Intelligent Control Strategies for Nonlinear Systems with Faults

点击翻译按钮获取中文摘要

In order to protect the public health and safety, there is substantial need for the reliable control and diagnosis of mechanical and electrical systems. This research program aims to develop intelligent controllers and strategies that can be used by a number of different engineering systems, including cars, planes, robots, and unmanned aerial systems, to name a few. A long-term research goal of this program is to provide techniques for a system to enable it to successfully predict the occurrence of faults and take appropriate control actions prior to the fault event. In other words, provide a system with some form of cognition. This offers a number of benefits to society (safety) as well as the economy (reduced maintenance costs, fewer catastrophic events, and less downtime). The proposed activities are important to the advancement of knowledge and understanding within the field. In addition, this work will have a positive impact in a number of industries, including: automotive, aerospace, defense, manufacturing, and space systems.*** My research group studies intelligent control and estimation methods. We seek to advance the use of machine learning (artificial neural networks) and Bayesian (probability-based) techniques on engineering control and estimation problems. Inherent to intelligence is the ability to remember the past (memory) and predict the future (estimation) based on a knowledge base (built through human training or system interactions) and the external environment (measurements and senses). Sometimes referred to as ‘cognitive', intelligent controllers are able to learn (adapt) over time and are able to predict and anticipate the future environment. Intelligent systems are important to protect the public health and safety. In addition to technical advancements, my research group trains highly-qualified personnel who help meet Canada's demand for engineers and will contribute to its economic success.