logo

SCIENTIA SINICA Informationis, Volume 50, Issue 4: 496-510(2020) https://doi.org/10.1360/SSI-2019-0220

Data-driven fault diagnosis for dynamic traction systems in high-speed trains

More info
  • ReceivedOct 7, 2019
  • AcceptedDec 7, 2019
  • PublishedApr 3, 2020

Abstract

Traction systems are an important aspect of high-speed trains, and their reliable operation is crucial. With data available from trains, this paper proposes an optimal fault detection and diagnosis (FDD) strategy for dynamic traction systems. Based on the established dynamic model, using sensor measurements, a correlation-aided subspace identification technique is proposed to formulate residual signals and corresponding test statistics for fault detection. Then, a modified support vector machine (SVM) is designed for optimally solving the diagnosis bias caused by the difference in the apparent probabilities of multiple fault scenarios. The feasibility and effectiveness of the proposed optical FDD performance are illustrated in the CRRC experimental platforms.


Funded by

国家自然科学基金(61490703,61922042)


References

[1] Givoni M. Development and Impact of the Modern High?speed Train: A Review. Transp Rev, 2006, 26: 593-611 CrossRef Google Scholar

[2] Liu Q, Kong D, Qin S J. Map-Reduce Decentralized PCA for Big Data Monitoring and Diagnosis of Faults in High-Speed Train Bearings. IFAC-PapersOnLine, 2018, 51: 144-149 CrossRef Google Scholar

[3] Zhou D H, Ji H Q, He X. Fault diagnosis techniques for the information control system of high-speed trains. Acta Automat Sin, 2018, 44: 1153--1164. Google Scholar

[4] Yang C, Yang C, Peng T. A Fault-Injection Strategy for Traction Drive Control Systems. IEEE Trans Ind Electron, 2017, 64: 5719-5727 CrossRef Google Scholar

[5] Chen H, Jiang B. A Review of Fault Detection and Diagnosis for the Traction System in High-Speed Trains. IEEE Trans Intell Transp Syst, 2020, 21: 450-465 CrossRef Google Scholar

[6] Chen H, Jiang B, Chen W. Data-driven Detection and Diagnosis of Incipient Faults in Electrical Drives of High-Speed Trains. IEEE Trans Ind Electron, 2019, 66: 4716-4725 CrossRef Google Scholar

[7] Jiang B, Wu Y K, Lu N Y, et al. Review of fault diagnosis and prognosis techniques for high-speed railway traction system. Control Decision, 2018, 33: 841--855. Google Scholar

[8] Rules for Operation and Maintenance of Railway EMU. http:$//$www.nra.gov.cn$/$jgzf$/$flfg$/$gfxwj$/$fbsj$/$2017_$/$201803$/$t201 80302_52968.shtml. Google Scholar

[9] Zhang S G. CRH$_2$-type Electric Multiple Units. Beijing: China Railway Publishing House, 2008. 146--150. Google Scholar

[10] Zhou D, Ji H, He X. Fault Detection and Isolation of the Brake Cylinder System for Electric Multiple Units. IEEE Trans Contr Syst Technol, 2018, 26: 1744-1757 CrossRef Google Scholar

[11] Guzinski J, Abu-Rub H, Diguet M. Speed and Load Torque Observer Application in High-Speed Train Electric Drive. IEEE Trans Ind Electron, 2010, 57: 565-574 CrossRef Google Scholar

[12] Chen Z, Ding S X, Peng T. Fault Detection for Non-Gaussian Processes Using Generalized Canonical Correlation Analysis and Randomized Algorithms. IEEE Trans Ind Electron, 2018, 65: 1559-1567 CrossRef Google Scholar

[13] Chen H, Jiang B, Lu N. Deep PCA Based Real-Time Incipient Fault Detection and Diagnosis Methodology for Electrical Drive in High-Speed Trains. IEEE Trans Veh Technol, 2018, 67: 4819-4830 CrossRef Google Scholar

[14] Feig P, Schatz J, Gschwendtner K. Prospective and Monetary Effectiveness Assessment Method for Advanced Driver Assistance Systems-Usage of Naturalistic Driving Studies and Experimental System Tests. IEEE Intell Transp Syst Mag, 2018, 10: 10-23 CrossRef Google Scholar

[15] Youssef A B, El Khil S K, Slama-Belkhodja I. State Observer-Based Sensor Fault Detection and Isolation, and Fault Tolerant Control of a Single-Phase PWM Rectifier for Electric Railway Traction. IEEE Trans Power Electron, 2013, 28: 5842-5853 CrossRef ADS Google Scholar

[16] Dai C, Huang K, Hu K. Fault diagnosis approach of traction transformers in high-speed railway combining kernel principal component analysis with random forest. IET Elect Syst Transp, 2016, 45: 202-206 CrossRef Google Scholar

[17] Zhou D, Qiu H, Yang S. Submodule Voltage Similarity-Based Open-Circuit Fault Diagnosis for Modular Multilevel Converters. IEEE Trans Power Electron, 2019, 34: 8008-8016 CrossRef ADS Google Scholar

[18] Bennett S M, Patton R J, Daley S. Sensor fault-tolerant control of a rail traction drive. Control Eng Practice, 1999, 7: 217-225 CrossRef Google Scholar

[19] Gou B, Ge X, Wang S. An Open-Switch Fault Diagnosis Method for Single-Phase PWM Rectifier Using a Model-Based Approach in High-Speed Railway Electrical Traction Drive System. IEEE Trans Power Electron, 2016, 31: 3816-3826 CrossRef ADS Google Scholar

[20] Estima J O, Marques Cardoso A J. A New Algorithm for Real-Time Multiple Open-Circuit Fault Diagnosis in Voltage-Fed PWM Motor Drives by the Reference Current Errors. IEEE Trans Ind Electron, 2013, 60: 3496-3505 CrossRef Google Scholar

[21] Henao H, Kia S H, Capolino G A. Torsional-Vibration Assessment and Gear-Fault Diagnosis in Railway Traction System. IEEE Trans Ind Electron, 2011, 58: 1707-1717 CrossRef Google Scholar

[22] Hu K, Liu Z, Lin S. Wavelet Entropy-Based Traction Inverter Open Switch Fault Diagnosis in High-Speed Railways. Entropy, 2016, 18: 78 CrossRef ADS Google Scholar

[23] Jiang Y, Yin S. Recursive Total Principle Component Regression Based Fault Detection and Its Application to Vehicular Cyber-Physical Systems. IEEE Trans Ind Inf, 2018, 14: 1415-1423 CrossRef Google Scholar

[24] Dong H, Ning B, Cai B. Automatic Train Control System Development and Simulation for High-Speed Railways. IEEE Circuits Syst Mag, 2010, 10: 6-18 CrossRef Google Scholar

[25] Yang C, Gui W, Chen Z. Voltage Difference Residual-Based Open-Circuit Fault Diagnosis Approach for Three-Level Converters in Electric Traction Systems. IEEE Trans Power Electron, 2020, 35: 3012-3028 CrossRef ADS Google Scholar

[26] Chen Y, Lv Y, Wang F Y. Traffic Flow Imputation Using Parallel Data and Generative Adversarial Networks. IEEE Trans Intell Transp Syst, 2019, : 1-7 CrossRef Google Scholar

[27] Review on the Traction System Sensor Technology of a Rail Transit Train.. Sensors, 2017, 17: 1356 CrossRef PubMed Google Scholar

[28] Yang X, Yang C, Peng T. Hardware-in-the-Loop Fault Injection for Traction Control System. IEEE J Emerg Sel Top Power Electron, 2018, 6: 696-706 CrossRef Google Scholar

[29] Ding S X, Yang Y, Zhang Y. Data-driven realizations of kernel and image representations and their application to fault detection and control system design. Automatica, 2014, 50: 2615-2623 CrossRef Google Scholar

[30] Qin S J. An overview of subspace identification. Comput Chem Eng, 2006, 30: 1502-1513 CrossRef Google Scholar

[31] Chen Z, Ding S X, Zhang K. Canonical correlation analysis-based fault detection methods with application to alumina evaporation process. Control Eng Practice, 2016, 46: 51-58 CrossRef Google Scholar

  • Figure 1

    (Color online) Schematic diagram of traction systems in CRH2-type high-speed trains

  • Figure 2

    (Color online) Platform of fault injection and diagnosis for traction systems of high-speed trains

  • Figure 3

    (Color online) Detection results for 4 types of faults. (a) Fault $f_1$: fault detection ratio is $100%$, and false alarm ratio is $0.3%$; (b) fault $f_2$: fault detection ratio is $59.69%$, and false alarm ratio is $0.92%$; (c) fault $f_3$: fault detection ratio is $99.46%$, and false alarm ratio is $0.77%$; (d) fault $f_4$: fault detection ratio is $71.5%$, and false alarm ratio is $0.23%$

  • Figure 4

    (Color online) Diagnosis results using (a) traditional SVM and (b) the proposed modified SVM

  •   

    Algorithm 1 离线辨识算法

    根据给定的 $N,s$, 得到输入矩阵 $U$ 与输出矩阵 $Y$;

    通过式 (14), 得到 $\mathcal{M}$ 矩阵;

    依据式 (15), 定义数据驱动的残差信号 $r(k)$;

    通过式 (17) 定义统计量, 并根据式 (18) 得到用于故障检测的阈值.

  •   

    Algorithm 2 离线故障特征提取算法

    对于 $\mathcal{T}$ 种故障数据, 通过式 (15) 得到所有类型故障的残差信号;

    采用一对多法, 定义故障诊断的目标函数 (20);

    根据诊断误差, 更新权重向量 $m$;

    通过式 (23) 得到新的惩罚因子;

    生成新的故障诊断超平面.

  •   

    Algorithm 3 在线 FDD 算法

    读取在线数据 $u$ 和 $y$;

    根据式 (15) 得到当前时刻残差;

    依据式 (17) 中的统计量, 进行故障检测. 当系统无故障时, 返回第 1 步;否则继续执行算法;

    以报警的残差作为算法 2 得到分类器的输入, 输出故障类型, 并返回第 1 步.

Copyright 2020 Science China Press Co., Ltd. 《中国科学》杂志社有限责任公司 版权所有

京ICP备18024590号-1       京公网安备11010102003388号