Fault Diagnosis of Fan Based on Fuzzy Neural Expert System

This paper combines neural network with expert system organically, adopts fuzzy membership function to calculate fuzzy relationship matrix to reduce the neural network algorithm´s dependence on training data, and on this basis, a new fashioned fuzzy neural expert system of fan is established. The system uses modular design and provides examples of diagnosis. The results show that the system performances high reasoning efficiency, good reliability, and satisfies the effectiveness and practicality of fault diagnosis of fan as well. The study can be extended to fault diagnosis of other rotating machinery.