USING A NEURAL FUZZY SYSTEM TO EXTRACT HEURISTIC KNOWLEDGE OF INCIPIENT FAULTS IN INDUCTION-MOTORS .2. APPLICATION

The use of electric motors in industry is extensive. These motors are exposed to a wide variety of environments and conditions which age the motor and make it subject to incipient faults. These incipient faults , if left undetected, contribute to the degradation and eventual failu re of the motors. Part I of this paper introduced a hybrid neural/fuzz y fault detector to perform fault detection tasks. Part I also discuss ed the purpose and methodology for combining the technologies of artif icial neural networks and fuzzy logic for fault detection applications . This paper uses the hybrid neural/fuzzy fault detector to solve the motor fault detection problem. As an illustration, the neural/fuzzy fa ult detector will be used to monitor the condition of the motor bearin g and the stator winding insulation. The initialization and training o f this fault detector is in accordance with the procedures outlined in Part I of this paper. Once the neural/fuzzy fault detector is trained , the detector not only can provide accurate fault detector performanc e, but can also provide the heuristic reasoning behind the fault detec tion process and the actual motor fault conditions. With better unders tanding of the heuristics through the use of fuzzy rules and fuzzy mem bership functions, we can have a better understanding of the fault det ection process of the system, thus we can design better motor protecti on systems.