Research on fault diagnosis based on fuzzy neural network

A method for fault diagnosis of Aircraft Subsystem based on the fuzzy neural network (FNN) is put forward. To automatically acquire the fuzzy rule-base and the initial parameters of the fuzzy model, the improved method based on fuzzy c-means clustering algorithm is used in structure identification. Then a initial FNN is constructed to match with the fuzzy model. The FNN is trained by its learning algorithm to obtain a precise fuzzy model and realize parameter identification. The model of aircraft subsystem fault diagnosis is set up by using the measured data at the fixed-check state on the ground as learning samples. Finally as testing and analysising the fault diagnosis model, the results show that this method has the advantages of anti-noise, anti-sensitive and higher diagnosing precision.