Skip to Main Content
A fault diagnosis method for analog circuits based on hierarchical support vector machine (HSVM) and Dempster-Shafer (D-S) theory is developed in this paper. Firstly, output voltage signals from the test nodes are obtained from analog circuits test points and the fault feature vectors are extracted from Haar wavelet transform coefficients. Then, after training the HSVM by faulty feature vectors, the HSVM model of the circuit fault diagnosis system is built. Finally, combing D-S theory with probabilistic interpretation of SVM scores estimate confidences over the prediction, which improves diagnosis results. Simulation results of diagnosing a four op-amp biquad high-pass filter circuit have confirmed the validity of the proposed technique.