| Summary: | Technology advances continue to revolutionise military equipment. The development of new firepower induces an interest in the enhancement of protection gear, both for transportation vehicles and personnel. There has been a significant amount of research of methods to increase protection capabilities without increases in the weight of a given defence system. This dissertation seeks to develop an optimisation tool that results in light-weight armour plates without compromising protection capabilities. A thorough study on the propagation of elastic and plastic stress waves aims for a better understanding of how an armour system behaves upon ballistic impact. The first part of this dissertation focuses on the development of a Python script that provides an efficient approach to model generation in Abaqus. It enables the user to avoid time consuming actions when designing ballistic test models to later simulate through the software. This script is also used to validate the theory behind elastic and plastic stress wave propagation while also being able to access output databases and interpret obtained results. The importance of the script is relevant for the second part of the dissertation, which takes advantage of the Abaqus Python Application Programming interface (API) to perform optimisation procedures automatically. Focusing particularly on the application of the particle swarm optimisation algorithm, this work continuously improves the efficiency and accuracy of the mentioned algorithm by dividing three different optimisation problems into several experiments. Each one of the experiments is carefully defined to highlight the impact of a specific operating parameter of the algorithm. A validation of the stress wave propagation and how it is affected upon contact with layered media is carefully conducted through a series of different analysis approaches. It is shown that the plastic stress wave propagates slower than the elastic one and that plastic deformation affects the properties of the generated stress wave, such as wavelength. The implemented particle swarm optimisation algorithm proved to be an effective approach to problem solving, however, for complex problems the operational parameters must be carefully chosen.
|
|---|