The frontal airbag in a vehicle is considered supplemental to the safety belt restraint system and is important in lowering measured injury assessment values for Anthropomorphic Test Devices (ATD) during vehicle crash testing. The probability of neck and chest injuries is an important factor for a vehicle’s performance rating under the United States-New Car Assessment Program (US-NCAP) protocol. A shorter lower tether was incorporated into the driver frontal airbag (DAB) to mitigate chest deformation injury, however higher neck injuries were observed with this change.
The purpose of this study is to identify the main factors influencing neck injury assessment values through the use of Design Of Experiments (DOE) techniques and find an optimum airbag design which mitigates neck and chest injury assessment values by using optimization techniques. Four different airbag designs were used in the first stage of the DOE, and one DAB design was chosen for the best performance in US-NCAP. Traditional meta model based optimization of the chosen DAB design followed.
The direct optimization method requires a great deal of computational resource, whereas meta model based optimization methods use comparatively little computational resource once there are sufficient sample data from the DOE. Dynamic meta model based optimization methods were introduced with combined CAE runs to reduce computing resource in this study. CAE runs were periodically sampled to update the meta model and provide improved accuracy. Two different optimization methods with dynamic meta models were demonstrated and compared with traditional meta model based optimization.