Support Vector Machine for Fault Diagnosis in Electrical Circuits

Author

Siwek, Krzysztof ; Osowski, Stanislaw ; Markiewicz, Tomasz

Author_Institution

Electr. Eng. Dept., Warsaw Univ. of Technol.

fYear

2006

fDate

38869

Firstpage

342

Lastpage

345

Abstract

The paper presents an automatic system for diagnosis of the parametric fault of the individual elements of the electrical circuit on the basis of external measurements of the voltage and current in this circuit. The fundamental role fulfills here the nonlinear signal processing performed by the support vector machine of the Gaussian kernel. The paper presents the results of numerical experiments on the basis of the measurements made in real RC ladder network

Keywords

Gaussian processes; RC circuits; electric current measurement; fault diagnosis; ladder networks; signal processing; support vector machines; voltage measurement; Gaussian kernel; RC ladder network; current measurement; electrical circuits; fault diagnosis; nonlinear signal processing; support vector machine; voltage measurement; Circuit analysis; Circuit faults; Circuit testing; Current measurement; Fault diagnosis; Frequency measurement; Shape measurement; Support vector machine classification; Support vector machines; Voltage;

fLanguage

English

Publisher

ieee

Conference_Titel

Signal Processing Symposium, 2006. NORSIG 2006. Proceedings of the 7th Nordic

Conference_Location

Rejkjavik

Print_ISBN

1-4244-0412-6

Electronic_ISBN

1-4244-0413-4

Type

conf

DOI

10.1109/NORSIG.2006.275251

Filename

4052246