Fault diagnosis for on-board equipment of train control system based on BERT+CNN_BiLSTM

Abstract

The on-board equipment as core equipment of train control system plays an important role in the process of high-speed train operation. At present, its fault diagnosis only depends on the experience of on-site operators and diagnosis efficiency is relatively low. To realize automatic fault diagnosis and improve diagnosis efficiency of the on-board equipment of train control system, a fault diagnosis model called BERT+CNN_BiLSTM was proposed, which combined bidirectional encoder representations from transformers(BERT) model, convolutional neural network(CNN) and bidirectional long short-term memory(BiLSTM). Firstly, the BERT model was used to transform the application event log(AElog) into a text vector representation that can mine semantic information recognized by computer. Secondly, CNN and BiLSTM were used to extract fault features and combine them to enhance spatial and temporal capability of the model. Finally, fault classification and diagnosis of on-board equipment of train control system was realized by using Softmax. In the experiment, taking an actual on-board equipment as the research object, the AElog generated during the train operation was selected as texperimental data to verify the performance of BERT+CNN_BiLSTM model. The results showed that compared with traditional machine learning algorithm, BERT+BiLSTM model and BERT+CNN model, the pecision, recall and F1 of BERT+CNN_BiLSTM model were 92.27%, 91.03% and 91.64%, respectively, which indicates that the proposed BERT+CNN_BiLSTM model has a better overall performance in the fault diagnosis of on-board equipment of high-speed train control system.

Keywords

on-board equipment / fault diagnosis / bidirectional encoder representations from transformers(BERT) / application event log(AElog) / bidirectional long short-term memory(BiLSTM) / convolutional neural network(CNN)

References

[1] LIANG X, WAN H F, GUO J, et al. Fault diagnosis method for train control on-board equipment based on Bayesian network.Journal of Railways, 2017,39(8): 93-100.

[2] LIU H. Research on fault diagnosis method for CTCS on-board equipment of high-speed railway based on association rules. Beijing: Beijing Jiaotong University, 2018.

[3] YANG L B, SHEN X, LI X Q, et al. Research on fault classification model of high-speed railway turnout based on text analysis. China Railway, 2020(8): 13-18.

[4] YANG J M. Fault diagnosis method for train control on-board equipment based on LSTM-BP neural network. Beijing: Beijing Jiaotong University, 2018.

[5] ZHOU L J, DANG J W, WANG Y X, et al. Research on fault classification of train control on-board equipment based on convolutional neural network. Journal of Railways, 2021, 43(6): 70-77.

[6] HINTON G E, SALAKHUTDINOV R R. Reducing the dimensionality of data with neural networks. Science, 2006, 313(5786): 504-507.

[7] WANG N Y, YE Y X, LIU L, et al. Research progress of language model based on deep learning. Journal of Software, 2021,32: 1082-1115.

[8] HE L, ZHENG Z X, XIANG F T, et al. Research progress of text classification technology based on deep learning. Computer Engineering, 2021, 47(2): 1-11.

[9] DAI L R, ZHANG S L, HUANG Z Y. Current situation and prospect of speech recognition technology based on deep learning. Data Collection and Processing, 2017, 32(2): 221-231.

[10] DEVLIN J, CHANG M W, LEE K, et al. BERT: Pre-training of deep bidirectional transformers for language understanding.2023-03-04].

[11] GUO Y Z, ZONG D G. Principle and application of CTCS-3 train operation control system. 1st edition. Beijing: China Railway Press, 2014.

[12] WANG X X. Failure analysis of CTCS3-300T train control vehicle equipment. Science and Technology Innovation and Application, 2017(3): 296.

[13] China Railway Corporation. Typical failure cases of train control on-board equipment. 1st edition. Beijing: China Railway Press, 2013.

[14] VASWANI A, SHAZEER N, PARMAR N, et al. Attention is all you need//31st International Conference on Neural Information Processing Systems, December 4-9, 2017, Long Beach, California, USA. New York: Curran Associates Inc., 2017: 5998-6008.

[15] HOCHREITER S. The vanishing gradient problem during learing recurrent neural nets and problem solutions. International Journal of Uncertainty, Fuzziness and Knowledge-Based Systems, 1998, 6(2): 107-116.

[16] XU G X, MENG Y T, QIU X Y, et al. Sentiment analysis of comment texts based on BiLSTM. IEEEE Access, 2019, 7: 51522-51532.