An Analog Circuit Fault Diagnosis Method Based on Electromagnetism-like Mechanism Algorithm

摘要

An analog circuit diagnosis method based on BP neural network optimized by electromagnetism-like mechanism algorithm was proposed for characteristics of the analog circuit. Firstly, fault features are extracted by using wavelet packet. Then, feature vectors are input to the BP neural network optimized by using electromagnetism-like mechanism algorithm. The method avoids parameters prematurely trapped in local optima, improves the classification capabilities neural network. The simulation and experiment results demonstrate the effectiveness of the proposed method.