Student Number : 9801145J - MSc dissertation - School of Electrical and Information Engineering - Faculty of Engineering / In this document, a paper is presented that reports on the optimisation of a system that assists in time series prediction. Daily closing prices of a stock are used as the time series under which the system is being optimised. Concepts of machine learning, Artificial Neural Networks, Genetic Algorithms, and Agent-Based Modeling are used as tools for this task. Neural networks serve as the prediction engine and genetic algorithms are used for optimisation tasks as well as the simulation of a multi-agent based trading environment. The simulated trading environment is used to ascertain and optimise the best data, in terms of quality, to use as inputs to the neural network. The results achieved were positive and a large portion of this work concentrates on the refinement of the predictive capability. From this study it is concluded that AI methods bring a sound scientific approach to time series prediction, regardless of the phenomena that is being predicted.

Current European legislation aims to limit vehicle noise emissions since many people are exposed to road traffic noise in urban areas. Vehicle pass-by noise is measured according to the international standard ISO 362 in Europe. More recent investigations of urban traffic have led to the proposal of a revised ISO 362 which includes a constant-speed test in addition to the traditional accelerated test in order to determine the pass-by noise value. In order to meet the legal pass-by noise requirements, vehicle manufacturers and suppliers must analyse and quantify vehicle noise source characteristics during the development phase of the vehicle. In addition, predictive tools need to be available for the estimation of the final pass-by noise value. This thesis aims to contribute to the understanding of vehicle pass-by noise and of the characteristics of the vehicle noise sources contributing to pass-by noise. This is supported through an extensive literature review in which current pass-by noise prediction methods are reviewed as well. Furthermore, three vehicle noise sources are replicated experimentally under laboratory conditions. This involves an orifice noise source, represented by a specially designed loudspeaker on a moving trolley, shell noise, represented by a metal cylinder structure, and tyre cavity and sidewall noise, represented by an annular membrane mounted on a tyre-like structure. The experimentally determined directivity characteristics of the acoustically excited noise sources are utilised in the pass-by noise prediction method. The predictive results are validated against experimental measurements of the three vehicle-like noise sources made within an anechoic chamber.