p-version of FEM (p-FEM), where the linear enrichment function is obtained from the interface enriched generalized finite element method (IGFEM) is combined with hierarchical shape functions of p-FEM. Convergence study shows that the proposed method accurately captures the discontinuity kinematics leading to an exact solution when compared with the analytical solution. A high condition number of the stiffness matrix is observed in the stability study which leads to loss of stability. A Jacobi like diagonal preconditioner is used to reduce the condition number but does not work well for high-order interpolations.

The purpose of this study is to investigate how metamaterials with a negative Poisson ratio, so-called auxetic metamaterials, can be used for conversion of pressure from a fluid to strain in a solid for application in a medical pressure sensor. As a first step, a literature review was carried out to identify the potential of auxetic metamaterials as a solution direction for pressure to strain conversion. From considering three optimal case scenarios for solution directions to create strain in a fiber, the approach of auxetic metamaterials stood out. High axial strains and a linear straining behavior, which is important for the read-out of the sensor, were expected. Auxetic Chiral structures were designed and fabricated and their effective Poisson ratios were experimentally evaluated. It was found that the boundary effects that are introduced during experiments influence the behavior of the auxetic structure, especially on the unit cells close to the boundaries. A region was found where the unit-cells behave as theory describes and the boundary effects are dissipated. Additionally, it was found that for the an-isotropic Meta Tetra Chiral structure the Poisson ratio is influenced by applied strain. The isotropic Anti Tetra Chiral structure did not show strain dependent behavior. The relation between pressure to strain conversion and the effective properties of auxetic structures indicated that an-isotropic effective properties are preferred. Using this finding Re-entrant Honeycomb structures were designed for high pressure to strain conversion. From measuring the force input and output while deforming the structure, the expectations were verified. In this research it was found that auxetic metamaterials with an-isotropic effective properties, such as Re-entrant Honeycomb structures, are an effective intermediate for converting pressure from the fluid domain to strain in a fiber.

The goal of this study is to understand the processes and wing characteristics that influence lift generation of a flapping wing, and apply the acquired knowledge in an optimization setting to obtain an optimal wing thickness distribution which maximizes effective wing lift. To achieve this, an extensive literature survey was conducted to identify geometrical characteristics of the wing that are known to help maximize wing lift, and loading conditions under which a given wing design must be numerically studied. Additionally, a cubic pressure load distribution was developed to approximately mimic the aerodynamic pressure acting on the surface of the wing. Changes were made to the research code, Charles, to achieve some of the desired features. Using some of the identified parameters and load cases, an optimization problem was formulated and thickness distributions were obtained for two wing planform shapes. The results were compared and discussed, and general patterns were identified which give insight on flapping wing design.

Origami structures have become of increasing importance in the field of engineering, due to their lightweight, compact and stiff properties. To design these structures, progress is being made into incorporating origami modeling in the Finite Element Method. To implement an arbitrarily located fold on an existing mesh, either re-meshing or enriched finite elements are required. In fold pattern optimisation of origami structures, re-meshing each intermediate design would not be time efficient, whereas enriched elements may be very time efficient. In this thesis foldable Kirchhoff-Love plate elements are derived using a mixed/hybrid element formulation in combination with an enriched Finite Element formulation. The use of a mixed/hybrid element formulation enables great simplification of the enrichment functions, since the discontinuous rotation field is evaluated at the boundaries of the enriched plate elements, instead of the element domain. Six foldable plate elements are derived and tested for accuracy and stability. The stability is improved by either local condensation of the enriched elements or by applying a precondition matrix.

Externally triggered lengthening or shortening (actuation) of one or more members in a lattice material can be used to achieve macroscopic shape changes. The three main attributes looked for in 2D lattice materials to be deemed suitable for actuation are in-plane isotropy, high specific elastic properties and limited energy requirement for actuation. However, no infallible topological criteria have yet been discovered that determine which micro-architectures meet these requirements. The Kagome micro-architecture yields the best-performing lattice material for actuation known to date.

This thesis contributes to the search for micro-architectures that perform similar to Kagome lattice material, or even outperform it. Four novel micro-architectures are introduced. Each of the new micro-architectures is verified to constitute a stiff (i.e. stretching-dominated) isotropic lattice material. Still, not all of them result in optimal elastic moduli. One of the proposed designs does perform identically to the Kagome structure in terms of elastic moduli. Moreover, it requires less energy for actuation in the range of relative densities of interest. On the other hand, its attenuation distance is shorter; actuation deformations damp out within a shorter distance than in Kagome lattice material.

Generalized finite element methods have proved a great potential in the mesh-independent modeling of both weak and strong discontinuities, such as the ones encountered when treating materials with inclusions or cracks. By removing the constraint of a conforming mesh, more freedom is offered to modeling exact geometries by means of splines. However, very few studies have been published which combine Non-Uniform Rational B-Splines (NURBS) to interface-enriched methods, addressing uniquely weak discontinuities. Therefore, the aim of this thesis is to propose a NURBS-based enhancement to the Discontinuity-Enriched Finite Element Method (DE-FEM) in two dimensions and to discuss the potential of its application. The main advantage of this method is the possibility to study problems that present discontinuities with arbitrary smooth shapes, while maintaining exact geometries throughout the analysis: in this way, the equivalence between design and computational geometry is preserved. To this purpose, a suitable NURBS-based analysis technique is selected and implemented within the framework offered by the group's finite element library, Hybrida.
The capabilities of the NURBS-enhanced DE-FEM to solve several weakly discontinuous problems are assessed for composites of different complexities. Furthermore, a novel study is presented which extends this technique to the treatment of strong discontinuities, in the context of fracture mechanics. The accuracy, convergence properties and numerical efficiency of the proposed method are investigated, in particular in comparison with the standard DE-FEM. Based on these observations, further insights are provided into the convenience and the limitations of adopting NURBS enhancements within the DE-FEM formulation. Lastly, some recommendations about possible directions of improvement are provided.