Analog circuit fault diagnosis method based on CPSO-SVM

To realize feature extraction and fault signal recognition efficiently in analog circuit fault diagnosis, a novel diagnosis method called CPSO-SVM is proposed. In this method, Chaos theory and Particle Swarm Optimization (PSO) are combined to form a new approach called CPSO and this approach is applied to the optimization of SVM parameters. A fault diagnosis experiment on a certain filter circuit indicates that the proposed diagnosis method is feasible and efficient.