Sensitivity Analysis and Interval Multi-Objective Optimization for an Occupant Restraint System Considering Craniocerebral Injury
Qiming Liu,Xingfu Wu,Xu Han,Jie Liu,Zheyi Zhang,Shijie Guo
DOI: https://doi.org/10.1115/1.4044328
2019-10-17
Journal of Mechanical Design
Abstract:Abstract In vehicle collision accidents, an occupant restraint system (ORS) is crucial to protect the human body from injury, and it commonly involves a large number of design parameters. However, it is very difficult to quantify the importance of design parameters and determine them in the ORS design process. Therefore, an approach of the combination of the proposed approximate sensitivity analysis (SA) method and the interval multi-objective optimization design is presented to reduce craniocerebral injury and improve ORS protection performance. First, to simulate the vehicle collision process and obtain the craniocerebral injury responses, the integrated finite element model of vehicle-occupant (IFEM-VO) is established by integrating the vehicle, dummy, seatbelt, airbag, etc. Then, the proposed approximate SA method is used to quantify the importance ranking of design parameters and ignore the effects of some nonessential parameters. In the SA process, the Kriging metamodel characterizing the relationships between design parameters and injury responses is fitted to overcome the time-consuming disadvantage of IFEM-VO. Finally, according to the results of SA, considering the influence of uncertainty, an interval multi-objective optimization design is implemented by treating the brain injury criteria (BRIC, BrIC) as the objectives and regarding the head injury criterion (HIC) and the rotational injury criterion (RIC) as the constraints. Comparison of the results before and after optimization indicates that the maximum values of the translational and rotational accelerations are greatly reduced, and the ORS protection performance is significantly improved. This study provides an effective way to improve the protection performance of vehicle ORS under uncertainty.
engineering, mechanical