Research on fault prediction model of ZPW-2000A track circuit based on LSTM algorithm

Chong Zhen Zhang

doi:10.1117/12.2639909

19 October 2022 Research on fault prediction model of ZPW-2000A track circuit based on LSTM algorithm

Chong Zhen Zhang

Proceedings Volume 12294, 7th International Symposium on Advances in Electrical, Electronics, and Computer Engineering; 1229457 (2022) https://doi.org/10.1117/12.2639909
Event: 7th International Symposium on Advances in Electrical, Electronics and Computer Engineering (ISAEECE 2022), 2022, Xishuangbanna, China

Abstract

Track circuit is an important basic equipment to ensure safe and reliable train operation. With the transformation from "fault repair" to "state repair" of railway signal equipment, a fault prediction model method based on LSTM algorithm was proposed in order to analyze track circuit monitoring information and formulate scientific and reasonable maintenance strategy. Firstly, the structural principle of track circuit is analyzed, and then LSTM model is proposed to predict its faults. In order to verify the prediction effect of the model, MAE, MAPE, NMSE and REP are used to measure the predicted results. Secondly, the prediction effect of the improved algorithm is compared with that of the traditional SVM algorithm, BP neural network and RNN algorithm, and the results show that all indicators of LSTM are optimal. Finally, an example is used to predict LSTM, which verifies the efficiency of the model.

Citation Download Citation

Chong Zhen Zhang "Research on fault prediction model of ZPW-2000A track circuit based on LSTM algorithm", Proc. SPIE 12294, 7th International Symposium on Advances in Electrical, Electronics, and Computer Engineering, 1229457 (19 October 2022); https://doi.org/10.1117/12.2639909

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available