基于特征相关性的牵引系统主回路接地故障诊断

陈志文; 李学明; 徐绍龙; 彭涛; 阳春华; 桂卫华

doi:10.16383/j.aas.c190588

基于特征相关性的牵引系统主回路接地故障诊断

doi: 10.16383/j.aas.c190588

陈志文^{1, 2, 3,},
李学明^{4, 5,},
徐绍龙^5,,
彭涛^1,,
阳春华^1,,
桂卫华^1,

1.
中南大学自动化学院长沙 410083
2.
鹏城实验室深圳 518000
3.
中南大学高性能复杂制造国家重点实验室长沙 410083
4.
湖南大学机械与运载工程学院长沙 410082
5.
株洲中车时代电气股份有限公司株洲 412001

基金项目: 国家自然科学基金(61803390, 61790571, 61773407), 轨道交通节能控制与安全监测湖南省重点实验室(2017TP1002), 高性能复杂制造国家重点实验室自主研究课题(ZZYJKT2020-14), 博士后基金(2019T120713)资助

详细信息

作者简介:
陈志文：中南大学自动化学院副教授. 2016年获得德国杜伊斯堡−埃森大学博士学位. 主要研究方向为基于模型和数据驱动的故障诊断技术. E-mail: zhiwen.chen@csu.edu.cn

李学明：株洲中车时代电气股份有限公司高级工程师. 2011年在中南大学获硕士学位. 主要研究方向为牵引系统故障诊断与预测. 本文通信作者. E-mail: lxm_hnu@163.com

徐绍龙：株洲中车时代电气股份有限公司高级工程师. 2016年在浙江大学获硕士学位. 主要研究方向为牵引系统控制, 故障诊断与预测. E-mail: xusl@csrzic.com

彭涛：中南大学自动化学院教授. 2005年获得中南大学博士学位. 主要研究方向为复杂系统的故障诊断与容错控制. E-mail: pandtao@csu.edu.cn

阳春华：中南大学自动化学院教授. 1985年获得中南工业大学学士学位. 主要研究方向为复杂工业过程建模与优化, 故障诊断与智能化系统. E-mail: ychh@csu.edu.cn

桂卫华：中国工程院院士, 中南大学自动化学院教授. 1981年获得中南矿冶学院硕士学位. 主要研究方向为复杂工业过程建模, 优化与控制应用和知识自动化. E-mail: gwh@csu.edu.cn

计量
- 文章访问数: 1388
- HTML全文浏览量: 411
- PDF下载量: 250
- 被引次数: 0
出版历程
- 收稿日期: 2019-08-17
- 录用日期: 2019-11-25
- 刊出日期: 2021-07-27

Feature Correlation-based Ground Fault Diagnosis Method for Main Circuit of Traction System

CHEN Zhi-Wen^{1, 2, 3
,},
LI Xue-Ming^{4, 5
,},
XU Shao-Long^5
,,
PENG Tao^1
,,
YANG Chun-Hua^1
,,
GUI Wei-Hua^1
,

1.
School of Automation, Central South University, Changsha 410083
2.
the Peng Cheng Laboratory, Shenzhen 518000
3.
State Key Laboratory of High Performance Complex Manufacturing, Central South University, Changsha 410083
4.
College of Mechanical and Vehicle Engineering, Hunan University, Changsha 410082
5.
Zhuzhou CRRC Times Electric Co., Ltd., Zhuzhou 412001

Funds: National Natural Science Foundation of China (61803390, 61790571, 61773407), Key Laboratory of Energy Saving Control and Safety Monitoring for Rail Transportation (2017TP1002), the Project of State Key Laboratory of High Performance Complex Manufacturing, Central South University (ZZYJKT2020-14), Postdoctoral Foundation (2019T120713)

More Information

Author Bio:
CHEN Zhi-Wen　Associate professor at the School of Automation, Central South University. He received his Ph.D. degree from University of Duisburg-Essen, Germany in 2016. His research interest covers model-based and data-based fault diagnosis

LI Xue-Ming　Senior engineer at Zhuzhou CRRC times Electric Co., Ltd. He received his master degree from Central South University in 2011. His research interest covers traction system control, fault diagnosis and prediction. Corresponding author of this paper

XU Shao-Long　Senior engineer at Zhuzhou CRRC times Electric Co., Ltd. He received his master degree from ZheJiang University in 2016. His research interest covers traction system control, fault diagnosis and prediction

PENG Tao　Professor at School of Automation, Central South University. She received her Ph.D. degree from Central South University in 2005. Her research interest covers fault diagnosis and fault tolerant control for complex systems

YANG Chun-Hua　Professor at the School of Automation, Central South University. She received her bachelor degree from Central South University of Technology in 1985. Her research interest covers complex industrial process modeling and optimization, fault diagnosis and intelligent system

GUI Wei-Hua　Academician of the Chinese Academy of Engineering, and professor at the School of Automation, Central South University. He received his master degree from Central South Institute of Mining and Metallurgy in 1981. His research interest covers complex industrial process modeling, optimization and control applications, and knowledge automation

摘要

摘要:
本文针对目前机车、动车牵引系统中主回路接地故障的精确定位问题, 提出了一种基于特征相关性的故障诊断方法. 该方法通过在线计算与故障关联的特征变量, 提取相关故障特征指标, 并考虑各故障特征指标间的相关性, 利用典型相关分析得到残差, 以实现快速故障检测. 进一步, 构建基于残差方向的故障隔离方法, 实现准确地故障定位. 现场实验表明, 与传统基于相关性的故障诊断方法以及实际工程应用方法相比, 在存在较大测量噪声与暂态工况变化时, 本文所提方法能实现更好的故障检测与隔离性能, 具有良好的应用价值.
- 接地故障 /
- 特征相关性 /
- 典型相关分析 /
- 故障隔离
Abstract:
A fault diagnosis method based on feature correlation is proposed in this paper to accurately locate the main circuit ground fault in the traction system of electrical locomotive and electric multiple unit (EMU). The characteristic variables and fault features associated with faults are calculated online, and canonical correlation analysis (CCA) is carried out to generate residual signal based on the correlation among the fault features to achieve fast fault detection. Accurate fault location is achieved based on the residual signal direction method. Field tests show that, compared with traditional CCA-based and on-board fault detection method, the proposed method has better fault detection and isolation performance in the presence of large measurement noise and transient condition changes and is also applicable to practice.
- Ground fault /
- feature correlation /
- canonical correlation analysis (CCA) /
- fault isolation

HTML全文

图 1 牵引系统典型主电路原理图

Fig. 1 Schematic diagram of main circuit of traction system

下载: 全尺寸图片幻灯片

图 2 故障类型F1、F2和F5的波形特征

Fig. 2 Features of F1、F2 and F5 faults

下载: 全尺寸图片幻灯片

图 3 基于特征相关性的故障检测算法流程图

Fig. 3 Flowchart of the feature correlation-based fault detection algorithm

下载: 全尺寸图片幻灯片

图 4 基于特征相关性的牵引主回路接地故障诊断算法原理框图

Fig. 4 Diagram of the feature correlation-based fault diagnosis algorithm for main circuit of the traction system

下载: 全尺寸图片幻灯片

图 5 机车牵引变流器实物图

Fig. 5 Photo of the traction converter for field test

下载: 全尺寸图片幻灯片

图 6 牵引变流器对外连接端子布置图

Fig. 6 Layout diagram of connection terminals under traction converter cabinet

下载: 全尺寸图片幻灯片

图 7 存在故障类型F1时的检测诊断结果

Fig. 7 Diagnosis results in F1 case

下载: 全尺寸图片幻灯片

图 10 故障残差方向显示

Fig. 10 Illustration of five fault directions

下载: 全尺寸图片幻灯片

表 1 牵引系统常见主回路接地故障点

Table 1 The ground points of typical faults

序号	主回路接地点	主电路中位置	故障代号
1	四象限输入侧正端接地	图1接地点①、③	F1
2	四象限输入侧负端接地	图1接地点②	F2
3	正母排接地	图1接地点④	F3
4	负母排接地	图1接地点⑤	F4
5	逆变输出侧接地	图1接地点⑥	F5

下载: 导出CSV

表 2 不同接地故障类型时故障特征变量$ I_{x1} $变化规律

Table 2 The change rules of $ I_{x1} $with different ground faults under working condition C5

故障类型	$ I_{x1} $变化规律	$ I_{x1} $相关指标
F1	其值在−0.5$ {\bar U_{s1}} $与0.5$ {\bar U_{s1}} $间变化,	均值, 方差, 最大值, 最大峰值
F2	且频率与四象限模块开关频率相同
F3	其值约为−0.5$ {\bar U_{s1}} $
F4	其值约为0.5$ {\bar U_{s1}} $
F5	其值在−0.5$ {\bar U_{s1}} $与0.5$ {\bar U_{s1}} $间变化, 且频率与逆变模块开关频率相同

下载: 导出CSV

表 3 不同接地故障类型时故障特征变量 $ I_{x2} $变化规律(工况5)

Table 3 The change rules of $ I_{x2} $ with different ground faults under work condition C5

故障类型	$ I_{x2} $变化规律	$ I_{x2} $相关时域指标
F1	$ I_{x2} $均值小于一定门槛值 (负值)
F2	$ I_{x2} $均值大于一定门槛值 (正值)	均值
F5	$ I_{x2} $均值约为 0

下载: 导出CSV

表 4 算法对比验证所采用数据

Table 4 Test data for comparing and verifying algorithms

序号	y	本文所提的FC-CCA方法		基于原始数据的CCA方法
		指标/变量	含义	指标/变量	含义
1	$ y_a $	$ J_1 $	$ I_{x1} $均值	$ U_{s1} $	中间直流电压
2	$ y_a $	$ J_2 $	$ I_{x1} $方差	$ U_{s2} $	半中间电压
3	$ y_a $	$ J_5 $	$ I_{x2} $均值	$ U_2 $	牵引变压器次边电压
4	$ y_b $	$ J_3 $	$ I_{x1} $最大值	$ I_{r1} $	四象限一输入电流
5	$ y_b $	$ J_4 $	$ I_{x1} $最大峰值	$ I_{r2} $	四象限二输入电流

下载: 导出CSV

表 5 不同故障类型时测量值变化情况说明

Table 5 Description of changes in measured in different fault types

故障类型	可检测工况
故障类型	C3	C4	C5
F1	√	√	√
F2	√	√	√
F3	√	√	√
F4	√	√	√
F5	×	×	√

下载: 导出CSV

表 6 不同故障类型时FDR与CIR结果对比

Table 6 Comparison results of FDR and CIR in different fault types

故障代号	故障检测率(FDR)			正确隔离率(CIR)
	FC-CCA 方法	CCA 方法	传统工程方法	FC-CCA 方法	CCA 方法	传统工程方法
F1	100%	99.99%	64.09%	100%	27.48%	0%
F2	100%	100%	79.74%	100%	14.44%	0%
F3	100%	100%	100%	100%	61.12%	0%
F4	100%	100%	79.03%	100%	21.93%	0%
F5	99.84%	99.68%	94.16%	99.32%	1.42%	0%

下载: 导出CSV

参考文献(25)

[1]	2018年中国轨道交通行业分析报告—市场运营态势与发展前景预测. 中国报告网, http://baogao.chinabaogao.com/yunshufuzhusheshi/342590342590.html.
[2]	Dong H. R, Ning B, Cai B. G, Hou Z. Automatic train control system development and simulation for high-speed railways. IEEE Circuits & Systems Magazine, 2010, 10(2): 6−18
[3]	周东华, 纪洪泉, 何潇. 高速列车信息控制系统的故障诊断技术. 自动化学报, 2018, 44(7): 1153−1164 Zhou Dong-Hua, Ji Hong-Quan, He Xiao. Fault diagnosis techniques for the information control system of high-speed trains. Acta Automatica Sinica, 2018, 44(7): 1153−1164
[4]	杨超, 彭涛, 阳春华, 陈志文, 桂卫华. 高速列车牵引传动系统故障测试与验证仿真平台研究综述. 自动化学报, 2019, 47(12): 2218−2232 Yang Chao, Peng Tao, Yang Chun-Hua, Chen Zhi-Wen, Gui Wei-Hua. Review of fault testing and its validation simulation platform for traction drive system of high-speed trains. Acta Automatica Sinica, 2019, 47(12): 2218−2232
[5]	Chen H. T, Jiang B. A review of fault detection and diagnosis for the traction system in high-speed trains. IEEE Transactions on Intelligent Transportation Systems, 2019: 1−16
[6]	陶宏伟, 彭涛, 杨超, 阳春华, 陈志文, 桂卫华. 高速列车牵引整流器多类故障联合诊断方法. 自动化学报, 2019, 45(12): 2294−2302 Tao Hong-Wei, Peng Tao, Yang Chao, Yang Chun-Hua, Chen Zhi-Wen, Gui Wei-Hua. Joint fault diagnosis method of multiclass faults for traction rectififier in high-speed train. Acta Automatica Sinica, 2019, 45(12): 2294−2302
[7]	刘强, 方彤, 董一凝, 秦泗钊. 基于动态建模与重构的列车轴承故障检测和定位. 自动化学报, 2019, 45(12): 2233−2241 Liu Qiang, Fang Tong, Dong Yi-Ning, Qin S. Joe. Dynamic modeling and reconstruction based fault detection and location of train bearings. Acta Automatica Sinica, 2019, 45(12): 2233−2241
[8]	田光兴, 郝凤荣, 代兴军. HXD3系列电力机车接地方式及接地故障检测环节探析. 铁道机车与动车, 2016, 5: 27−33 doi: 10.3969/j.issn.1003-1820.2016.02.009 Tian Guang-Xing, Hao Feng-Rong, Dai Xing-Jun. Discussion on grounding mode and ground fault detection circuit of HXD3 serial electric locomotive. Railway Locomotive and Motor Car, 2016, 5: 27−33 doi: 10.3969/j.issn.1003-1820.2016.02.009
[9]	El-Sherif N. Ground fault protection − all you need to know. IEEE Transactions on Industry Applications, 2017, 53(6): 6047−6056 doi: 10.1109/TIA.2017.2746558
[10]	RumMdkarn J, Ngaopitakkul A. Behavior analysis of winding to ground fault in transformer using high and low frequency components from discrete wavelet transform. In: Proceedings of International conference on applied system innovation, Sapporo, Japan, 2017: 13−17
[11]	Gruhn T, Glenney J, Savostianik M. Type B ground-fault protection on adjustable frequency drives. IEEE Transactions on Industry Applications, 2017, 54(1): 934−939
[12]	Wei L, Liu Z, Kerkman R. J, Skibinski G. L. Identifying ground-fault locations: Using adjustable speed drives in high-resistance grounded systems. IEEE Industry Applications Magazine, 2013, 19(2): 47−55
[13]	Hu J, Wei L, McGuire J, Liu Z. Ground fault location self-diagnosis in high resistance grounding drive systems. In: Proceedings of IEEE Energy Conversion Congress and Exposition, Pittsburgh, PA, USA, 2014: 3179−3185
[14]	Hu J, Wei L, McGuire J, Liu Z. Flux linkage detection based ground fault identification and system diagnosis in high-resistance grounding systems. IEEE Transactions on Industry Applications, 2016, 53(3): 2967−2975
[15]	李伟, 郭晓燕, 张波. “和谐”系列电力机车传动系统接地检测比较. 机车电传动, 2010, 6: 67−69 Li Wei, Guo Xiao-Yan, Zhang Bo. Comparisons of HXD serial locomotives drive system grounding detection. Electric Drive for Locomotives, 2010, 6: 67−69
[16]	徐培刚, 彭军华, 罗铁军. HXD1C型机车主回路接地故障分析. 机车电传动, 2013, 36: 103−107 Xu Pei-Gang, Peng Jun-Hua, Luo Tie-Jun. Analysis on main circuit ground fault of HXD1C locomotive. Electric Drive for Locomotives, 2013, 36: 103−107
[17]	陈立胜, 颜罡. 交流传动电力机车主电路接地故障的检测、定位及对策分析. 电力机车与城轨车辆, 2013, 36(3): 84−86 Chen Li-Sheng, Yan Gang. Location and countermeasure analysis of main circuit grounding fault of AC drive electric locomotive. Electric Locomotives & Mass Transit Vehicles, 2013, 36(3): 84−86
[18]	李耘笼, 刘可安. DJJ2型动力车主接地故障分析. 电力机车与城轨车辆, 2005, 28(2): 54−55 Li Geng-Long, Liu Ke-An. Ground fault analysis on main circuit for power car with type DJJ2. Electric Locomotives & Mass Transit Vehicles, 2005, 28(2): 54−55
[19]	Liu Y, Liu B, Zhao X, Xie M. A mixture of variational canonical correlation analysis for nonlinear and quality-relevant process monitoring. IEEE Transactions on Industrial Electronics, 2018, 65(8): 6478−6486 doi: 10.1109/TIE.2017.2786253
[20]	Chen Z. W, Ding S. X, Peng T, Yang C. H, Gui W. H. Fault detection for non-Gaussian processes using generalized correlation analysis and randomized algorithms. IEEE Transactions on Industrial Electronics, 2018, 65(2): 1559−1567 doi: 10.1109/TIE.2017.2733501
[21]	Zhang K, Peng K. X, Ding S. X, Chen Z. W, Yang X. A correlation-based distributed fault detection method and its application to a hot tandem rolling mill process. IEEE Transactions on Industrial Electronics, 2019, 67(3): 2380−2390 doi: 10.1109/TIE.2019.2901565
[22]	Chen Z. W, Li X. M, Yang C, Peng T, Yang C. H, Karimi H, Gui W. H. A data-driven ground fault detection and isolation method for main circuit in railway electrical traction system. ISA Transactions, 2019, 87: 264−271 doi: 10.1016/j.isatra.2018.11.031
[23]	Anderson T. An Introduction to Multivariate Statistical Analysis. Second Edition. New York: John Wiley and Sons, LTD: 1984
[24]	Gustafsson F. Statistical signal processing approaches to fault detection. Annual Reviews in Control, 2006, 31(1): 41−54
[25]	He Z. M, Chen Z. W, Zhou H. Y, Wang D. Y, Xing Y, Wang J. Q. A visualization approach for unknown fault diagnosis. Chemometrics and Intelligent Laboratory Systems, 2018, 172: 80−89 doi: 10.1016/j.chemolab.2017.11.013