摘要 : Computer models are widely used to simulate dynamic systems in automobile industry. It is imperative to have high quality CAE models with good predictive capability. This requires CAE engineers to conduct model calibration with ph... 展开 Computer models are widely used to simulate dynamic systems in automobile industry. It is imperative to have high quality CAE models with good predictive capability. This requires CAE engineers to conduct model calibration with physical tests. The challenges in the occupant restraint system model calibration are: (1) the dynamic system usually consists of multiple responses, (2) most of the responses are functional data or time histories, and (3) the traditional trial-and-error calibration approach is time consuming and highly depends on analyst's expertise. These call for the development of an automatic and effectivemodel calibration method. This paper presents a newly developed automatic model calibration method, based on the Error Assessment of Response Time Histories (EARTH) metric. The EARTH metric is used to perform model assessment on various important features of the functional responses. A new multi-objective optimization problem is formulated and solved by a Non-dominated Sorting Genetic Algorithm to automatically update CAE model parameters. A real-world example is used to demonstrate the use of the proposed method. 收起