Fault prediction of power supply vehicle based on multi-state time series prediction learning

被引：0
作者：
Li W. [1 ,2 ,3 ]
Zhou B.-X. [1 ,2 ,3 ]
Jiang D.-N. [1 ,2 ,3 ]
Sun X.-J. [4 ]
机构：
[1] College of Electrical and Information Engineering, Lanzhou University of Technology, Lanzhou
[2] Key Laboratory of Gansu Advanced Control for Industrial Processes, Lanzhou University of Technology, Lanzhou
[3] National Demonstration Center for Experimental Electrical and Control Engineering Education, Lanzhou University of Technology, Lanzhou
[4] Lanzhou Power Supply Vehicle Research Institute Co. Ltd., Lanzhou
来源：
| 1600年 / Editorial Board of Jilin University卷 / 50期
关键词：
Fault prediction; Improved k-nearest neighbor algorithm; Long short-term memory network; Power supply vehicle; State time series;
D O I：
10.13229/j.cnki.jdxbgxb20181290
中图分类号：
学科分类号：
摘要：
The existing fault prediction methods are difficult to apply to large and complex equipment. Aiming at this situation, a fault prediction method based on multi-state time series dynamic trend prediction learning is proposed for power supply vehicle. Firstly, this method establishes a time series prediction model of power supply vehicle operation status based on Long Short Term Memory (LSTM) network, and predicts the future operation situation by combining the history and real-time operation data of power supply vehicle. Then, on the basis of obtaining the prediction situation, the improved -Nearest Neighbor (kNN) algorithm is used to analyze the correlation between the state change trend and the fault, and to predict the possible faults in the future. Experimental analysis is carried out on the simulation system of power supply vehicle. The results verify the validity and applicability of the proposed method. © 2020, Jilin University Press. All right reserved.
引用
收藏
页码：1532 / 1544
页数：12
相关论文
共 31 条
  • [1] Jiang Bin, Wu Yun-kai, Lu Ning-yun, Et al., Review of fault diagnosis and prognosis techniques for high-speed railway traction system, Control and Decision, 33, 5, pp. 841-855, (2018)
  • [2] Zeng Jun, Chen Yan-feng, Yang Ping, Et al., Review of fault diagnosis methods of large-scale wind turbines, Power System Technology, 42, 3, pp. 849-860, (2018)
  • [3] Mellit A, Tina G M, Kalogirou S A., Fault detection and diagnosis methods for photovoltaic systems: a review, Renewable and Sustainable Energy Reviews, 91, pp. 1-17, (2018)
  • [4] Wang De-jun, Lyu Zhi-chao, Wang Qi-ming, Et al., Engine misfire fault diagnosis based on cylinder pressure identification, Journal of Jilin University(Engineering and Technology Edition), 47, 3, pp. 917-923, (2017)
  • [5] Pei Hong, Hu Chang-hua, Si Xiao-sheng, Et al., Remaining life prediction information-based maintenance decision model for equipment under imperfect maintenance, Acta Automatica Sinica, 44, 4, pp. 719-729, (2018)
  • [6] Sun Qiang, Yue Ji-guang, Review on fault prognostic methods based on uncertainty, Control and Decision, 29, 5, pp. 769-778, (2014)
  • [7] Zhang Pan, Su Yi-fang, Han Yin-feng, Et al., Judgment on failure distribution type and estimation of failure rate for distribution terminals, Proceedings of the CSU-EPSA, 30, 2, pp. 61-65, (2018)
  • [8] Zhang Ying-zhi, Liu Jin-tong, Shen Gui-xiang, Et al., Reliability modeling of CNC machine tools system based on failure correlation analysis, Journal of Jilin University(Engineering and Technology Edition), 47, 1, pp. 169-173, (2017)
  • [9] Zhao Z, Liang B, Wang X, Et al., Remaining useful life prediction of aircraft engine based on degradation pattern learning, Reliability Engineering & System Safety, 164, pp. 74-83, (2017)
  • [10] Ai Hong, Zhou Dong-hua, Fault prediction approach for dynamic system, Journal of Huazhong University of Science and Technology(Natural Science Edition), 37, pp. 222-225, (2009)
1234