Wavelet-transformation-based advanced state chi-square test method and its application

In view of the problem that traditional state chi-square test can't accurately determine the fault duration in soft fault detection, an improved state chi-square test method based on wavelet-transform was presented. Firstly, the signal jumping point was identified by using the decay speed of wavelet transform modulus maxima on the different scales. Then, the detection results were feedback to the Kalman filter. By introducing state correction term to correct the state error, the soft fault duration can be determined by the curve of fault detection. The simulation results show that the proposed method solve the problem of soft fault duration, and greatly improve the accuracy of fault detection.