Fault diagnosis method for energy storage mechanism of high voltage circuit breaker based on CNN characteristic matrix constructed by sound-vibration signal

Shutao Zhao1 , Erxu Wang2 , Jiawei Hao3

1, 2, 3School of Electrical and Electronic Engineering, North China Electric Power University, Baoding, China

1Corresponding author

Journal of Vibroengineering, Vol. 21, Issue 6, 2019, p. 1665-1678. https://doi.org/10.21595/jve.2019.20781
Received 7 May 2019; received in revised form 1 July 2019; accepted 6 August 2019; published 30 September 2019

Copyright © 2019 Shutao Zhao, et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Creative Commons License
Abstract.

Aiming at the problem that some traditional high voltage circuit breaker fault diagnosis methods were over-dependent on subjective experience, the accuracy was not very high and the generalization ability was poor, a fault diagnosis method for energy storage mechanism of high voltage circuit breaker, which based on Convolutional Neural Network (CNN) characteristic matrix constructed by sound-vibration signal ,was proposed. In this paper, firstly, the morphological filtering was used for background noise cancellation of sound signal, and the time scale alignment method based on kurtosis and envelope similarity were proposed to ensure the synchronism of the sound-vibration signal. Secondly, the Pearson correlation coefficient was used to construct two-dimensional image characteristic matrix for the expanded sound-vibration signal. Finally, the characteristic matrix was trained by utilizing CNN. Local Response Normalization (LRN) and core function decorrelation were utilized to improve the structure of CNN model, which reduced the bad impact of large data fluctuation of energy storage process on the diagnostic accuracy of circuit breaker energy storage mechanism. Compared with the traditional method, the proposed method has obvious advantages, whose total accurate rate up to 98.2 % and generalization performance is excellent.

Graphical Abstract

Highlights
  • The combination of sound signal and vibration signal
  • CNN optimization model based on Local Response Normalization (LRN) and core function decorrelation
  • Signal preprocessing based on morphological denoising and time scale alignment method
  • data expansion to obtain the large amount of data
  • Pearson correlation coefficient to construct sound-vibration signal characteristic matrix for CNN

Keywords: sound-vibration combination, CNN characteristic matrix, time scale alignment, data expansion, fault diagnosis.

References

  1. Xu Guozheng Principle and Application of High Voltage Circuit Breakers. Tsinghua University Press, Beijing, 2000. [CrossRef]
  2. Song Youwen Discussion on several problems of line breaker failure protection. Power System Protection and Control, Vol. 36, Issue 23, 2008, p. 88-91. [CrossRef]
  3. Zhao Shutao, Wang Erxu Fault diagnosis of circuit breaker energy storage mechanism based on current-vibration entropy weight characteristic and grey wolf optimization-support vector machine. IEEE Access, Vol. 7, 2019, p. 86798-86809. [CrossRef]
  4. Zhao Shutao, Wu Chengjian, Li Ming, et al. Research on the testing method of mechanical characteristics of circuit breakers based on NCC-P-S optimization algorithm. Journal of Electrical Engineering of China, Vol. 37, Issue 14, 2017, p. 4265-4271. [CrossRef]
  5. Lin Jintao Research on Mechanical Fault Diagnosis Method of Circuit Breaker Based on Vibration Signal. Shandong University, 2019. [CrossRef]
  6. Jérôme Gilles Empirical wavelet transform. IEEE Transactions on Signal Processing, Vol. 61, Issue 16, 2013, p. 3999-4010. [Publisher]
  7. Zhao Shutao, Zhang Pei, Shenlu, et al. Vibration-sound combined fault diagnosis method for high voltage circuit breakers. Journal of Electrical Technology, Vol. 29, Issue 7, 2014, p. 216-221. [CrossRef]
  8. Zhao Shutao, Wang Yaxiao, Li Mufeng, et al. Circuit breaker fault diagnosis method based on sound-vibration joint characteristic entropy. Journal of North China Electric Power University (Natural Science Edition), Vol. 43, Issue 6, 2016, p. 20-24. [CrossRef]
  9. Yang Yuanwei, Guan Yonggang, Chen Shigang, et al. Mechanical fault diagnosis method of high voltage circuit breaker based on sound signal. China Journal of Electrical Engineering, Vol. 38, Issue 22, 2018, p. 6730-6737. [CrossRef]
  10. Sun Shuguang, Yu Hao, Du Taihang, et al. Fault vibration diagnosis method of universal circuit breaker based on multi-feature fusion and improvement of QPSO-RVM. Journal of Electrical Technology, Vol. 32, Issue 19, 2017, p. 107-117. [CrossRef]
  11. He Mengyuan, Ding Qiaolin, Zhao Shutao, et al. Research of circuit breaker intelligent fault diagnosis method based on double clustering. IEICE Electronics Express, Vol. 14, Issue 17, 2017, p. 20170463. [CrossRef]
  12. Qu Jianling, Yu Lu, Yuan Tao, Tian Yanping, et al. an adaptive fault diagnosis algorithm for rolling bearings based on one-dimensional convolution neural network. Journal of Instruments and Instruments, Vol. 39, Issue 7, 2018, p. 134-143. [CrossRef]
  13. Ren Hao, Qu Jianfeng, Chai Yi, et al. Research status and challenges of in-depth learning in the field of fault diagnosis. Control and decision-making, Vol. 32, Issue 8, 2017, p. 1345-1358. [CrossRef]
  14. Huang Xinbo, Hu Xiaowen, Zhu Yongcan, et al. Fault diagnosis of high voltage circuit breakers based on convolutional neural network algorithm. Power automation equipment, Vol. 38, Issue 5, 2018, p. 136-140. [CrossRef]
  15. Liu Cheng Fault Diagnosis of High Voltage Circuit Breakers Based on Vibration Signal Processing. Xiamen Institute of Technology, 2017. [CrossRef]
  16. An Jing, Ai Ping, Xu Sen, et al. An intelligent fault diagnosis method for rotating machinery based on one-dimensional convolution neural network. Journal of Nanjing University (Natural Science), Vol. 55, Issue 1, 2019, p. 133-142. [CrossRef]
  17. Li Dongdong, Wang Hao, Yang Fan, et al. Fault detection of planetary gearbox of wind turbine based on one-dimensional convolution neural network and Soft-Max classifier. Motor and Control Applications, Vol. 45, Issue 6, 2018, p. 80-87. [CrossRef]
  18. Xiao Liangjun Research on fault diagnosis method of offset press rolling bearing based on convolution neural network. Xi’an University of Technology, 2018. [CrossRef]
  19. Li Guoli, Huo Mingxia, Gao Xinzhi, et al. Mechanical fault diagnosis method of circuit breaker based on LMD and time-frequency fractal dimension. Instruments and Analysis Monitoring, Vol. 4, 2018, p. 1-5. [CrossRef]
  20. Meng Qinghua, Hou Zhoubo, Sun Xiaohong Research on noise reduction algorithm for vibration signal of automobile hub unit based on mathematical morphology. Mechatronics Engineering, Vol. 30, Issue 4, 2013, p. 411-416. [CrossRef]
  21. Menger Research on Insulator State Recognition Based on CNN. North China Electric Power University, 2018. [CrossRef]
  22. Krizhevsky A., Sutskever I., Hinton G. E. ImageNet classification with deep convolutional neural networks. International Conference on Neural Information Processing Systems, 2012, p. 1097-1105. [CrossRef]