Dynamic Modal Evolution of High-Speed Train Car Bodies Under Complex Boundary and Load Conditions: A Field Test Study

Stochastic Subspace Identification (SSI) theory offers the distinct advantage of extracting modal parameters directly from operational ambient excitations without requiring artificial force, ensuring completely true boundary conditions and providing extensive field measurement data. In this study, we systematically investigate the operational modal characteristics of Electric Multiple Units (EMUs) in the Chinese high-speed railway network under multi-dimensional coupling conditions, including wide speed ranges, axle load perturbations, air spring faults, and coupled operation. The results reveal that while car body modal frequencies remain largely insensitive to operating speed—indicating negligible effects of aerodynamic stiffness—they exhibit distinct sensitivities to mass and boundary variations. Specifically, an increase in axle load induces a significant attenuation (exceeding 5%) in low-order vertical bending frequencies, conforming to the dynamic mass law. Conversely, air spring deflation triggers a sharp increase in boundary stiffness, resulting in a 13.6% surge in torsional modal frequency, which serves as a critical indicator for fault diagnosis. Furthermore, coupled operation is found to primarily enhance system damping. Based on these findings, we establish a “condition-modal” vehicle sensitivity matrix, quantifying dynamic evolution mechanisms under complex boundaries and providing a vital baseline for monitoring the structural health of railway vehicles and conducting intelligent maintenance.