Experimental Frequency-Based Substructuring for Full-Vehicle NVH Characterization in Electric Vehicles

Electric vehicles introduce new NVH challenges due to high-frequency excitations from e-motors and associated electronics. These excitations, often in the kilohertz range, are primarily transmitted as structure-borne noise and require advanced experimental methods for accurate characterization and mitigation. Traditional simulation approaches are insufficient to capture the true dynamic behavior of complex components under operational boundary conditions. This study presents a fully experimental Frequency-Based Substructuring (FBS) methodology applied to a complete electric vehicle. Frequency Response Functions (FRFs) of individual subsystems—such as the e-motor, mounting system, inverter housing, and body connection points—are measured and transformed using Virtual Point Transformation to obtain consistent six-degree-of-freedom representations. These Virtual Point models are assembled into a full-vehicle system to investigate the structure-borne transfer paths to the vehicle interior. The substructuring approach enables transparent decomposition of source-path-target relationships, providing deep insight into how dynamic interactions between subsystems influence interior acoustic response. Special attention is given to identifying dominant paths of high-frequency noise transmission, understanding local resonances introduced by mounting interfaces, and evaluating the contribution of individual components to cabin noise. Experimental validation is performed at system level, ensuring the accuracy of the hybrid model across a wide frequency spectrum. The method enables reliable NVH performance assessments in the early design phase and supports data-driven optimization of isolation strategies without reliance on numerical simulation or FE modeling.