Fault Diagnosis Method of Analog Circuit Based on CEEMD-ELM

An analog circuit fault diagnosis method based on complementary ensemble empirical mode decomposition and extreme learning machine (CEEMD-ELM) algorithm is proposed. The test signals of the analog circuit were decomposed into multiple intrinsic mode functions (IMF), and then the features of each IMF, such as relative entropy, kurtosis, RMS, margin factor, were obtained to establish the fault feature vectors. The established fault feature vectors were used as input to train the ELM to realize the fault diagnosis of the analog circuit. Simulation results show the effectiveness and accuracy of the proposed method.