119. Power electronic circuits fault diagnosis based on wavelet packet transform and LSSVM

Deqiang He1, Kai Lu2, Qiong Xiao3

1, 2Guangxi Key Laboratory of Manufacturing System and Advanced Manufacturing Technology,
College of Mechanical Engineering, Guangxi University, Nanning, China

3Nanning China Railway Rail Transit Group Co. Ltd, Nanning, China

1Corresponding author

E-mail: 1hdqianglqy@126.com, 2mesut20150706@163.com, 3xq0735@163.com

Received 15 May 2017; received in revised form 25 May 2017; accepted 26 May 2017

DOI https://doi.org/10.21595/jme.2017.18631

 

Abstract. Power electronic circuits play a vital role in industry application and get more attention in fault diagnosis fields in recent years. In this paper, a method based on wavelet packet transform and least square support vector machine is proposed to diagnose fault of power electronic circuits. We use single-phase half-bridge rectifier as example. Output voltage signal at the main circuit DC side is selected as research object. Wavelet packet transform is used to extract fault feature samples and then multi-class LSSVM classification is built for fault identification. Results show that the performances based on LSSVM are better than that of RBPNN.

Keywords: power electronic circuits, fault diagnosis, wavelet packet transform, LSSVM.

References

[1]        Tao C., Shan-xu D., et al. A survey of fault diagnosis technology for power electronics system. Electrical Measurement and Instrumentation, Vol. 5, 2008, p. 1‑7.

[2]        Hu Zhi-kun, Gui Wei-hua, et al. Wavelet fractal fault detection method of power electronic circuit. Control Engineering of China, Vol. 3, 2008, p. 337‑341.

[3]        Jing-ding Cai, Ren-wu Yan Fault diagnosis of power electronic circuit based on wavelet analysis and random forests algorithm. Journal of Electronic Power Science and Technology, Vol. 2, 2011, p. 54‑60.

[4]        Jiang Cui, You-ren Wang Testing the parametric faults of power electronic circuits applying classifiers fusion method based on fuzzy inference. Proceedings of the CSEE, Vol. 18, 2009, p. 54‑59.

[5]        Yi Wu, Youren Wang Multiple parametric faults diagnosis for power electronic circuits based on hybrid bond graph and genetic algorithm. Measurement, Vol. 92, 2016, p. 365‑381.

[6]        Hu Qing, Wang Rongjie, Zhan Yiju Fault diagnosis technology based on SVM in power electronics circuit. Proceedings of the CSEE, Vol. 12, 2008, p. 107‑111.

[7]        Cai Jin-ding, Yan Ren-wu Fault diagnosis of power electronic circuit applying ARMA bispectrum and discrete hidden Markov model. Proceedings of the CSEE, Vol. 24, 2010, p. 54‑60.

[8]        Gong G. S., Wang X. Load test and detection system of DC 600 V locomotive power supply device. Electric Drive for Locomotives, Vol. 5, 2011, p. 4‑58.

[9]        Zhang B. K. Design and realization of fault diagnosis software for HXD3C electric locomotive power supply. Electric Drive for Locomotives, Vol. 5, 2012, p. 93‑95.

[10]     Cai Jin-ding, Yan Ren-wu Fault diagnosis of power electronic circuit based on wavelet analysis and random forests algorithm. Journal of Electric Power Science and Technology, Issue 2, 2011, p. 54‑60.

[11]     Liu M. C. Wavelet Analysis and Application. Tsinghua University Press, Beijing, 2013.

[12]     Han X. J. Research on Fault Diagnosis of Power Electronic Circuits Based on Wavelet Transform and Neural Network. Nanjing University of Aeronautics and Astronautics, 2008.

[13]     Astaf’eva N. M. Wavelet analysis: basic theory and some applications. Physics-Uspekhi, Vol. 39, 11, p. 1996‑1085.

[14]     Chen P., Chen H. Y. Wind speed forecasting based on combination of wavelet packet analysis with support vector regression. Power System Technology, Vol. 35, Issue 51, 2011, p. 178‑182.

[15]     Cortes C., Vapnik V. N. Support vector network. Machine Learning, Vol. 20, Issue 3, 1995, p. 237‑297.

[16]     Suykens J. A. K., Vandewalle J. Least squares support vector machine classifers. Neural Processing Letters, Vol. 9, Issue 3, 1999, p. 293‑300.

[17]     Vapnik V N. The Nature of Statistical Learning Theory. Tsinghua University Press, Beijing, 2005.

[18]     Bai P., Zhang X. B. Support Vector Machine and Its Application. Xidian University Press, Xi’an, 2008.

[19]     Chen P., Yuan L. F. An improved SVM classifier based on double chains quantum genetic algorithm and its application in analogue circuit diagnosis. Neurocomputing, Vol. 211, 2016, p. 202‑211.

Cite this article

He Deqiang, Lu Kai, Xiao Qiong Power electronic circuits fault diagnosis based on wavelet packet transform and LSSVM. Journal of Measurements in Engineering, Vol. 5, Issue 2, 2017, p. 68‑76.

 

Journal of Measurements in Engineering. June 2017, Volume 5, Issue 2

© JVE International Ltd. ISSN Print 2335-2124, ISSN Online 2424-4635, Kaunas, Lithuania