New method of analog circuit fault diagnosis using fuzzy support vector machine

Authors:
  • JTJingyuan Tang
    Jingyuan Tang
  • YSYibing Shi
    Yibing Shi

Abstract
In order to solve the problem of correctly identifying fault classes in analog circuit fault diagnosis and improve classification ability, a novel fault diagnosis method based on fuzzy support vector machine (FSVM) is proposed in this paper. Firstly, the fault feature vectors are extracted by joint time-frequency domain feature extraction method and the fuzzy membership of the feature vectors is computed by a novel proposed membership function to overcome the sensitivity to noise and outliers. Then, after training the FSVM by faulty feature vectors, the FSVM model of the circuit fault diagnosis system is built. Finally, test samples’ feature vectors are input into the trained FSVM model to identify the different fault cases. The simulation results for analog and mixed-signal test benchmark CTSV filter circuits demonstrate that the proposed method can not only correctly classify the single fault classes but also the multiple fault classes with a highly average classification success rate 98.2%. The method develops a new direction for the fault diagnosis of analog circuit.

This publication has 7 references indexed in Scilit: