To address the issue of long periods in extracting load spectra based on virtual iteration (VI), this paper presents a method for extracting chassis load spectra based on a spatial attention gated recurrent unit (SAGRU) neural network. Firstly, load spectra such as the six-axis wheel force, the acceleration and the displacement on multiple monitoring locations were acquired through reinforced proving ground testing. Then, a full vehicle multi-body dynamics (MBD) model is established, and the vertical displacement excitation at the wheel center under road test conditions is inversely solved based on SAGRU. Finally, the response of monitoring locations and chassis loads under road test conditions are calculated through MBD simulation analysis using the vertical displacement excitation and the five-component force in other directions as input. By comparing with the virtual iteration method for obtaining load spectra, it is shown that the proposed method achieves a 22.5% efficiency improvement while ensuring accuracy.
JINH J, ZHOUZ S, ZANGL G,et al .Simulation and optimization method of vehicle frame structure based on measured load spectrum and its application[J].Journal of Vibration and Shock,2023,42(1):181-189.(in Chinese)
DONGG J, YANF, HANJ,et al. Load spectrum extraction method for automobile suspension components[J].Journal of Vibration and Shock,2021,40(2):103-110.(in Chinese)
[7]
ZHOUD L, CHANGJ L .Fatigue analysis of a light truck rear axle based on virtual iteration method[J].Shock and Vibration,2022, 2022(1): 8598491.
[8]
FANGJ, LIX W, ZHANGD L,et al .Research on load reverse engineering and vibration fatigue analysis technology of rapid box wagon[J]. Materials,2022,15(23):8322.
[9]
LIUR X, DOBRIBANE, HOUZ C,et al. Dynamic load identification for mechanical systems:a review[J]. Archives of Computational Methods in Engineering,2022,29(2): 831-863.
[10]
GAOY D, YUX Y, CHENL K,et al .Impact load identification algorithm of helicopter weapon pylon based on time-domain response signal[J]. Aerospace,2022,9(7):388.
[11]
ZHOUJ, DONGL L, GUANW,et al .Impact load identification of nonlinear structures using deep recurrent neural network[J]. Mechanical Systems and Signal Processing,2019,133:106292.
[12]
YANGH J, JIANGJ H, CHENG P,et al. A recurrent neural network-based method for dynamic load identification of beam structures[J]. Materials, 2021, 14(24): 7846.
[13]
CANDONM, ESPOSITOM, FAYEKH,et al .Advanced multi-input system identification for next generation aircraft loads monitoring using linear regression,neural networks and deep learning[J]. Mechanical Systems and Signal Processing,2022,171: 108809.
[14]
DONGM Y, CHENS Q, XIEB,et al .A quantitative detection method for wheel polygonization of heavy-haul locomotives based on a hybrid deep learning model[J].Measurement,2024,227:114206.
[15]
SCHMIDHUBERJ. A local learning algorithm for dynamic feedforward and recurrent networks[J]. Connection Science,1989, 1(4): 403-412.
[16]
HOCHREITERS, SCHMIDHUBERJ. Long short-term memory[J]. Neural Computation,1997,9(8): 1735-1780.
WUX J, WEIY H, WANGA C,et al .Vehicle trajectory prediction based on LSTM-GRU integrating Dropout and attention mechanism[J]. Journal of Hunan University (Natural Sciences),2023, 50(4): 65-75.(in Chinese)
[19]
YANGK, DINGY L, GENGF F,et al .A multi-sensor mapping Bi-LSTM model of bridge monitoring data based on spatial-temporal attention mechanism[J]. Measurement,2023,217:113053.
[20]
DINGY K, ZHUY L, FENGJ,et al. Interpretable spatio-temporal attention LSTM model for flood forecasting[J]. Neurocomputing, 2020, 403: 348-359.
ZENGJ W, JIY J, RENL H, et al. Radial load identification method of APM vehicle tires based on 1D CNN and Bi-GRU[J]. China Mechanical Engineering, 2023, 34(3): 359-368.(in Chinese)
[23]
SANDHUH K, BODDAS S, YANE,et al .A comparative study on deep learning models for condition monitoring of advanced reactor piping systems[J]. Mechanical Systems and Signal Processing, 2024, 209: 111091.
[24]
ZHANGC, WANR Q, HEJ R,et al. Non-stationary vibration fatigue life prediction of automotive components based on long short-term memory network[J]. International Journal of Fatigue,2024,187: 108459.
基金资助
国家重点研发计划资助项目(2020YFB1807204)
National Key Research and Development Program of China(2020YFB1807204)
国家自然科学基金资助项目(U2001213)
National Natural ScienceFoundation of China(U2001213)
江西省人工智能交通信息传输与处理重点实验室资助项目(20202BCD42010)
Jiangxi Key Laboratory of Artificial Intelligence Transportation Information Transmission and Processing(20202BCD42010)