| This is a practical subject co-developed by Northeastern University Equipment Diagnosis Center and military Region to meet the need of reality.In this paper, an algorithm using wavelet neural network for fault diagnosis in analog circuit witholerance is proposed, and is simulated on the direct circuit and the alternating circuit andreceives better result.Wavelet neural network is the combination of Wavelet Analysistheory and Artificial Neural Networks theory. It not only possesses the wavelet transform's time-frequency localizationability, but also makes full use of the property of self-learning of Artificial Neural Networks, which enables Wavelet neural network to have better ability of approximation and better fault-tolerance capacity. Because of its particular merits, Wavelet neural network is widely used in many fields, such as signal processing, functionapproximation, data forecast, system identification and patternrec-ognition.At last, corresponding programs are designed for training the general BP neural network and wavelet neural network by MATLAB, LABVIEW and simulation analysis proved that wavelet neuralnetwork is adaptive in complex pattern classifying.In this paper, some results are obtained, but there is also some difference from thepractical application, and it still should be studied in the future. |
