Analog circuit fault diagnosis using Ada Boost and SVM

A fault diagnosis method for analog circuits based on Support Vector Machine (SVM) and AdaBoost algorithm 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 packet transform coefficients. Then, after training the AdaBoost SVM by faulty feature vectors, the SVM ensemble model of the circuit with tolerances fault diagnosis system is built. Simulation results of diagnosing a two stage four op-amp biquad low-pass filter circuit, compared with several existent fault diagnosis methods, show us the proposed technique has the highest classification accuracy and have confirmed the validity of the method.