U-Net-Driven Method for Identifying Vehicle Moving Forces from Bridge Displacement Data

The structural safety of bridges depends not only on their inherent performance but also on the external loads they endure. Traffic loads serve as the primary source of live loads during bridge operation, making accurate identification of vehicle-bridge interaction forces critical for bridge design and assessment. Current Moving Force Identification (MFI) faces challenges such as noise sensitivity, prolonged computation times, and reliance on vehicle speed, axle count, and axle spacing information. This study proposes a U-Net-based vehicle dynamic load identification method that utilizes structural response data to estimate dynamic axle loads without requiring prior information such as axle count or spacing. The method's accuracy and robustness are validated through real-bridge vehicle tests, demonstrating its feasibility for engineering applications.