Study on life prediction method for rail vehicle critical components based on deep learning models and track load spectra

Abstract

Deep learning and fatigue life prediction remain focal research areas in rail vehicle engineering. This study addresses the vibration fatigue of wheelset lifting lug in Chengdu Metro Line 1 bogies, aiming to develop a fatigue life prediction method for critical bogie components using deep learning models and measured track load spectra. Extensive field tests on Chengdu Metro Line 1 were conducted to acquire acceleration and stress response data of the wheelset lifting lug, generating training samples for the neural network system. Component stress responses were calculated via time-domain track acceleration and validated against in-situ stress measurements. Results show that neural network-fitted dynamic stress values exhibit excellent consistency with measured data, with errors constrained within 5 %. This study validates the proposed small-sample deep learning approach as an effective and accurate solution for fatigue life prediction of critical bogie components under operational load conditions.

Keywords

Railway vehicle / Deep learning / Neural network / Life prediction / Vibration fatigue

References

[1] Y. Cao, Z. Chen, T. Wen, et al., Rail fastener detection of heavy railway based on deep learning, High-speed Railw. 1 (1) (2023) 63-69.

[2] B. Indraratna, M. Babar Sajjad, T. Ngo, et al., Improved performance of ballasted tracks at transition zones: A review of experimental and modelling approaches, Transp. Geotech. 21 (2019) 100260.

[3] X. Wei, X. Yin, Y. Hu, et al., Squats and corrugation detection of railway track based on time-frequency analysis by using bogie acceleration measurements, Veh. Syst. Dyn. 58 (2020) 1167-1188.

[4] A. Rodríguez, R. Sanudo, M. Miranda, et al., Smartphones and tablets applications in railways, ride comfort and track quality. Transition zones analysis, Measurement 182 (2021) 109644.

[5] H. Chen, N. Hu, Z. Cheng, et al., A deep convolutional neural network based fusion method of two-direction vibration signal data for health state identification of planetary gearboxes, Measurement 146 (2019) 268-278.

[6] M. Azamfar, J. Singh, I. Bravo-Imaz, et al., Multisensor data fusion for gearbox fault diagnosis using 2D convolutional neural network and motor current signature analysis, Mech. Syst. Signal Process. 144 (2020) 106861.