A new analog circuit fault diagnosis approach based on GA-SVM

Shaowei Chen, Shuai Zhao, Cong Wang

Research output: Contribution to book/anthology/report/conference proceedingArticle in proceedingResearchpeer-review

7 Citations (Scopus)

Abstract

Fault diagnosis is crucial for analog circuits. In this paper, a new fault diagnosis method based on genetic algorithm and support vector machine (GA-SVM) is proposed. We design fault mode and collect the fault datasets on the basis of a quad high pass filter circuit. Wavelet packet analysis is employed to extract fault samples information. Sampled data's dimension is further reduced by Principal Component Analysis(PCA). To improve the efficiency of SVM, we use GA to search optimized parameters for its kernel function. After being trained with sampled data, the optimized SVM can steadily classify circuit faults. Simulation results show that the new algorithm classifies circuit faults at an accuracy of 92.69%. Our approach provides a new direction for analog circuit fault diagnosis.
Original languageEnglish
Title of host publication2013 IEEE International Conference of IEEE Region 10, IEEE TENCON 2013 - Conference Proceedings
Publication date2013
Article number6718926
ISBN (Print)9781479928262
DOIs
Publication statusPublished - 2013
Externally publishedYes
Event2013 IEEE International Conference of IEEE Region 10, IEEE TENCON 2013 - Xi'an, Shaanxi, China
Duration: 22 Oct 201325 Oct 2013

Conference

Conference2013 IEEE International Conference of IEEE Region 10, IEEE TENCON 2013
CountryChina
CityXi'an, Shaanxi
Period22/10/201325/10/2013
SponsorIEEE Region 10 (Asia Pacific Region), IEEE Xi'an Section, National Natural Science Foundation of China, Northwestern Polytechnical University, Xi'an Jiaotong University
SeriesIEEE Region 10 Annual International Conference, Proceedings/TENCON
ISSN2159-3442

Keywords

  • Analog Circuits
  • GA
  • PCA
  • SVM
  • Wavelet

Fingerprint

Dive into the research topics of 'A new analog circuit fault diagnosis approach based on GA-SVM'. Together they form a unique fingerprint.

Cite this