Fault diagnosis method for train control on-board interface equipment of CTCS-3 based on temporal knowledge graph completion

doi:10.11772/j.issn.1001-9081.2024070990

Journal of Computer Applications ›› 2025, Vol. 45 ›› Issue (2): 677-684.DOI: 10.11772/j.issn.1001-9081.2024070990

• Frontier and comprehensive applications • Previous Articles

Fault diagnosis method for train control on-board interface equipment of CTCS-3 based on temporal knowledge graph completion

Meng WANG¹, Daqian ZHANG², Bingyan ZHOU³, Qianying MA³, Jidong LYU⁴()

^1.Beijing HollySys System Engineering Company Limited，Beijing 100176，China
^2.Signal Communication Research Institute，China Academy of Railway Sciences Corporation Limited，Beijing 100081，China
^3.School of Automation and Intelligence，Beijing Jiaotong University，Beijing 100044，China
^4.National Engineering Research Center of Rail Transportation Operation and Control System，Beijing Jiaotong University，Beijing 100044，China

Received:2024-07-15 Revised:2024-11-16 Accepted:2024-11-22 Online:2024-12-03 Published:2025-02-10
Contact: Jidong LYU
About author:WANG Meng， born in 1983， M. S.， senior engineer. His research interests include railroad signal.
ZHANG Daqian， born in 1997， M. S. His research interests include railroad signal.
ZHOU Bingyan， born in 2001， M. S. candidate. Her research interests include transportation system simulation and testing
MA Qianying， born in 2001， M. S. candidate. Her research interests include transportation intelligent control and optimization
Supported by:
National Natural Science Foundation of China(52272329);Development plan of China Railway Corporation(N2023G058)

基于时序知识图谱补全的CTCS-3级列控车载接口设备故障诊断方法

王猛¹, 张大千², 周冰艳³, 马倩影³, 吕继东⁴()

^1.北京和利时系统工程有限公司，北京 100176
^2.中国铁道科学研究院集团有限公司通信信号研究所，北京 100081
^3.北京交通大学自动化与智能学院，北京 100044
^4.北京交通大学轨道交通运行控制系统国家工程研究中心，北京 100044

通讯作者: 吕继东
作者简介:王猛（1983—），男，河南濮阳人，高级工程师，硕士，主要研究方向：铁路信号
张大千（1997—），男，河北保定人，硕士，主要研究方向：铁路信号
周冰艳（2001—），女，河南商丘人，硕士研究生，主要研究方向：交通系统仿真与测试
马倩影（2001—），女，河南平顶山人，硕士研究生，主要研究方向：交通智能控制与优化；
基金资助:
国家自然科学基金资助项目(52272329);中国铁路合作发展计划(N2023G058)

Abstract

Abstract:

Chinese Train Control System level 3 （CTCS-3） train control on-board equipment plays a crucial role in ensuring train safety and improving operational efficiency. On-board interface equipment enables interaction between the on-board Automatic Train Protection （ATP） system， and ground equipment， drivers and trains. However， faults in on-board interface equipment account for a relatively high proportion of on-board equipment faults. In order to identify fault causes and ensure safety， a fault diagnosis method for on-board interface equipment based on temporal knowledge graph completion was proposed. In the method， travel logs and fault statistical data were integrated by introducing the temporal series， which extracted fault phenomena， performed entity alignment， and constructed a temporal knowledge graph. On the basis of the above， a fault diagnosis network based on knowledge graph completion was constructed； Temporal-Translating Embedding （T-TransE） vectorization， and Bidirectional Long Short-Term Memory （Bi-LSTM） network as well as Self-Attention （SA） mechanism were integrated for temporal feature extraction. Finally， the T-TransE vectorization model was pretrained using on-board interface equipment fault data from a railway administration in recent years， and the temporal introduction method with the best effect was selected. In order to validate superiority of the proposed method and effectiveness of the data integration method， the diagnostic network without data integration or temporal relationship introduction， as well as other common fault diagnostic networks， were tested using the on-board fault data. Experimental results show that with the same corpus， the temporal knowledge graph completion-based fault diagnosis model achieves the highest accuracy of 96.69% compared to other fault diagnosis frameworks.

Key words: train operation control system, on-board interface equipment, temporal series, knowledge graph, fault diagnosis

摘要：

CTCS-3级（Chinese Train Control System-3）列控车载设备在保障列车安全和提高运行效率方面发挥着重要作用。车载接口设备实现车载列车自动防护（ATP）系统与地面设备、司机和列车的交互，然而它的故障在车载设备故障中占比高。为了确定故障原因并保证行车安全，提出一种基于时序知识图谱补全的列控车载接口设备故障诊断方法。首先，采用引入时序的方式整合行车日志和故障统计数据，从而提取故障现象并对齐实体，构建时序知识图谱；其次，构建基于图谱补全的故障诊断网络，融合时序翻译（T-TransE）向量化算法、双向长短期记忆（Bi-LSTM）网络和自注意力（SA）机制提取时序特征；最后，使用某铁路局近几年的车载接口设备故障数据对T-TransE向量化模型进行预训练，选出效果最佳的时序引入方式。为验证所提方法的优越性以及数据结合方式的有效性，使用车载故障数据对不进行数据结合且不进行时序关系引入的故障诊断网络以及其他常见的故障诊断网络进行测试。实验结果表明，在同一语料的情况下，与其他故障诊断框架相比，基于时序知识图谱补全的故障诊断模型正确率最高，达到96.69%。

关键词: 列车运行控制系统, 车载接口设备, 时间序列, 知识图谱, 故障诊断

CLC Number:

TP391.1

Meng WANG, Daqian ZHANG, Bingyan ZHOU, Qianying MA, Jidong LYU. Fault diagnosis method for train control on-board interface equipment of CTCS-3 based on temporal knowledge graph completion[J]. Journal of Computer Applications, 2025, 45(2): 677-684.

王猛, 张大千, 周冰艳, 马倩影, 吕继东. 基于时序知识图谱补全的CTCS-3级列控车载接口设备故障诊断方法[J]. 《计算机应用》唯一官方网站, 2025, 45(2): 677-684.

Figures/Tables 18

Fig. 1 Structure of CTCS-3 train control on-board equipment

Fig. 2 Composition of Log statement

Tab. 1 Statistics of fault equipment， causes and phenomena

1.无线通信超时； 2.网络资源不可用；

3.MT硬件故障； 4.MT软件故障

1.实施最大常用制动

2.降级运行

BTM

1.BTM硬件故障； 2.BTM端口无效； 3.BSA错误；

4.全零应答器； 5.应答器丢失

1.启动失败； 2.ATPCU故障、停车； 3.C2CU故障

4.应答器报文错误、制动、停车； 5.主机与DMI通信中断；6.BTM报文错误

TCR

1.TCR传输不良； 2.TCR状态异常

3.TCR信息不合理； 4.TCR硬件故障

1.C2CU安全软件故障、制动停车； 2.TCR故障； 3.载频核对不一致

SDU

1.ODO无服务； 2.雷达故障

3.速传故障； 4.SDU硬件故障

1.速度传感器故障、SDU故障； 2.速度传感器故障

3.雷达故障； 4.ATPCU故障、制动、停车

DMI

1.DMI硬件故障； 2.DMI通信故障

3.DMI软件故障

1.DMI无法启动； 2.ATP故障、制动、停车； 3.DMI显示异常

4.无法输入； 5.主机与人机界面通信暂时终止、停车

VDX

1.继电器故障

2.接口故障

1.启动失败； 2.制动故障；

3.制动测试失败； 4.主机与DMI通信中断、制动、停车

Fig. 3 Knowledge graph triplet

Fig. 4 Interface fault diagnosis framework based on temporal knowledge graph completion

Fig. 5 Interface fault diagnosis flowchart based on temporal knowledge graph completion

Tab. 2 Causes and percentages of faults

故障设备	编号	故障原因	故障占比/%
RTU	FI-1	网络资源不可用	21.3
	FI-2	电台故障	10.9
	FI-3	RBC异常	6.1
TCR	FII-1	TCR硬件故障	4.7
TCR	FII-2	TCR传输异常	5.5
BTM	FIII-1	BAS错误	12.2
BTM	FIII-2	BTM接口故障	7.4
SDU	FIV-1	速传故障	3.2
SDU	FIV-2	雷达故障	7.2
DMI	FV-1	DMI硬件故障	8.2
VDX	FVI-1	继电器故障	6.3
	FVI-2	VDX报文无效	2.9
	FVI-3	VDX端口无效	4.1

Tab. 3 Processed fault data

log_no	relation	equ_no	t
rd network resource available	cause	radio service lost	1
radio service lost	cause	connection closed	1
connection closed	cause	lstm nid=45	0
lstm nid=45	cause	radio timeout	0
radio timeout	cause	connection lost service	1
connection lost service	cause	timeout	0
timeout	equipment_is	RTU	-1

Tab. 4 Entity of fault phenomenon 1 when phe1_no is 1~7

phe1_no	fault_phe	phe1_no	fault_phe
1	无线连接超时	5	列车收到全零应答器
2	列车进入冒进模式	6	DMI报测速雷达故障
3	L5码突降为HU码	7	ATP报应答器一致性
4	DMI报BTM故障

Tab. 5 Parameters of optimal results of three temporal embedding methods

嵌入方式	$r$	$e m b e d d i n g_d i m$	$γ$	$n b a t c h e s$
基于组合	0.001	50	1	128
基于向量	0.001	100	1	64
基于系数	0.001	100	1	64

Tab. 5 Parameters of optimal results of three temporal embedding methods

嵌入方式	$r$	$e m b e d d i n g_d i m$	$γ$	$n b a t c h e s$
基于组合	0.001	50	1	128
基于向量	0.001	100	1	64
基于系数	0.001	100	1	64

Fig. 6 Training losses of three temporal embedding methods

Tab. 6 Evaluation results of three temporal embedding methods

嵌入方式	MRR	Hits@1	Hits@3	Hits@10
基于组合	0.62	0.36	0.47	0.68
基于向量	0.70	0.48	0.59	0.72
基于系数	0.34	0.25	0.33	0.56

Tab. 7 Fault diagnosis network hyperparameters

参数	值	参数	值
batch_size	128	学习率	0.001
epoch	100	学习率衰减因子	0.01
优化器	Adam	Bi-LSTM层Dropout	0.5
文本嵌入维度	100

Tab. 8 Structure setting of fault diagnosis network

层序号	类型	输出	层序号	类型	输出
1	T-TransE	（None，30，300）	5	Dense	（None，100）
2	Bi-LSTM	（None，30，100）	6	Dropout	（None，100）
3	Bi-LSTM	（None，400）	7	Dense	（None，13）
4	Self-Attention	（None，400）

Fig. 7 Loss and accuracy curves on training set and validation set

Fig. 8 Confusion matrix of fault diagnotor

Tab. 9 Control experimental setting

编号	模型	输入	结构	输出
M1	T-TransE- Bi-LSTM-SA^［26］	不引入故障设备的时序知识图谱	T-TransE	（None，30，300）
			Bi-LSTM	（None，30，100）
			Bi-LSTM	（None，400）
			Self-Attention	（None，400）
			Dense	（None，100）
			Dropout	（None，100）
			Dense	（None，13）
M2	Bi-LSTM-SA^［27］	引入故障设备的时序知识图谱	Embedding	（None，30，300）
			Bi-LSTM	（None，30，100）
			Bi-LSTM	（None，400）
			Self-Attention	（None，400）
			Dense	（None，100）
			Dropout	（None，100）
			Dense	（None，13）
M3	Bi-LSTM^［28］	引入故障设备的时序知识图谱	Embedding	（None，30，300）
			Bi-LSTM	（None，30，100）
			Bi-LSTM	（None，400）
			Dense	（None，100）
			Dropout	（None，100）
			Dense	（None，13）
M4	1D CNN-LSTM^［6］	原始故障数据	Embedding	（None，30，300）
			1D CNN	（None，30，100）
			LSTM	（None，400）
			Dense	（None，100）
			Dropout	（None，100）
			Dense	（None，13）
M5	TextCNN^［29］	原始故障数据	Embedding	（None，30，300）
			1D CNN	（None，30，100）
			1D CNN	（None，30，400）
			Flatten	（None，12 400）
			Dense	（None，100）
			Dropout	（None，100）
			Dense	（None，13）

Tab. 10 F1 scores of proposed method and control group

故障类别	本文方法	M1	M2	M3	M4	M5
FI-1	0.99	0.95	0.96	0.95	0.96	0.92
FI-2	0.98	0.92	0.90	0.92	0.90	0.95
FI-3	0.95	0.96	0.95	0.95	0.95	0.90
FII-1	0.95	0.82	0.80	0.77	0.84	0.93
FII-2	0.90	0.87	0.85	0.84	0.83	0.75
FIII-1	0.99	0.90	0.91	0.86	0.82	0.84
FIII-2	0.97	0.92	0.89	0.90	0.90	0.88
FIV-1	0.96	0.95	0.83	0.88	0.89	0.91
FIV-2	0.97	0.91	0.90	0.93	0.82	0.79
FV-1	0.95	0.94	0.89	0.85	0.89	0.90
FVI-1	0.93	0.94	0.95	0.91	0.94	0.90
FVI-2	1.00	0.96	0.94	0.88	0.95	0.92
FVI-3	0.94	0.93	0.90	0.91	0.86	0.90

References 29

1	安玉华. 列控车载设备（CTCS3-300T型）故障分析与研究［D］. 北京：中国铁道科学研究院， 2015： 69-69.
	AN Y H. The failure analysis and research of the train control on-board equipment （CTCS3-300T）［D］. Beijing： China Academy of Railway Sciences， 2015： 69-69.
2	王子琪. 数模驱动的高铁列控车载设备故障传播路径识别与测试［D］. 北京：北京交通大学， 2022.
	WANG Z Q. Identification and testing of fault propagation paths of on-board equipment of CTCS driven by data and model［D］. Beijing： Beijing Jiaotong University， 2022.
3	何涛，韩敬佳. CTCS-3级列控车载设备的形式化建模与验证［J］. 重庆大学学报， 2023， 46（9）：120-129.
	HE T， HAN J J. Formal modeling and verification of CTCS-3 level train control on-board equipment［J］. Journal of Chongqing University， 2023， 46（9）： 120-129.
4	林经源，何涛. 基于STPA和时序逻辑的RBC切换场景安全分析［J］. 电子技术与软件工程， 2023（6）：210-213.
	LIN J Y， HE T. Safety analysis of RBC switching scenarios based on STPA and temporal logic［J］. Electronic Technology and Software Engineering， 2023（6）： 210-213.
5	展鑫，孙颖. 列控系统车-地无线通信协议测试研究［J］. 铁路通信信号工程技术， 2023， 20（10）：30-36.
	ZHAN X， SUN Y. Research on testing and implementation of radio communication protocol for high-speed railway［J］. Railway Signalling and Communication Engineering， 2023， 20（10）： 30-36.
6	ZHANG D， CAO Y， ZHANG M， et al. Fault diagnosis of on-board equipment in CTCS-3 based on CNN-LSTM model［C］// Proceedings of the 2022 China Automation Congress. Piscataway： IEEE， 2022： 5484-5489.
7	杨嘉明. 基于LSTM-BP神经网络的列控车载设备故障诊断方法［D］北京：北京交通大学， 2018： 75-75.
	YANG J M. Fault diagnosis method of the on-board equipment of CTCS based on LSTM-BP neural network［D］. Beijing： Beijing Jiaotong University， 2018： 75-75.
8	HU H， TANG B， GONG X， et al. Intelligent fault diagnosis of the high-speed train with big data based on deep neural networks［J］. IEEE Transactions on Industrial Informatics， 2017， 13（4）： 2106-2116.
9	LIU H， MA R， LI D， et al. Machinery fault diagnosis based on deep learning for time series analysis and knowledge graphs［J］. Journal of Signal Processing Systems， 2021， 93（12）： 1433-1455.
10	董兴芝. 面向智能高铁安全保障的知识图谱构建及应用关键技术研究［D］. 北京：中国铁道科学研究院， 2022.
	DONG X Z. Research on key technology of knowledge graph construction and its application for intelligent high-speed railway safety assurance［D］. Beijing： China Academy of Railway Sciences， 2022.
11	薛莲，姚新文，郑启明，等. 高铁列控车载设备故障知识图谱构建方法研究［J］. 铁道科学与工程学报， 2023， 20（1）：34-43.
	XUE L， YAO X W， ZHENG Q M， et al. Research on construction method of fault knowledge graph of CTCS on-board equipment［J］. Journal of Railway Science and Engineering， 2023， 20（1）： 34-43.
12	李志欣. 简谈CTCS-3级列控系统无线链接超时问题［J］. 铁路通信信号工程技术， 2018， 15（3）：28-33.
	LI Z X. Wireless connection timeout problem of CTCS-3 train control system［J］. Railway Signalling & Communication Engineering， 2018， 15（3）： 28-33.
13	LIU X， TA Z， AN M， et al. Research on suppressing the interference of on-board BTM antenna by installing metal shielding plate［C］// Proceedings of the 5th Asia Conference on Energy and Electrical Engineering. Piscataway： IEEE， 2022： 123-127.
14	彭继宇. 应答器报文数据校验方法研究及软件设计［D］. 成都：西南交通大学， 2022.
	PENG J Y. Research and software design of data verification method for balise telegram［D］. Chengdu： Southwest Jiaotong University， 2022.
15	吕岩琪. CTCS3-300T列控车载应答器传输系统故障分析及处理建议［J］. 铁路通信信号工程技术， 2021， 18（9）：99-103.
	LYU Y Q. Fault analysis and suggestions on fault handling of onboard balise transmission system of CTCS3-300T train control system［J］. Railway Signalling and Communication Engineering， 2021， 18（9）： 99-103.
16	周德豹. 轨道电路读取器的设计原理及应用［J］. 铁道通信信号， 2008， 44（8）：4-5.
	ZHOU D B. Principle and application of track circuit reader［J］. Railway Signalling and Communication， 2008， 44（8）： 4-5.
17	LI M， SUN Z， ZHANG W， et al. Leveraging semantic property for temporal knowledge graph completion［J］. Applied Intelligence， 2023， 53（8）： 9247-9260.
18	SHEN T， ZHANG F， CHENG J. A comprehensive overview of knowledge graph completion［J］. Knowledge-Based Systems， 2022， 255： No.109597.
19	FAN B， SHAO C， LIU Y， et al. A new approach in developing an urban rail transit emergency knowledge graph based on operation fault logs［J］. Journal of Transportation Safety and Security， 2023， 15（10）： 1057-1085.
20	魏伟，赵小强，丁艳华，等. 基于BiLSTM和改进注意力机制的高铁车载设备故障诊断［J］. 铁道学报， 2022， 44（11）：53-62.
	WEI W， ZHAO X Q， DING Y H， et al. Fault diagnosis of vehicle on-board equipment based on BiLSTM and attention mechanism［J］. Journal of the China Railway Society， 2022， 44（11）： 53-62.
21	MOUSSELLY-SERGIEH H， BOTSCHEN T， GUREVYCH I， et al. A multimodal translation-based approach for knowledge graph representation learning［C］// Proceedings of the 7th Joint Conference on Lexical and Computational Semantics. Stroudsburg： ACL， 2018： 225-234.
22	ZHANG D， LV J， LUO Z， et al. Fault diagnosis of CTCS-3 on-board logic function based on SSGAN framework［C］// Proceedings of the IEEE 26th International Conference on Intelligent Transportation Systems. Piscataway： IEEE， 2023： 3349-3354.
23	AHMED S， SAIF A F M S， HANIF， M I， et al. Att-BiL-SL： attention-based Bi-LSTM and sequential LSTM for describing video in the textual formation［J］. Applied Sciences， 2021， 12（1）： No.317.
24	AL-ASWADI F N， CHAN H Y， GAN K H. Automatic ontology construction from text： a review from shallow to deep learning trend［J］. Artificial Intelligence Review， 2020， 53（6）： 3901-3928.
25	SHINDE O， KHOBRAGADE A. Knowledge graph creation on Windows malwares and completion using knowledge graph embedding［C］// Proceedings of the 4th International Conference on Advances in Computing， Communication Control and Networking. Piscataway： IEEE， 2022： 1980-1984.
26	HUANG Z， XU P， LIANG D， et al. TRANS-BLSTM： transformer with bidirectional LSTM for language understanding［EB/OL］. ［2023-05-11］..
27	LI W， QI F， TANG M， et al. Bidirectional LSTM with self-attention mechanism and multi-channel features for sentiment classification［J］. Neurocomputing， 2020， 387： 63-77.
28	SYED F， DI SIPIO R， SINERVO P. Bidirectional Long Short-Term Memory （BLSTM） neural networks for reconstruction of top-quark pair decay kinematics［EB/OL］. ［2023-05-11］..
29	KIM Y. Convolutional neural networks for sentence classification［C］// Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing. Stroudsburg： ACL， 2014： 1746-1751.

[1]	Xinran XU, Shaobing ZHANG, Miao CHENG, Yang ZHANG, Shang ZENG. Bearings fault diagnosis method based on multi-pathed hierarchical mixture-of-experts model [J]. Journal of Computer Applications, 2025, 45(1): 59-68.
[2]	Rui LI, Guanfeng LI, Dezhou HU, Wenxin GAO. Knowledge graph multi-hop reasoning model fusing path and subgraph features [J]. Journal of Computer Applications, 2025, 45(1): 32-39.
[3]	Xueqiang LYU, Tao WANG, Xindong YOU, Ge XU. HTLR： named entity recognition framework with hierarchical fusion of multi-knowledge [J]. Journal of Computer Applications, 2025, 45(1): 40-47.
[4]	Zidong CHENG, Peng LI, Feng ZHU. Potential relation mining in internet of things threat intelligence knowledge graph [J]. Journal of Computer Applications, 2025, 45(1): 24-31.
[5]	Jie WU, Ansi ZHANG, Maodong WU, Yizong ZHANG, Congbao WANG. Overview of research and application of knowledge graph in equipment fault diagnosis [J]. Journal of Computer Applications, 2024, 44(9): 2651-2659.
[6]	Guixiang XUE, Hui WANG, Weifeng ZHOU, Yu LIU, Yan LI. Port traffic flow prediction based on knowledge graph and spatio-temporal diffusion graph convolutional network [J]. Journal of Computer Applications, 2024, 44(9): 2952-2957.
[7]	Yubo ZHAO, Liping ZHANG, Sheng YAN, Min HOU, Mao GAO. Relation extraction between discipline knowledge entities based on improved piecewise convolutional neural network and knowledge distillation [J]. Journal of Computer Applications, 2024, 44(8): 2421-2429.
[8]	Jianjing LI, Guanfeng LI, Feizhou QIN, Weijun LI. Multi-relation approximate reasoning model based on uncertain knowledge graph embedding [J]. Journal of Computer Applications, 2024, 44(6): 1751-1759.
[9]	Youren YU, Yangsen ZHANG, Yuru JIANG, Gaijuan HUANG. Chinese named entity recognition model incorporating multi-granularity linguistic knowledge and hierarchical information [J]. Journal of Computer Applications, 2024, 44(6): 1706-1712.
[10]	Xiaoyan ZHAO, Yan KUANG, Menghan WANG, Peiyan YUAN. Device-to-device content sharing mechanism based on knowledge graph [J]. Journal of Computer Applications, 2024, 44(4): 995-1001.
[11]	Jie GUO, Jiayu LIN, Zuhong LIANG, Xiaobo LUO, Haitao SUN. Recommendation method based on knowledge‑awareness and cross-level contrastive learning [J]. Journal of Computer Applications, 2024, 44(4): 1121-1127.
[12]	Linqin WANG, Te ZHANG, Zhihong XU, Yongfeng DONG, Guowei YANG. Fusing entity semantic and structural information for knowledge graph reasoning [J]. Journal of Computer Applications, 2024, 44(11): 3371-3378.
[13]	Wenjuan JIANG, Yi GUO, Jiaojiao FU. Reasoning question answering model of complex temporal knowledge graph with graph attention [J]. Journal of Computer Applications, 2024, 44(10): 3047-3057.
[14]	Beijing ZHOU, Hairong WANG, Yimeng WANG, Lisi ZHANG, He MA. Recommendation method using knowledge graph embedding propagation [J]. Journal of Computer Applications, 2024, 44(10): 3252-3259.
[15]	Chunlei WANG, Xiao WANG, Kai LIU. Multimodal knowledge graph representation learning： a review [J]. Journal of Computer Applications, 2024, 44(1): 1-15.

Fault diagnosis method for train control on-board interface equipment of CTCS-3 based on temporal knowledge graph completion

基于时序知识图谱补全的CTCS-3级列控车载接口设备故障诊断方法

RichHTML

PDF

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 18

References 29

Related Articles 15

Recommended Articles

Metrics