SCIENCE CHINA Information Sciences, Volume 64 , Issue 1 : 119203(2021) https://doi.org/10.1007/s11432-018-9543-8

Fault diagnosis of high-speed train bogie based on LSTM neural network

More info
  • ReceivedMay 19, 2018
  • AcceptedAug 3, 2018
  • PublishedMar 11, 2020


There is no abstract available for this article.


This work was financially aided by National Natural Science Foundation of China (Grant Nos. 61773323, 61603316, 61433011 61733015) and Fundamental Research Funds for the Central Universities (Grant No. 2682018CX15).


[1] Bai W, Yao X, Dong H. Mixed H ?/H fault detection filter design for the dynamics of high speed train. Sci China Inf Sci, 2017, 60: 048201 CrossRef Google Scholar

[2] Mao Z H, Zhan Y H, Tao G, et al. Sensor fault detection for rail vehicle suspension systems with disturbances and stochastic noises. IEEE Trans Veh Technol, 2016, 66: 4691-4705. Google Scholar

[3] Mao Z, Tao G, Jiang B. Adaptive Compensation of Traction System Actuator Failures for High-Speed Trains. IEEE Trans Intell Transp Syst, 2017, 18: 2950-2963 CrossRef Google Scholar

[4] Graves A, Schmidhuber J. Framewise phoneme classification with bidirectional LSTM and other neural network architectures.. Neural Networks, 2005, 18: 602-610 CrossRef PubMed Google Scholar

[5] Kingma D P, Ba J. Adam: a method for stochastic optimization. In: Proceedings of International Conference on Learning Representations, SanDiego, 2014. 5. Google Scholar

  • Figure 1

    (Color online) (a) The workflow of LSTM network. (b) The location of sensors in the simulation model. The samples which generated by (b) are inputs to neural network (a). (c) The algorithm flow based on LSTM network. (d) Confusion martix for the fault diagnosis on the test set. The labels of true and predicted class are given in (e).