Reliability enhanced EV using pattern recognition techniques

Abstract

This is a copy of the author's accepted version of a paper subsequently published in the proceedings of 2014 IEEE International Electric Vehicle Conference (IEVC). IEEE, pp. 1-8. It is available online at:http://dx.doi.org/10.1109/IEVC.2014.7056199The WestminsterResearch online digital archive at the University of Westminster aims to make the research output of the University available to a wider audience. Copyright and Moral Rights remain with the authors and/or copyright owners.© 2014 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.Whilst further distribution of specific materials from within this archive is forbidden, you may freely distribute the URL of WestminsterResearch: (http://westminsterresearch.wmin.ac.uk/). Abstract-The following paper will contribute to the development of novel data transmission techniques from an IVHM perspective so that Electrical Vehicles (EV) will be able to communicate semantically by directly pointing out to the worst failure/threat scenarios. This is achieved by constructing an image-based data communication in which the data that is monitored by a vast number of different sensors are collected as images; and then, the meaningful failure/threat objects are transmitted among a number of EVs. The meanings of these objects that are clarified for each EV by a set of training patterns are semantically linked from one to other EVs through the similarities that the EVs share. This is a similar approach to wellknown image compression and retrieval techniques, but the difference is that the training patterns, codebook, and codewords within the different EVs are not the same. Hence, the initial image that is compressed at the transmitter side does not exactly match the image retrieved at the receiver's side; as it concerns both EVs semantically that mainly addresses the worst risky scenarios. As an advantage, connected EVs would require less number of communication channels to talk together while also reducing data bandwidth as it only sends the similarity rates and tags of patterns instead of sending the whole initial image that is constructed from various sensors, including cameras. As a case study, this concept is applied to DC-DC converters which refer to a system that presents one of the major problems for EVs.