Fault diagnosis based on discrete wavelet transform and ART2 neural network

In this paper we proposed a method of fault diagnosis for dynamic system based on DWT(discrete wavelet transform) and ART2 NN (adaptive resonance theory 2 neural network). In the proposed method, the fault is detected when the errors between the system output and the nominal system output cross a predetermined threshold. Once a fault in the system is detected, the ART2 NN fault classifier isolates the fault. The algorithm contains three main steps: a fault detection part by threshold test, a data preprocessing part via DWT and a fault isolation part by fault classifier. The simulation results demonstrate the effectiveness of the proposed fault diagnosis method based on DWT and ART2 NN.