|
Coupled Systems Mechanics Volume 14, Number 6, December 2025 , pages 507-530 DOI: https://doi.org/10.12989/csm.2026.14.6.507 |
||
|
Dynamic response prediction of vehicle-bridge system under random excitation based on SSA-LSTM model |
||
Tian Zhang, Yuanzhu Liu, Pengfei Li and Yunfeng Zou
|
||
| Abstract | ||
| To establish digital twin model of the vehicle-bridge interaction system under random track irregularity excitation, it is necessary to compute the system response in real time. Traditional methods are time-intensive and lack real-time capability, whereas surrogate model-based approaches can rapidly and accurately predict dynamic response. This study proposes a surrogate model that employs the Sparrow Search Algorithm to optimize Long Short-Term Memory neural networks for predicting the dynamic response of vehicle-bridge interaction system under random excitation. Initially, a physical model of the vehicle-bridge interaction system is established, incorporating track irregularities to calculate the dynamic response and generate training samples. Subsequently, an SSA-LSTM surrogate model is developed and trained. Finally, the surrogate model is utilized to predict the dynamic response of the vehicle-bridge interaction system under arbitrary track irregularity excitations. To validate the robustness of the proposed algorithm, the prediction results of various surrogate models are compared. The results indicate that the proposed surrogate model achieves higher computational efficiency compared to classical mechanical models of the vehicle-bridge interaction system. Moreover, the SSA-LSTM surrogate model outperforms traditional LSTM and Backpropagation surrogate models in terms of prediction accuracy for the dynamic response of the vehicle-bridge interaction system. | ||
| Key Words | ||
| dynamic response prediction; random excitation; sparrow search algorithm; surrogate model; vehicle-bridge interaction system | ||
| Address | ||
| Tian Zhang, Yuanzhu Liu: Transportation Engineering College, Dalian Maritime University, Dalian 116026, China Pengfei Li: Research Institute of Highway Ministry of Transport, Beijing 100088, China Yunfeng Zou: National Engineering Research Center of High-speed Railway Construction Technology, Changsha 410075, China; School of Civil Engineering, Central South University, Changsha 410075, China | ||
