Title :
Fault diagnosis method of non-linear analog circuits based on Volterra series and SVM
Author :
Huafeng He ; Yicheng Chen ; Zheng Yang ; Huixuan Deng
Author_Institution :
Second Artillery Eng. Univ., Xi´an, China
fDate :
May 31 2014-June 2 2014
Abstract :
A new fault diagnosis method of analog circuits is proposed in this paper. The method is based on Volterra series and SVM. This paper introduces the basic theory of Volterra series and SVM and deduces the proposed identification method in detail. The identification method is applied to the fault diagnosis of a non-linear analog circuit. The simulation result shows that the proposed method can obviously improve the fault diagnosis rate compared with the traditional method.
Keywords :
Volterra series; analogue circuits; fault diagnosis; support vector machines; SVM; Volterra series; fault diagnosis method; identification method; nonlinear analog circuits; support vector machine; Analog circuits; Circuit faults; Fault diagnosis; Feature extraction; Frequency-domain analysis; Kernel; Support vector machines; Fault Diagnosis; SVM; Volterra Series;
Conference_Titel :
Control and Decision Conference (2014 CCDC), The 26th Chinese
Conference_Location :
Changsha
Print_ISBN :
978-1-4799-3707-3
DOI :
10.1109/CCDC.2014.6852729
