Neural Fault Diagnosis and Fuzzy Fault Control for a Complex Dynamic System

Fault diagnosis has become an issue of primary importance in modern process automation as it provides the prerequisites for the task of fault detection. The ability to detect the faults is essential to improve reliability and security of a complex control system. Parameter estimation methods, state observation schemes, statistical likelihood ratio tests, rule-based expert system reasoning, pattern recognition techniques , and artiicial neural network approaches are the most common methodologies developed actively during recent years. In this paper, we describe a completed feasibility study demonstrating the merit of employing pattern recognition and an artiicial neural network for fault diagnosis through back propagation learning algorithm and making the use of fuzzy approximate reasoning for fault control via parameter changes in a dynamic system. As a test case, a complex magnetic levitation vehicle (MLV) system is studied. Analytical fault symptoms are obtained by system dynamics measurements and the classiication is carried out through a multilayer feed-forward network. The neural network is rst taught the diierent fault situations through training patterns. After the network is trained, it achieves an overall classiica-tion accuracy of 99.78% for a disturbance-free MLV model and 91.4% for a model with track disturbance irregularities. Proper actions are performed based on fuzzy reasoning of knowledge base results in a normal process operation recovered.