Machine Learning Unsupervised Methods in the Design of an On-board Health Monitoring System for Satellite Applications

The dissertation starts by providing a description of the phenomena related to the increasing importance recently acquired by satellite applications. The spread of such technology comes with implications, such as an increase in maintenance cost, from which derives the interest in developing advanced techniques that favor an augmented autonomy of spacecrafts in health monitoring. Machine learning techniques are widely employed to lay a foundation for effective systems specialized in fault detection by examining telemetry data. Telemetry consists of a considerable amount of information; therefore, the adopted algorithms must be able to handle multivariate data while facing the limitations imposed by on-board hardware features. In the framework of outlier detection, the dissertation addresses the topic of unsupervised machine learning methods. In the unsupervised scenario, lack of prior knowledge of the data behavior is assumed. In the specific, two models are brought to attention, namely Local Outlier Factor and One-Class Support Vector Machines. Their performances are compared in terms of both the achieved prediction accuracy and the equivalent computational cost. Both models are trained and tested upon the same sets of time series data in a variety of settings, finalized at gaining insights on the effect of the increase in dimensionality. The obtained results allow to claim that both models, combined with a proper tuning of their characteristic parameters, successfully comply with the role of outlier detectors in multivariate time series data. Nevertheless, under this specific context, Local Outlier Factor results to be outperforming One-Class SVM, in that it proves to be more stable over a wider range of input parameter values. This property is especially valuable in unsupervised learning since it suggests that the model is keen to adapting to unforeseen patterns.

Fatigue Design Challenges of Transition Pieces from Decommissioned Platforms for Offshore Wind Energy

All structures are subjected to various loading conditions and combinations. For offshore structures, these loads include permanent loads, hydrostatic pressure, wave, current, and wind loads. Typically, sea environments in different geographical regions are characterized by the 100-year wave height, surface currents, and velocity speeds. The main problems associated with the commonly used, deterministic method is the fact that not all waves have the same period, and that the actual stochastic nature of the marine environment is not taken into account. Offshore steel structure fatigue design is done using the DNVGL-RP-0005:2016 standard which takes precedence over the DNV-RP-C203 standard (2012). Fatigue analysis is necessary for oil and gas producing offshore steel structures which were first constructed in the Gulf of Mexico North Sea (the 1930s) and later in the North Sea (1960s). Fatigue strength is commonly described by S-N curves which have been obtained by laboratory experiments. The rapid development of the Offshore wind industry has caused the exploration into deeper ocean areas and the adoption of new support structural concepts such as full lattice tower systems amongst others. The optimal design of offshore wind support structures including foundation, turbine towers, and transition piece components putting into consideration, economy, safety, and even the environment is a critical challenge. In this study, fatigue design challenges of transition pieces from decommissioned platforms for offshore wind energy are proposed to be discussed. The fatigue resistance of the material and structural components under uniaxial and multiaxial loading is introduced with the new fatigue design rules whilst considering the combination of global and local modeling using finite element analysis software programs.