Vehicle Interior Noise Prediction Based on Elman Neural Network

Vehicle interior noise is an important factor affecting ride comfort. To reduce the inside vehicle at body design stage, a finite element model of must be established. While taking first-order global modal body-in-white, maximum sound pressure level target point in vehicle, mass, and side impact conditions into account, thickness panel as determined via sensitivity analysis treated input variable, sample by following Hamersley experimental design. Specifically, Elman neural network predicts value structure optimization method that comprehensively considers NVH performance safety The prediction errors algorithm were within 3%, which meets accuracy requirements. achieve satisfactory restraint performance, reduced 5.92 dB, intrusions two points on B-pillar inner are 31.1 mm 33.71 mm, respectively. improved while reduced. This study provides reference for multidisciplinary research aiming to optimize structures.