Prediction of Failure Rate of Metro Vehicle Bogie Based on Neural Network

  • Xiuqi Wang
  • Yong QinEmail author
  • Yong Fu
  • Meng Ye
Conference paper
Part of the Lecture Notes in Electrical Engineering book series (LNEE, volume 617)


Metro train bogie system is located between the car body and the track, which is one of the key subsystems to ensure the safety of train operation. As a complex system, bogie system is composed of many components, once the failure happened, it would impact the normal operation of the whole train. In order to better predict the failure rate of bogie system, radial basis function (RBF) neural network is introduced to predict the failure rate of the whole system through the fault data of bogie components, and genetic algorithm is used to optimize the model. Experimental results showed that the proposed method can accurately predict the bogie failure rate, and can be used as a system-level reliability prediction method, providing a data basis for later system improvement and optimization.


Bogie Genetic algorithm RBF neural network Prediction of failure rate 



The authors gratefully acknowledge the financial supports for this research from National Natural Science Foundation of China (61833002).


  1. 1.
    Xia J (2014) Research on reliability analysis and application of metro door system. Nanjing University of Science and Technology, NanjingGoogle Scholar
  2. 2.
    Hu C, Yao J (2012) Reliability analysis for electric multiple units based on fault tree Monte Carlo method. China Railway Sci z1:52–59Google Scholar
  3. 3.
    Wang H, Lu Z, Zhang B (2012) Analysis method for the operational reliability of EMU running gear based on fault tree and Bayesian network. China Railway Sci z1:60–64Google Scholar
  4. 4.
    Meng L, Liu Z, Diao L et al (2016) Reliability evaluation of high-speed train traction transmission system based on Markov model. J China Railway Soc 8:23–27Google Scholar
  5. 5.
    Qin Y, Fu Y, Li W et al (2018) Operational safety and reliability assessment of high speed train with intuitionistic fuzzy set and VIKOR method. J Beijing Univ Technol 44(1):112–119Google Scholar
  6. 6.
    Li L, Cheng X, Qin Y et al (2013) Reliability prediction of urban rail transit vehicle based on BP neural network. J Central South Univ (Science and Technology) 1:42–46Google Scholar
  7. 7.
    Zhou Q, Deng Y, Chen J (2010) The RBFNN application in fault diagnosis for the subway train. Comput CD Softw Appl 11:71–72Google Scholar
  8. 8.
    Song Y, Zhu M (2014) Subway sensor fault diagnosis based on radial neural network. Urban Mass Transit 5:94–97, 101Google Scholar
  9. 9.
    Yin H, Wang K, Zhang T (2015) Fault prediction based on PSO-BP neural network about wheel and Axle Box of Bogie in urban rail train. Complex Syst Complexity Sci 4:97–103Google Scholar
  10. 10.
    Hartman E, Keeler JD, Kowalski JM (2008) Layered neural networks with Gaussian hidden units as universal approximations. Neural Comput 2(2):210–215CrossRefGoogle Scholar
  11. 11.
    Zhou P (2013) Design and application of neural network based on Matlab. Tsinghua University Press, BeijingGoogle Scholar
  12. 12.
    Yu J (2012) Reliability analysis and application research of key system of metro vehicles. Beijing Jiaotong University, BeijingGoogle Scholar
  13. 13.
    Zeng Y (2015) RBF natural network based on genetic algorithm used in maximum power point tracking of photovoltaic system. Hunan University of Technology, ZhuzhouGoogle Scholar
  14. 14.
    Chen A (2007) Research on data prediction method based on BP and RBF neural networks. Central South University, NanjingGoogle Scholar
  15. 15.
    Chen M (2006) Study of submarine’s displacement and principal dimensions by using GA based optimum RBF neural network. Huazhong University of Science and Technology, WuhanGoogle Scholar
  16. 16.
    Shi J (2014) Study of Bogie failure prediction and maintenance of urban rail train based on reliability analysis. Beijing Jiaotong University, BeijingGoogle Scholar

Copyright information

© Springer Nature Singapore Pte Ltd. 2020

Authors and Affiliations

  1. 1.National Key Laboratory of Rail Control and Safety of Beijing Jiaotong UniversityBeijingChina
  2. 2.Beijing Engineering Research Center of Urban Traffic Information Intelligent Sensing and Service TechnologiesBeijingChina

Personalised recommendations