Hyperspectral imaging has a variety of commercially important applications in Earth observation because of its advanced functionalities that enable precise material identification. This thesis presents the structural design and analysis of a 3U CubeSat carrying a hyperspectral imaging payload, as part of the Earth Moon Education CubeSats mission by EuroMoonMars. Due to its high resource demands, hyperspectral imaging requires an efficient design of the structural subsystem to accommodate the payload and other subsystem components, but publicly available research on this is extremely limited. A CAD model of the complete CubeSat was first designed in Autodesk Fusion based on requirements from the dispenser. For every design version, structural FE analyses were performed in ANSYS Workbench to ensure that the CubeSat withstands the vibrational loads from the launch vehicle and the thermoelastic response from the thermal loads in orbit. After simplifying the geometry of the CAD model to reduce the computational effort, and applying appropriate loads and constraints, the analyses were run on a multi-core processor. The design of the CubeSat was finalised when the FE analyses resulted in positive margins of safety against yielding. Sensitivity analyses on uncertain parameters were conducted in optiSLang, and the results of these will be used for validating and updating the FE analyses, to be carried out in the next phase of the mission by testing a physical model of the CubeSat in dedicated setups.

This thesis presents the structural and thermal design of a CubeSat-based rover system for lunar surface exploration, developed with standard 3U CubeSat as foundation. The study focuses on the integration of a mobility platform, capable of being deployed from a lunar lander, on to the 3U CubeSat. Detailed finite element analysis (FEA) was conducted to ensure structural survivability under launch conditions, including quasi-static, random vibration, and shock loads. The first natural frequency was found to be 267 Hz, exceeding the 115 Hz threshold across major spacecraft launchers. Maximum von Mises stress reached 300 MPa under the worst-case shock loading, yielding a minimum margin of safety of 0.10. Thermal analysis using passive insulation strategies demonstrated that all subsystem components remained within -20°C to +65°C under lunar surface conditions. Mobility performance was validated through terramechanics modelling and field testing on Earth's sandy terrain, where the prototype exhibited static and dynamic sinkage within 11‰ and 9‰ of analytical predictions, respectively. These results demonstrate that with appropriate design strategies and material selection, a CubeSat-Based Rover configuration can be feasibly integrated within a 3U CubeSat. While further studies—such as radiation tolerance, dust mitigation, lunar analogue terrain testing, and structural testing—are required, the findings highlight the potential of this architecture as a standardized, low-cost platform for short-duration lunar surface missions.
Dataset: https://doi.org/10.4121/uuid:6bd1d545-e071-4f7b-a457-e213909b878f

This thesis investigates machine learning approaches for predicting the glass transition temperature of polymeric materials using chemical structure representations. Multiple molecular encoding strategies were evaluated, including character‑level tokenization of SMILES strings with count vectorization and TF‑IDF weighting, as well as structural fingerprints such as Morgan and MACCS keys. These representations were applied to a range of models, including Support Vector Regression, Random Forests, Gaussian Process Regression, and deep learning architectures such as Artificial Neural Networks (ANNs) and Convolutional Neural Networks (CNNs). Bayesian optimization via Optuna was employed to tune hyperparameters for optimal performance. Experimental results show that Random Forests combining tokenized SMILES with numerical descriptors achieved competitive predictive accuracy, while fingerprint‑based models provided interpretable and computationally efficient alternatives. The study highlights the trade‑offs between representation richness, interpretability, and prediction, offering insights into the design of data‑driven pipelines for polymer property prediction in aerospace and materials engineering applications.

Protecting sensitive objects during a rocket launch is imperative. For equipment going to the ISS this is done through packing them in polymer foam or bubble wrap. Simulating how well an object is packaged is computationally intensive and difficult to implement. To solve this, this thesis examines the performance of three surrogate models, LSTM, LSTM-FC-GP, and LSTM-PCA-GP, to predict the response of a foam packed object when it is subjected to a vibration. The results show that the models can accurately predict the response of the system in both the time and frequency domain. Furthermore, the LSTM models that are augmented with a GP can accurately predict the uncertainty of the system in addition to having better performance.

The W-type is a motion-compensated offshore crane capable of hoisting up to 2 mt of cargo, thereby supporting the maintenance processes for the rapidly growing offshore wind farms. However, the rated cargo mass is limited by the dynamic amplification factor (DAF), which is a governing parameter in the structural design of offshore cranes. During the design phase, the industry currently makes use of a one-dimensional spring-mass model to estimate the DAF. In this study, a simplified flexible multibody dynamic model was developed, and validated with experimental data, to simulate the DAF instead. It was found that the previous method underpredicts the DAF by up to 56%, because it neglects the effects of crane inertia and structural damping. Besides, the active controller of the luffing cylinders was modelled for the first time, which reduced the DAF up to 13%. Therefore, the simple spring-mass model is not deemed accurate for this application.