A large number of offshore wind turbines have been installed in recent years, and even
more are being erected in current and future projects. The majority of these turbines
have been or are being built of monopile foundations. Eventually, both the turbine and
the foundation have to be decommissioned. However, decommissioning options for
monopile foundations are limited, especially for complete removal of the foundation.
This thesis aims to study a new removal method patented by the company Iv-Consult,
which could potentially make entire monopile removal possible. The method uses
deformation of the monopile wall by a clamping tool to generate a vertical force. Since
the thesis served as a preliminary study into the removal method, it was chosen to limit
the scope of this study to the monopiles that are expected to be decommissioned in the
near future. With this information, a single case of monopile with dimensions that are
representative for these monopiles was presented. To analyse the removal method on the
case monopile the finite element method was used. In order to ensure results obtained
using the finite element method are accurate and representative, a convergence study was
carried out. To determine the element type and the mesh size that was the most suitable,
two different models were used: a ring model and a small cylinder model. Each was
studied using different combinations of inputs. After the results from the convergence
were obtained, these results were used on the models with which the removal method
was studied. Different configurations of the removal method were analysed with the
aim of maximizing the force that can be generated by the method to overcome resistance
forces. As a final step, the resistance forces that need to be overcome were calculated
for different soil types and compared to the results obtained from the removal method
models to verify the feasibility of the method.
From the results of the convergence study, it was shown that it is possible to model
the removal method using the finite element method and obtain representative results.
The final results of this thesis showed that clamping the monopile multiple times
does not have a positive influence on the vertical force that can be generated. This is
because clamping the monopile only once caused the highest vertical force. However, the
forces generated using the different setups of the clamping method were not sufficient
to overcome the resistance forces. This shows that the removal method in the current
configurations is not feasible, without applying any soil resistance reducing methods.

Structuresmade of fiber-reinforced polymer (FRP) composites are usually considered lightweight. On account of this feature, polymeric composites find their application in many engineering disciplines, where the expected service lifetimemay extend to decades. The advent of thermoplastics has brought about cost-effective processing and the possibility to recycle structural components, among other features. However, the viscous nature of the polymer matrix poses a challenge to the straightforward application of FRP composites in load-bearing situations: the material response is a function of time and temperature. In addition, structural components may be subjected to long-term cyclic loading, that may lead to failure irrespective of the viscous nature of the material constituents. Hence, for an optimal application of (thermoplastic) FRP composites in load-bearing situations, it is necessary to understand, and eventually to predict their long-termperformance....@en

The global focus on climate change and the transition away from fossil fuels has highlighted the importance of renewable energy sources. Offshore wind turbines are being optimized and are therefore growing in size and power.

This research focuses on bolted ring-flange connections, a connection type that plays a crucial role in the design of offshore wind turbines, as they transfer the external force between parts of a turbine. The objective of this thesis is to analyze how the increasing dimensions from current to future offshore wind turbines and geometric imperfections impact the reliability of analytical approaches for load transfer functions (LTFs) for these connections. Two components of this objective are considered: examining the influence of different dimensions of ring-flange connections and analyzing the impact of various gaps between flanges on LTFs for 'current generation' and 'next generation' turbines. Analytical calculations are compared to results obtained with finite element analyses, which are assumed to represent an actual connection.

Based on the research findings, the following conclusions are made. Firstly, the widely used tri-linear approach by Schmidt/Neuper [18] for obtaining the LTF in bolted ring-flange connections is found to be unreliable for current and future turbine sizes. This method highly underestimates the forces in the bolts when initial gaps are present between the flanges. Calculations performed with this approach could lead to an overestimation of the turbine's lifetime compared to reality by multiple years, possibly causing more maintenance or early failure. Alternative approaches, such as a very new and not yet approved polynomial approach, show reliable results, providing accurate estimations of bolt forces for large connection diameters. Additionally, currently verified tolerances for gaps between flanges (1 mm over 30° and 2 mm over entire circumference) are outdated, and larger gap heights or smaller gap lengths are expected in practice, especially for future turbines. These gaps lead to a larger bolt force in practice, decreasing the fatigue resistance and lifetime of the structure. Even though very small gaps are expected to occur often, lower bolt forces are obtained compared to larger gaps, both with an expected height to length ratio of 𝑢𝑔𝑎𝑝/𝑙𝑔𝑎𝑝= 0.53 ∗ 10^−3. In analytical design calculations, it therefore is recommended to consider larger sized gaps with a gap length of approximately 1600 mm with its expected gap height.

Offshore wind energy production is increasing rapidly and optimising the design of support structures has the potential to significantly reduce steel and costs. Jacket structures are commonly selected for deeper waters, and the design of these structures is primarily influenced by the fatigue performance of the tubular joints. The fatigue resistance of these joints can typically be evaluated using different stress assessment methods, each associated with a corresponding S-N curve. While the Hot Spot Stress (HSS) approach is commonly used for design purposes, the Effective Notch Stress (ENS) approach offers higher accuracy by taking into account the stress concentration originating from the weld profile. Consequently, the ENS approach has the potential to provide less conservative fatigue life estimations, leading to the reduction of steel and costs. However, the application to and research of the ENS approach for tubular joints is limited. Therefore, the objective of this thesis is to investigate the application of the ENS approach on a tubular joint. The research was divided into three parts, including an investigation on the required FE mesh, a study on sub-modeling of the ENS approach and a comparative analysis on the fatigue life predictions between the HSS and ENS approaches.

The FE mesh required for the ENS approach was evaluated on a cruciform joint according to the DNV-RP-C203 validation methodology and the Stress Concentration Factors (SCFs) values obtained for different mesh configurations were compared. Two mesh variants were developed, and different element sizes in the notch were evaluated. The results show that with a maximum element size in the notch equal to 0.39 mm, according to DNV, unacceptable SCF errors are obtained of -4% and 4%. Furthermore, with a maximum notch element size of 0.25 mm, according to IIW, SCF errors of -0.3% and -2.2% were obtained, indicating that accurate results can be obtained but are not guaranteed. Additionally, the accuracy of the mesh was found to be primarily influenced by the number of nodes surrounding the notch, rather than the size of the elements in the notch.

The accuracy of sub-modeling for the ENS approach on a tubular T-joint was evaluated by varying global modeling choices and sub-model sizes. The accuracy was measured by comparing sub-modeling stress results against those from a global T-joint FE model with the ENS approach applied. The results show that a global model composed of 3D elements with a weld profile that corresponds to the sub-model is recommended. Simplifying the global model by excluding the weld profile or using 2D elements has been found to provide inconsistent and inaccurate results. Moreover, a sub-model of the weld area with a size as small as 1.67 degrees of the brace radial angle has been found to provide consistent results. Thereby, sub-modeling reduces the required mesh size by a factor of 15, compared to the global modeling approach. Additionally, a rather surprising maximum principal stress error of approximately -6% was observed as a result of sub-modeling for all varied load types and sub-model sizes.

A comparative study between the HSS and the ENS approach was conducted for the computed fatigues of a tubular T-joint. Various FE models were composed for the T-joint subjected to different load cases, and equivalent SCFs were computed by introducing a correction factor based on the S-N curves of both approaches. The results show that the ENS approach provides a lower computed fatigue life over the HSS approach when applied to a tubular joint. This discrepancy is attributed to the FAT225 S-N curve, used for the ENS approach, which is not tailored for tubular joints but for welded straight plates. For the HSS approach, the choice of element type (2D vs 3D) and the inclusion of a weld profile have a significant effect on the stress distribution and the resulting fatigue life prediction. It was found that a model consisting of 3D elements with the inclusion of a weld profile yields the longest computed fatigue life for a tubular joint.

To conclude, the application of the ENS approach was investigated for a tubular joint, by evaluating the mesh sensitivity, examining the accuracy of sub-modeling and comparing the computed fatigue lives with those from the HSS approach. Further research is recommended to improve the computational efficiency of the ENS mesh and to achieve a deeper understanding of the -6% error found by sub-modeling. Additionally, the development of an S-N curve tailored for tubular joints is suggested to improve the accuracy of the ENS approach for this type of joint.

Both preloaded and injected bolted connections are commonly used in steel bridges in The Netherlands. Where preloaded bolted connections are already used for a longer time, the injection of bolt holes is a method used from the 1970s onwards. Injection of bolts became a good alternative for the replacement of riveted connections. This injection increases the slip resistance of the connection by filling the clearance between the bolt and the plate. Combining both methods increases the slip resistance of a bolted connection and is therefore even more beneficial.
According to the standard NEN-EN1993-1-8 [17], it is safe to assume that the resistance of a preloaded and injected bolted connection is equal to the sum of the two individual obtained resistances from preloading and injection. The aim of this research is to validate this statement and thus investigate the resistance of a slip-critical steel-to-steel connection including preloading and/or injection.

The study addresses this manner through both experiments and numerical analysis, addressing all three types of connections. In addition, this potential of summation is considered for two different levels of preloading, implementing plastic and elastic preload preloading. These levels of preloading were considered for two reasons, embodying different applications and identifying the effect of the preload force on the resistance. Each experiment is based on a short-term testing method that is in compliance with the described method of the EN1090-2. These resistances are verified by a numerical analysis with the goal of gaining more insight into the behaviour of the connection and validation of the obtained results.

From both experimental and numerical results, it is concluded that a summation of the individual resistances overestimates the actual resistance of a preloaded and injected bolted connection when measuring at the edge of the plates. This overestimation is similar for both preload levels.
Based on these results, it is concluded that it is unsafe to assume that the resistance of a preloaded and injected bolted connection is equal to the sum of the individual resistances obtained from preloading and injection. However, this conclusion might change if a different criterion for measuring relative displacements at specific locations is implemented.

The resistance of a connection is defined at a fixed magnitude of relative displacement measured at the edge of the plate, assuming that this is representing the displacement of the bolt. However, it turned out that the assumed limit displacement of 0.15mm did not result in the same displacement at the bolt(group) for all connections due to the amount of preload force and the geometry of the plates. Therefore, it is recommended to investigate the sensibility of the measured displacements for different geometries of and preload levels in the connection.

Munition articles may be sensitive to case-crushing loading. This is a potential safety hazard since it may lead to accidental detonation. A better understanding of the mechanism causing the munition articles to react can help improve safety. A finite element model combined with post-processing is used to model a Steven test. This thesis specifically looks at the absence or occurrence of a self-sustaining reaction in energetic material PBX-9501, a polymer bonded HMX based explosive. A reference paper (Gruau et al., 2009) was used to model the behaviour and validate some of the results.
The PBX is predicted to react if it has satisfied a certain reaction criterion based on the hot spot theory of ignition. This theory poses that known chemical properties of the material, the plastic shear strain rate and the hydrostatic pressure can predict the occurrence of a reaction. A finite element model was used to find the plastic strain rate and hydrostatic pressure in the energetic material subjected to a case-crushing loading. Next, the reaction criterion is computed in a separate script. The energetic material was modelled as a homogeneous continuum with a Drucker-Prager yield criterion. This material model did not include any post-yielding behaviour, like strain hardening.
First, a finite element model was constructed that was used as the baseline model throughout the thesis. The combined FEM and post-processing model was shown to be more sensitive than the model in the reference paper. The combined model also predicted a reaction at a lower velocity, which, according to tests in the reference paper, should have not produced a reaction.
The baseline model, and consequent models in the parameter study, also showed excessive deformation. This can be attributed to the use of a simpler Drucker-Prager material model for the PBX as opposed to a more complex material model. The absence of any hardening behaviour, and the resulting deformation, also contributed to the instability of the finite element model.
A parameter study was conducted by adapting the baseline model. Varying the velocity of the projectile was the most impactful change.
In conclusion: the developed model gives a first indication of parameters that contribute to the sensitivity of certain munition articles to case crushing. However, the material model used should be improved to create a more stable finite element model. An improved material model could also better capture the mechanical behaviour of the complex energetic material. It is recommended that future research should focus on this improvement.

In the presented work, we have developed a method to analyse the interaction between the wind and membrane structures. The complex behaviour of light weighted membrane structures can induce dynamic and wind effects, so this requires an appropriate method to analyse the structure. When the deformation of membrane structures is large, it becomes necessary to consider the interaction between wind flow and membrane. This interaction leads to aeroelastic problems.
The necessary properties to be considered for the aeroelastic coupled problem are considered and presented. The major focus of the thesis is to simulate the geometric non-linear behaviour of the membrane structures when subjected to wind loading. To derive the initial equilibrium shape, the method of form finding is applied. To properly simulate the wind flow around the structure, existing research regarding wind tunnel testing of a hemispherical air dome model is used. First, hemispherical model similar to existing research was made to verify the boundary conditions. The results are compared to the wind-tunnel testing results to get proper boundary conditions in Computational Fluid Dynamics models. In order to keep the simulation simple, the wind velocity is considered constant with height and constant wind flow with respect to time.
Partitioned analysis is used to simulate the physics behind the wind-membrane interaction simulation. By using this method, the multi-physics problem is separated into individual fields. Considering the wind-membrane interaction, the problem is separated into fluid domain and structural domain.
The structural and fluid domain mechanics are discussed in detail. For the single field solvers, methods are introduced based on the fundamentals. Finite Element method is used for the form-finding as well as numerical simulation of the structural field and SOFiSTiK software is used. For the fluid simulation, ANSYS CFX is used.
The strong physical coupling is done between the fluid and the structural domain. For this, the partitioned coupling simulation is used. The requirements and methods for the partitioned analysis are presented as well. For the coupling of separate solvers, a central coupling tool is made using Grasshopper. The transfer of coupling data such as displacement and wind pressure on the membrane structure is done using the developed coupling tool. In the development of this coupling method using Grasshopper, Python programming language is used. The developed method was used in iterations to analyse the effect of geometric non-linearity.
In order to make the developed method reusable for the air-supported membrane structures, the whole process is made parametric using grasshopper. This enables us to analyse the membrane structures with known base shape, internal air pressure. Ultimately this method can be used to analyse the effect of geometric non-linearity on membrane structures when subjected to wind loading. To show the use case of developed method, two different models are developed to analyse the fluid structure interaction. First model has been considered with a circular base shape and the second one has been considered with a square base shape. In order to analyse the effect of geometric non-linearity on these structures, the results such as maximum wind pressure on the membrane structure and maximum structural displacement are compared to the results without considering the geometric non-linearity.
Moreover, wind variation analysis is done to analyse the extent to which geometric non-linearity is dependent on the wind speed. Similarly, size variation analysis is also done to analyse the extent to which geometric non-linearity is dependent on the structure size, keeping the wind velocity constant.

Energetic materials have widespread applicability in various fields of engineering as propellants, explosives, and pyrotechnics. The Netherlands Organisation for Applied Scientific Research (TNO) is investigating additive manufacturing of solid propellants. As part of this investigation, TNO developed a new polymer binder mix for solid propellants. The thermomechanical behavior of the solid propellants is not well understood yet. Therefore, TNO is interested in developing computational modelling schemes for characterizing their behavior. Computational schemes based on the finite element method (FEM) are developed in this thesis. They are calibrated with matrix-only and solid propellant uniaxial tension experiments provided by TNO. The developed schemes are compared with the experiments to better understand the materials behavior and improve the computational models.

The macro-structural behavior of solid propellants is significantly affected by the behavior of the micro-structure. It is computationally unfeasible to directly consider the micro-structure in macro-structural FEM. Therefore, computational homogenization (CH) is employed for idealizing the micro-structure with representative volume elements (RVEs) at a finite number of macroscopic locations. Micro-structural behavior consists of numerous nonlinear thermomechanical processes. This thesis focuses in characterizing micro-structural matrix viscoelasticity, continuum matrix damage, and debonding. The brittle and rate dependent behavior of the matrix and the solid propellants is described with these thermomechanical processes. Other relevant processes such as temperature effects, nonlinear elasticity, anisotropy, and large strains are not considered.

The experiments exhibit a clear rate dependence throughout the entire loading process, and viscoelasticity is hypothesized to play a major role in this dependence. The experimental samples experience a brittle failure that is believed to be caused by matrix micro-crack damage. Particle debonding is observed in the failure planes of the solid propellants and is hypothesized to significantly affect their behavior.

Matrix-only and solid propellant computational results confirm that viscoelasticity is a major source of rate dependence and that (continuum) matrix micro-crack damage causes brittle failure. They also suggest that continuum matrix damage is a significant source of rate dependence in the post-damage regime, and matrix-only experiments agree. Comparing the results for perfectly bonded solid propellants to the results for solid propellants with debonding and to the solid propellant experiments, it is clear that debonding indeed has a large effect in reducing the ultimate strength of the material. The full CH damage-debonding-viscoelastic scheme shows promise as a first step towards full characterization. Future work can improve the scheme by addressing the issue of free surface propagation and can build on it by including relevant thermomechanical processes that are not yet implemented. Recommendations on how to achieve this are given in the final chapter.

Designing engineering structures relies on accurate numerical simulations to predict the behaviour of a structure before its realization. In the design process, many variables influence the final structure. There is an incentive for optimizing the design based on the desire for cheaper structures and less material usage. Although a wide variety of optimization techniques exist, in practice many structures are not fully optimized and could fulfill their purpose with less material usage. This partly results from optimization techniques generally requiring many simulations for problems with a high number of variables. In creating increasingly complex models, simulating its performance can take up to days or weeks to compute, incentivizing the development of optimization methods that use as little simulations as possible. The goal of this thesis is to provide a method of optimization which significantly reduces the required number of simulations to be performed, aiming to overcome the issue of required computational effort. The optimization problem used in this thesis is the geometric optimization of a fiber reinforced polymer (FRP) microstructure. Each evaluation requires a Finite Element analysis, and many parameters are required to describe the geometry. The goal is to find the maximum stress at perfect plasticity during uniaxial tension, interpreted as a measure of strength. While generally only fibers with a circular cross-section are used, here a morphing parameter is introduced that changes this shape between being circular and square. Furthermore, fibers are allowed to overlap, effectively creating a single fiber with a complex shape. This geometry optimization provides a challenging problem for which different optimization techniques can be compared. The approach taken in attempting to minimize the number of function evaluations is a combination of several machine learning methods. Based on a limited number of initial samples, Bayesian optimization (BO) is applied. In BO a prediction model in the form of a Bayesian Neural Network (BNN) is created and used to inform further sampling. This prediction model provides a mean and a standard deviation in its prediction, both of which are used to find the best point to sample next. The number of initial samples required to create an accurate prediction model increases exponentially with the number of parameters. It is therefore opted to first use a variational autoencoder (VAE) to reduce the number of parameters in which BO is performed, by encoding the parameters in a lower dimensional representation. The variational autoencoder does this without requiring any function evaluation. Results show that when encoding the original parameters using the VAE, the encoded representation is unable to recreate all possible configurations of the original parameters. Furthermore, by transforming to a lower dimensional representation, within this representation the complexity of a function increases compared to the original function, generally leading to many local optima. This complexity proves difficult for the BNN to accurately predict based on a limited number of samples. As a result, BO does optimize the result, but does not reliably find the global optimum of the reduced parameter space. No clear conclusion can be made on the overall performance of the method compared to alternative optimization methods. Recommendations are made for what additional studies could be performed. Further recommendations are given for solving issues limiting the current performance as well as possible additions to the framework. The fiber geometry optimization serves as a good case to compare different optimization methods. It is shown that the geometry has a significant influence on the mechanical performance of a FRP microstructure, and optimization is therefore beneficial. The load case considered is however too specific for the optimized result to have direct practical use. Still, it does demonstrate that the results of an optimization study can be generalized. As the optimization framework is data driven, the optimization objective could easily be extended to study more realistic problems.

The Rotterdam municipality is constantly busy replacing old footbridges that have reached their lifetime. Lately, this is done using modern materials, such as fibre-reinforced polymers (FRP). FRP is a relatively light-weight material that has a high yield strength, which allows for aesthetically pleasing, very slender designs. Governing for such designs is often the dynamic behaviour of the bridge: its eigenfrequency may very well lie within the `critical range' of common footfall frequencies. If the step frequency of a pedestrian crossing the bridge lies near the eigenfrequency, a high dynamic amplification factor occurs: high vertical acceleration values for the bridge deck will occur, which means discomfort for the pedestrians.

In this thesis, a sustainable solution to that problem was sought. Rather than using pile foundations to clamp the bridgeheads (and thereby increase the eigenfrequency out of the critical range), it was investigated whether the kinetic energy can be harvested from the bridge. Ideally, this both dampens the vibrations and provides free, sustainable energy.

Several theoretical models were developed to describe the behaviour of an existing footbridge for which data is available. Adaptations to these models were made in the form of either extra dashpots or a tuned mass damper (TMD). In both cases, the dampers are regenerative, i.e. able to harvest energy.

It was shown that adding a TMD provides the best energy harvesting performance. From a practical perspective, it has some advantages as well, such as universal applicability. The parameters in the single degree of freedom (SDOF) model were optimized for maximum energy harvesting while still keeping the vertical deck acceleration at reasonable levels. Finite element analysis largely confirmed these design choices, and showed that reducing the vibration levels is certainly possible with only a relatively small TMD. For a single person crossing at near-resonant footfall frequency only leads to a vertical acceleration of about 1.5 m/s.

The amount of energy that can be harvested in such a way is small but non-negligible: about 26 Joules worth of energy is estimated to be gained by an `average' person crossing the bridge. This is taking into account system efficiency as well.

A literature study concerning real-world energy harvesting systems was performed, and showed the possibility to closely emulate the theoretical behaviour that was assumed for the regenerative dashpots. This can be done using electromagnetic dampers in combination with proper power electronics design.

Reinforced composites are used in many industrial and multi-functional applications. The efficiency of the reinforcements depends mainly on the aspect ratio, material properties, and the adhesion between matrix and reinforcement. Particularly, high aspect ratio fillers and inclusions have gained popularity due to their unique material and geometrical features, where a fundamental understanding of composites hierarchical structure and behavior is crucial for the optimal design and performance. There is however a lack of robust numerical modeling frameworks that are able to accurately represent composites with high aspect ratio reinforcements. Ideally the expensive mesh generation of the standard finite element method or the simplifying assumptions adopted by smeared type or mean-field approaches should be avoided.
A group of numerical techniques here referred to as "embedded methods" eliminate mesh conformity restrictions and significantly reduce the computational cost of the standard finite element method, while still benefiting from the advantages of a direct numerical analysis. In formulating the embedded models, enrichment techniques and different element technologies are considered, and physical assumptions are investigated. Limitations of the classical embedded models are highlighted through numerical examples, on the basis of which possible enhancements are discussed. We specifically highlight the important roles of field gradients continuity/discontinuity and the element size, order, and regularity extensions on the smoothness of the solutions.
A computationally efficient embedded model is then applied to the study of failure and inclusion orientation effects in planar composites. A detailed study is also performed for dense fiber-reinforced composites, where homogenized mechanical properties are extracted and various forms of neutrality of thin fibers are demonstrated. In this context, a part of this thesis is dedicated to one-to-one comparisons between results obtained using the standard finite element method and embedded techniques. This led to a range of model and geometry parameters under which predictions of embedded technique are reliable. Comparisons are reported in terms of homogenized properties and local field variables, namely relative displacement between inclusions and matrix (slips).
Finally as a preliminary step towards multi-functional fiber-based structural batteries, an electro-chemical system characterized by composite cathode in a half cell configuration is considered. The main point of difference with common composite batteries is that active material particles are cast in form of high aspect ratio fibers, which are efficiently discretized by use of the embedded technique. A discrete definition of fibers, unlike the case of mean-field approaches, allows to define local fields and interfacial conditions between fibers and electrolyte and is crucial for the accurate modelling of a battery cell with fiber-based electrodes.@en

Numerical simulations have become an essential part of design in every field of engineering, and the boundaries of technology are pushed further out every year. In structural engineering, the desire to design structures that have complex shapes or that are simply cheaper and more efficient, has necessitated the use of complex numerical simulations. This necessity is further substantiated when taking into consideration structures such as wind turbines that are subjected to extreme environmental conditions. In many cases, however, such simulations are prohibitively expensive due to complex material behaviour or the many-query nature of design optimization. The lack of knowledge and understanding of the behaviour and failure mechanisms is compensated by adopting less complex designs and/or high safety factors, which leads to less efficient and more expensive designs. In recent years, methods to circumvent such high computational costs have been developed. Acceleration techniques such as model-order reduction (MOR) are now widely researched, and significant developments are being made to overcome the issues of prohibitively expensive high-fidelity models. This thesis uses Proper Orthogonal Decomposition (POD) to drastically reduce the degrees of freedom of a simply supported beam that is loaded in the downwards direction along the span. The result of the MOR process is a reduced-order model (ROM) that accurately approximates the behaviour of the full-order model (FOM). The ROM is constructed by determining a set of basis vectors that contain compressed information of representative full-order solutions collected in an \textit{offline} training phase. The goal in this thesis is so collect the information and construct the reduced basis as efficiently as possible while guaranteeing a given accuracy of the ROM. Two methods are presented to iteratively construct the reduced basis. The first one is the \textit{Surrogate Parameter Space} (SPS) method, where greedy sampling is performed on incrementally finer grids of points along the beam. Each individual grid is referred to as an SPS, and they are exhausted by selecting the load location that in each iteration will add the most new information to the ROM. The other method is the \textit{Gaussian Process Regression} (GPR). Greedy sampling using a Bayesian machine learning algorithm involving GPR is used to predict load locations along the beam that add the most new information to the ROM. A method to efficiently combine and compress information obtained in each iteration was also developed. The results showed that the SPS method is efficient to construct an accurate ROM for the beam model in this thesis, but the GPR managed to recognize that some areas in the span did not have to be sampled as much as others. This makes GPR the more promising method for other high-fidelity models when high accuracy is desired. The results also showed that both methods depend greatly on input parameters that define how much information is kept in each iteration, and for GPR an additional parameter that determines how much accurate the regression itself should be. In order to execute an efficient offline phase, these parameters must be chosen carefully, and it is recommended for future work to develop methods to adaptively choose them. It was also shown that the number of ROMs that have to be run as part of the greedy sampling is the bottleneck of efficiency for both methods. For high-fidelity models with higher-dimensional parameter spaces, this bottleneck necessitates the use of hyper-reduction techniques such as the Empirical Cubature Method (ECM) to reduce the computation time of the ROM itself. It is recommended that the methods investigated and the corresponding results in this thesis are used as a stepping stone to implement automatic and efficient sampling methods to other high-fidelity models.

The service life of concrete structures depend largely on the durability of concrete. The durability of concrete is influenced by mass transport mechanisms that can have severe deteriorating effects. Transport of water in concrete is of paramount importance as water can act as a carrier of ions such as chlorides which can corrode the reinforcement and reduce the service life of concrete structures. The main objective of this thesis is to study moisture transport through capillary absorption in concrete. Numerical simulation of moisture transport is performed through lattice elements with an irregular mesh configuration. New computational tools are developed and compared with the existing tools in terms of effectiveness to simulate moisture transport in homogeneous uncracked concrete. The existing numerical model uses approximate volume of transport elements to determine the volumetric capacity of elements. During discretization of the governing equation, the variation of diffusivity of the elements is approximated as a uniform mean distribution and an explicit time stepping scheme is implemented which has consequences on the flow equilibrium at each time step. In the proposed numerical model, exact volume of transport elements is considered and during the discretization of governing equations, exponential variation of diffusivity is considered along with an implicit time stepping scheme. Moisture transport is accurately simulated in uncracked homogeneous concrete even by using the existing model as validated through experimental results. Both the models are compared through error analysis by varying mesh size and time step. Moisture flow through different diffusivity coefficients is simulated using both the models and it is observed that the existing model suffers from oscillations in the saturation level during initial stages of flow due to inability of the existing model to maintain flow equilibrium at each time step. The proposed model on the other hand shows no such oscillations due to the flow equilibrium being maintained at each time step. Mesh size, magnitude of time step and diffusivity coefficient are shown to be the parameters limiting the effectiveness of either of the models. A single discrete crack is considered within the domain and its influence on moisture transport is observed. The nature of the crack is considered through two approaches. In the first approach, crack is considered as an additional porous phase in cement matrix and the entire volume occupied by the crack is considered to be a void. In the second approach, in addition to considering the porous nature of crack, moisture surface interaction between the water surface and crack walls is also considered. Horizontal moisture distribution around a planar crack is simulated using both the approaches which show similar results as in the experiment. The vertical moisture distribution is simulated within a wedge shaped crack and the results are compared with experimental observations. The first approach shows a slower rate of saturation of the crack as compared to the experiment. The second approach shows the vertical saturation of the crack similar to the experiment. A comparison between the Delaunay and Voronoi modelling techniques of moisture simulation in cracked domain is discussed. It is shown that the presence of a crack accelerates the moisture transport in concrete as it exposes additional surface of concrete from where water can penetrate in the material and also increases the diffusivity of the concrete material lying in the vicinity of the crack. Moisture transport is simulated in concrete considering its multi-phase nature. Concrete is considered to be composed of three phases i.e. cement paste, aggregates and ITZ around aggregates. Aggregates are considered to be impervious which do not allow the flow of moisture through them which slows down the flow in concrete. The ITZ around the aggregates have higher transport properties and accelerate the flow. A numerical framework is formulated in which spherical aggregates are projected on lattice elements which are assigned transport properties pertaining to the phase of concrete they represent. Moisture transport is simulated and compared with experimental results where an increase of volume fraction of aggregates leads to a decrease of cumulative saturation level. The influence of ITZ is observed on a local and global scale by varying the ITZ diffusivity coefficient. The moisture simulation is also modelled on a finer mesh to see the effect of mesh refinement. Finally, moisture transport is simulated in a numerical model that combines a planar crack along with heterogeneities and flow is observed at different time stages.

Multiscale computational homogenization is an efficient method to upscale the microstructural behavior of micro-heterogeneous materials. In this method, a representative volume element (RVE) is assigned to a macroscale material point and the constitutive law for the macroscopic model at that point is obtained by solving a boundary value problem on the RVE. Among the conventional boundary conditions, the so-called strong periodic boundary conditions tend to converge faster towards the actual microstructural response. Nonetheless, applying strong periodic boundary conditions to a batch of 48 fiber-matrix RVEs under uniaxial load with varying orientations introduces a dependency between the average ultimate principal stress (σ₁) and the orientation angle (θ).

This treatise investigates the effects on this dependency by applying so-called weak periodic boundary conditions instead. These boundary conditions soften the strong requirement of periodicity of displacements at the RVE boundary by coarsening the traction mesh and requiring periodicity to hold only in an average sense over this coarser mesh. Three main questions are asked: Do weak periodic boundary conditions alleviate the dependency between σ₁ and θ? Is this dependency the result of RVEs prone to localization under shear? Do smaller coarsening factors widen the range of θ over which localization with a single shear band is permissible? Overall, it is concluded that only the weakest form of weak periodic boundary conditions reduces the dependency between σ₁ and θ, which is indeed caused by RVEs that are prone to strain localization under shear, particularly towards θ = 45^o. Increasing coarsening factors quickly introduce stricter periodicity requirements, thus limiting the possibility of strain localization with a single shear band at angles other than θ = 45^o. Recommendations are provided to alleviate the dependency between σ₁ and θ as well as how to more realistically model the behavior of RVEs used in this research.

Auxetic materials are increasingly used in the field of sports, defense, bio-medicine, and acoustic-filtering due to their unconventional behavior of exhibiting a negative Poisson's effect. The prospective use of auxetics to mitigate shock and impact forces have proven the relevance of research on auxetics. It becomes of paramount importance to understand their behavior and mechanics of deformation, especially regarding their response to dynamic loading conditions. Although auxetics have caught attention from several researchers, a detailed overview of computational modeling of auxetics is still missing and the deformation behavior of auxetics subjected to different loading conditions is an unfolded topic in structural mechanics.

In this thesis, computational tools are developed and validated to investigate the effect of auxeticity in a linear elastic continuum and a re-entrant honeycomb structure under different static and dynamic loading conditions. The possibility to assess the global behavior of an assembly of re-entrant honeycomb unit cells by the analysis of an equivalent continuum model with the same global dimensions and homogenized mechanical properties has been examined. The dynamic analysis of a continuum model of an isotropic auxetic material presented the influence of auxeticity in increasing the ratio of shear wave velocity to dilatation wave velocity. The finite element analysis of a simply supported auxetic beam subjected to impact at mid-span affirmed the hypothesis that in auxetics, the material flows towards the point of impact, thus increasing the indentation resistance. Computational challenges in analyzing a beam impact problem are discussed through the comparison of results from the standard finite element analysis, modal analysis, and the B-Bar method. It is recommended to further develop the proposed models to have a more robust approach on analyzing the continuum beam impact problem.

The assembly of a re-entrant honeycomb structure used in the finite element simulations were modelled with different unit cells by the variation of the two dominant geometrical parameters of the unit cell members. The results highlighted the strong correlation between the effective mechanical properties and the geometrical parameters, therefore indicating the possibility of achieving the desired values of the Poisson's ratio by simply modifying the necessary geometrical parameters. The results also showed that it is possible to obtain a Poisson's ratio more negative than -1 which can be exploited to design a displacement amplifier especially suitable for defense applications. The close comparison of the analytical and the computational results provided a groundwork to extend the numerical study to geometrical nonlinear analysis and other loading conditions like bi-axial loading and bending at the structural level. The comparison of results obtained from the analysis of an assembled cellular structure and an equivalent continuum model subjected to bi-axial loading and bending demonstrated the possibility of assessing the global behavior of cellular structure using the continuum model.

Dynamic analyses of a re-entrant honeycomb structure revealed that the wave propagation phenomenon in a re-entrant honeycomb structure is a result of several reflections, transmissions, mode conversions and superposition of waves occurring at the connection of cell members. The proportion of reflection, transmission and mode conversion of incident wave when it passes the connection can be controlled by varying the re-entrant angle between the cell members. For the lower frequency regime, the speed of wave propagation in re-entrant structures can be estimated using the continuum approach.

Furthermore, in a secondary study, the influence of the geometrical parameters of a re-entrant honeycomb structure to exhibit the stop-pass band was examined. Recommendations are provided for the further development of computational tools to analyze the three-dimensional and disordered re-entrant structures. In addition, it is also suggested to investigate the beam impact problem in the cellular model, the effect of rigidity of connection of cell members and research the possibility of having an isotropic re-entrant honeycomb structure as the present model is orthotropic in nature.

Epoxy resins are increasingly used in critical structural components with widespread applications in the transportation, construction and energy industries. The wind energy sector is one of the fastest growing commercial markets for epoxy resins, meaning that the structural behaviour of epoxy resins is becoming a key area of research, especially regarding its application to wind turbines. Wind turbines, particularly those in off-shore installations, are subject to a wide range of environmental conditions, most notably large variations in humidity and temperature. Both moisture and increased temperatures have been observed to have a significant impact on the stiffness and strength of epoxy resins. These environment effects, coupled with complex time dependent mechanical behaviour, means that the accurate prediction of the structural performance of epoxy resins has not yet been fully described.

This thesis presents a multiphysics framework for the simulation of hygrothermal ageing in epoxy resins. The constitutive model formulated in this thesis consists of a non-linear viscoelastic and viscoplastic mechanical model, physically coupled with a Fourier heat conduction model and a Fickian diffusion model. Degradation based on a glass transition surface is implemented to describe the multi-state behaviour of epoxy resins. To justify the model assumptions, DMA and creep tests are performed on epoxy resin specimens and their temperature dependent mechanical behaviour is established. A number of numerical benchmark tests and case studies are performed using a finite element implementation of the numerical framework. It is shown that the multiphysics framework can capture the characteristic mechanical and hygrothermal ageing behaviour exhibited by epoxy resins. Recommendations are provided for further development of the numerical model and calibration of the material properties. In a secondary study, a mesh sensitivity analysis is performed on an existing viscoelasitc-viscoplastic-damage model and recommendations for an improved formulation are provided.

Numerical modelling of pile foundations can be done in several ways. In commercial finite element packages two main options are available; using volume elements with interface elements between the pile and soil domains and the embedded beam approach. The embedded beam element was first proposed by Sadek and Shahrour (2004) and considers a beam element that can cross a solid element at any arbitrary location with any arbitrary inclination. This has several advantages to the volume pile method, such as the need for fewer elements and the mesh uncoupling of the pile and soil, which make this method much more efficient and leads to a significant reduction in calculation time. However, the embedded beam element also deals with a number of limitations and drawbacks. This research focuses on overcoming the mesh sensitivity, which is caused by the stress singularity that is introduced in the soil by the beam element. Also, the inability to take into account the pile surface will be resolved, aiming to improve the lateral pile-soil interaction.

The idea of Turello et al. (2016) of an embedded beam element with explicit interaction surface is extended and generalised leading to a new embedded beam formulation. In the proposed model the beam displacements at the interaction surface are obtained by a mapping scheme that takes into account Timoshenko beam theory and which is generalised to model inclined piles as well. A constitutive equation that describes the relation between the interface stresses and relative displacements between the pile and soil is defined along the shaft and at the foot of the pile. Along the shaft of the pile a shear stress limit is defined based on the Mohr-Coulomb failure criterion in order to incorporate plasticity in lateral direction. Furthermore, a more practical and efficient assembly procedure is proposed.

Validation of the proposed method proofs that the proposed model leads to a significant mesh sensitivity reduction in case of axially loaded models compared to the existing implementation. The overall response of laterally loaded piles is improved considerably as well. However, the proposed method is still unable to capture lateral interface behaviour in order to model soil slippage around the pile. Furthermore, it is recommended to formulate a generally applicable foot interface stiffness and to optimise the code in order to reduce the computation time. The description of the interaction surface opens up many new possibilities for future research, such as modelling the true cross-section shape.

Ceramic is a hard and brittle material with little ductility. It has wide applications in diverse industries viz. metallurgy, atomic energy, electronics, communication, space, military, insulation, biomechanical appliances etc. It is preferred material for armour protection due to its high strength-weight ratio. Its plastic deformation and failure behavior have, therefore, become a subject of extensive research in the recent past. Consequently, several material models were formulated in the second half of the nineteenth century by researchers such as
Johnson- Holumquist, Rajendran-Grove, Deshpande-Evans etc. to study ceramic response under high-velocity impact. But these models are complex and require extensive calibration, while the Drucker-Prager (DP) Model is easy to implement. It was developed for study in soil and rock and is chosen to analyze ceramic quasi-plastic and tensile behaviour as part of research through the current thesis.
Tensile ring and cone cracks, first observed by Hertz, are one of the modes of failure in brittle materials like ceramic. The prime objective of the thesis is the study of ceramic failure as well as cone crack initiation and its propagation through the ceramic body under the influence of varying material parameters such as cohesion, friction angle, dilatancy and softening of ceramics besides confinement by indentation. The Drucker Prager model is employed to study
indentation by simulation through numerical methods in JEM JIVE FEM library. Verification of the model was done through simulation of unit cubes subjected to unidirectional stresses at prescribed displacements. Suitable modification for pressure dependent softening behaviour of the ceramic is also made into the model and simulations undertaken in order to get insight into the nonlinear strength degradation of the ceramic post-elastic limit. Since the DP criterion
tends to overestimate material strength, a suitable mechanism to limit the material strength is integrated with DP yield function and also to facilitate the comparison of results arrived through DP and modified yield functions.
The results obtained by a simulated indentation in accordance with the DP formulation suggest that ceramic having high dilatancy, low friction angle and small softening modulus under confined conditions is more suited for use as armor protection. The pressure dependent softening behavior of ceramic is favorable for it being a good armor protection material. The DP Yield function modified to limit the material strength to a finite value, did not have any significant impact on crack initiation though the zone of compressive plastic strain grew in size during indentation. The DP based numerical model suffered from mesh sensitivity. The introduction of viscoplasticity to the numerical model was observed to have a positive influence in mitigating mesh sensitivity.

In an accidental explosion of ammunition magazine, concrete debris throw hazard is a major safety concern. The current approach to collect explosively-damaged debris after test does not yield representative test data on debris launch condition. This research is initiated to study the possibilities of using polymeric foam material to soft-catch concrete debris, without inducing additional damage. Preliminary design of the soft-catcher are first performed using a modified analytical shock wave model. Numerical analyses using LS-DYNA are subsequently carried out to verify the analytical model assumptions and account the effect of rigid end backing of the soft-catcher. Following then, the debris penetration problem is escalated into 2-D space domain, in order to include the influence of tearing resistances within the soft-catcher. The observed phenomena are eventually translated into design parameters for the soft-catcher.

This master thesis is built up around two questions. First: Is it possible to calibrate a ductile failure model, where the strain at failure is a function of stress triaxiality and Lode angle, using only a single SENB specimen? The failure model that is referred to describes local failure in a finite element model to simulate ductile fracture. In this thesis, an orthotropic damage constitutive material model is constructed to accommodate the most realistic fictitious crack growth. Simple isotropic element erosion or deletion was found to have severe effects on the stress states of neighbouring elements due to a major loss of constraint. The damage model is based on the stress triaxiality and Lode angle, derived from the invariants of the stress tensor. For any combination of stress triaxiality and Lode angle, an equivalent plastic strain to failure is required in the form of a three-dimensional failure surface. Typically, multiple experiments with varying dominant stress states are needed to accurately calibrate such a failure surface or model. This research project shows that one can not only calibrate a failure surface from a single three-point bending specimen, but also obtains more calibration points than other approaches that are typically used for this purpose along the way. These calibration points are collected during a fully automated iterative procedure, where points are added every time the force in the simulation exceeds the force found in the experiment by more than 1%. As soon as enough calibration points with distinct stress states were found, the Hosford-Coulomb model was fitted using the least squares method. Finally, the obtained failure model was verified by using it to simulate the specimen on which it was calibrated. From these results, it can be concluded that it is possible to calibrate a failure model using only a single SENB specimen. In addition to this thesis, this procedure is outlined in a paper (Coppejans & Walters, 2017). The second part of the thesis revolves around whether a ductile failure model obtained from the procedure presented in this thesis can be successfully used to predict failure in different geometries. To verify this, the differences in stress states that occur in specimens with different dimensions and aspect ratios is researched both theoretically and numerically. From that study, it is found that indeed the geometry has an influence on the stress states that are present and where they are present. This influence is great enough that simple fracture mechanics approaches such as the J-Integral and CTOD are only transferable if strict requirements are met between SENB specimen from the same material but with different dimensions. Standards such as BS7448 and ASTM E1820 describe the validity of the experimentally obtained fracture toughness parameters with respect to the specimen size to deal with these requirements. Where the global approach (fracture mechanics) fails, the local approach (damage mechanics) finds a natural application due to its revolving around stress states. Application of the failure model obtained from the smallest specimen to the specimen with greater size and to the specimen with a different aspect ratio strengthens this claim. The failure model is able to predict the onset of element failure, used to simulate crack growth, within 2.5% deviation in experimental and simulated force at crack initiation. The force during crack propagation is simulated with 99% accuracy for crack mouth opening diameters (CMOD) at least twice and up to four times the CMOD due to plastic deformation. For greater amounts of CMOD, the reaction force is simulated with decreasing accuracy. The crack path on the free surface is verified by comparing the simulated crack path, visualised by removing all damaged elements from view, to a picture that was taken after the experiment. This reveals that while the extent of the crack is fairly accurately simulated, the width of the crack is both not accurately represented by simply removing all damaged elements from view, and the created damage model may require more development.