ZHENG Yunshui， LI Yujie
（College of Automation & Electrical Engineering, Lanzhou Jiaotong University, Lanzhou 730070, China）
Abstract： Aiming at fault prediction for track circuit is restricted by data, the six-terminal network model is established based on the working principle and transmission line theory of ZPW-2000A track circuit to simulate and analyze three kinds of shunting malfunction signals. A combined analysis method combining variational mode decomposition and convolutional neural network is proposed. The energy spectrum feature is extracted from the original signal by variational mode decomposition, and the deep feature is extracted by convolution neural network. Then, the sensitive feature for fault prediction is obtained by weighted fusion of the two. The simulation experiments indicate that the proposed method can predict the shunting malfunction accurately and effectively, which can achieve a prediction accuracy of 99.87% and provide a new idea for the prediction of the shunting malfunction for track circuit.
Key words： track circuit; shunting malfunction; variational mode decomposition; deep learning; feature fusion
References
［1］HU Y S. Improvement plan of track circuit bad shunting. Railway Signaling and Communication, 2008(5): 24-26.
［2］WANG R F, XI X N. On the safety warning to be made based on the fuzzy entropy and asymmetric proximity. Journal of Safety and Environment, 2017, 17(3): 829-834.
［3］FENG D, ZHAO L H. Online estimation method of shunt resistance based on cab-signal remote monitoring system. Journal of the China Railway Society, 2017, 39(4): 62-67.
［4］ZHANG M Q, ZHAO H B, SUN S P. Prediction of shunt malfunction of track circuit based on PSO-SVM. Journal of the China Railway Society, 2015, 37(10): 68-74.
［5］XU K, ZHAO L H. A rapid diagnosis method for multiple compensation capacitor faults of jointless track circuits. Journal of the China Railway Society, 2018, 40(2): 67-72.
［6］ZHANG Y P, QI H, ZHAO B. Research on method for detection of shunt state of track circuits. Journal of the China Railway Society, 2017, 39(1): 70-75.
［7］FENG D, ZHAO L H. Modeling of JTC shunt states based on train structure and running scenarios. Journal of Beijing Jiaotong University, 2019, 43(2): 41-47+57.
［8］NAGASE K, NOMURA T, HIRAMA J. Influence of contact condition between rails and wheels upon track circuit shunting (1st report quantitative evaluation of factors that cause inadequate track circuit shunting). Transactions of the Japan Society of Mechanical Engineers, Part C, 1997, 63(616): 4187-4194.
［9］HIRAMA J, WAKABAYASHI Y, KITAGAWA T, et al. Influence of contact condition between rails and wheels upon track circuit shunting (4th report, result of continuously measured shunting resistance of the running wheelsets). Transactions of the Japan Society of Mechanical Engineers, Part C, 2005, 71(4): 1263-1268.
［10］DING C J, FENG Y B, WANG M N. Rolling bearing fault diagnosis using variational mode decomposition and deep convolutional neural network. Journal of Vibration and Shock, 2021, 40(2): 287-296.
［11］CHEN Z, HU Y Q, TIAN S Q, et al. Non-stationary signal combined analysis based fault diagnosis method. Journal on Communications, 2020, 41(5): 187-195.
［12］SHAO H D, JIANG H K, ZHANG H Z, et al. Rolling bearing fault feature learning using improved convolutional deep belief network with compressed sensing. Mechanical Systems and Signal Processing, 2018, 100: 743-765.
［13］HUANG B. Research on fault optimal combination prediction model of ZPW-2000A track circuit based on improved metabolic GM(1,1) and SVM. Journal of Railway Science and Engineering, 2019, 16(11): 2852-2858.
［14］SU L N. Research on fault diagnosis of shunt malfunction for ZPW-2000A track circuit based on improved VMD and feature selection. Chengdu: Southwest Jiaotong University, 2019.
［15］KI B L, SEJUNE C, CHANG O K. A convolutional neural network for fault classification and diagnosis in semiconductor manufacturing processes. IEEE Transactions on Semiconductor Manufacturing, 2017, 30(2): 135-142.
［16］CHEN Z Q, DENG S C, CHEN X D, et al. Deep neural networks-based rolling bearing fault diagnosis. Microelectronics Reliability, 2017, 75: 327-333.
［17］JIANG G Q, HE H B, YAN J, et al. Multiscale convolutional neural networks for fault diagnosis of wind turbine gearbox. IEEE Transactions on Industrial Electronics, 2019, 66(4): 3196-3207.
基于组合分析法的轨道电路分路不良预测
郑云水， 李钰婕
（兰州交通大学 自动化与电气工程学院， 甘肃 兰州 730070）
摘要：针对轨道电路故障预测研究受数据制约的现象， 本文基于ZPW-2000A型轨道电路的工作原理和传输线理论， 建立了轨道电路六端网模型， 来仿真模拟3类分路不良故障信号， 并对其进行分析。 同时， 提出变分模态分解和卷积神经网络结合的组合分析法， 对原始信号进行变分模态分解提取能量谱特征， 再与卷积神经网络提取的深度表达特征进行加权融合， 得到敏感特征用于故障预测。 轨道电路分路不良预测的实验结果表明： 组合分析法能够准确有效预测分路不良故障， 预测的准确率达到99.87%， 为轨道电路分路不良的预测提供了新的思路。
关键词：轨道电路； 分路不良； 变分模态分解； 深度学习； 特征融合
引用格式：ZHENG Yunshui， LI Yujie． Prediction of shunting malfunction for track circuit based on combined analysis． Journal of Measurement Science and Instrumentation， 2023， 14（2）： 233-241. DOI： 10．3969／j．issn．1674-8042．2023．02．013
[full text view]
