System parameter identification of dynamic system using surrogate AI/ML technique

In the automobile industry, intense competition often necessitates final design judgments in the early stages of product release. When these designs are built on physical vehicles, variations can occur due to tolerances, the fitment of stiffness elements, and other parameters. These variations directly influence the dynamic behavior of the product, which may differ from its digital twin created during the design phase. Repeatedly measuring all these system parameters can be costly or physically impractical. After identifying some system parameters by performing five dynamic load cases on a 4-poster under controlled environments and acquiring the dynamic response at different suspension locations of the passenger vehicle. These target responses are then correlated with the responses from the digital twin to identify the optimal combination of digital model parameters using surrogate AI/ML models in modeFRONTIER.