|About this Abstract
||NUMISHEET 2021: The 12th International Conference on Numerical Simulation of 3D Sheet Metal Forming Processes
||An Artificial Intelligence Tool for Predicting the Crash Pulse Response in a Crashworthiness Application of Vehicle Lightweighting
||Christopher Kohar, Daniel S Connolly, Timofei Liusko, Kaan Inal
|On-Site Speaker (Planned)
Automakers are tasked with the challenge of selecting the optimal combination of material properties to produce next-generation lightweight vehicles that maintain structural integrity while minimizing mass and cost. Vehicle designers rely on finite element (FE) simulations of vehicle performance to evaluate the suitability of a material substitution. These virtual studies can produce a vast amount of data that can be leveraged by artificial intelligence (AI) techniques to accelerate the design process. This work presents an AI framework to predict the vehicle crashworthiness response to different material properties. A virtual experimental study of FE simulations of the frontal crash condition of a pick-up truck with different material compositions are used to generate the data for this method. A new type of recurrent neural network is used to relate the occupant crash-pulse response, which is a key crashworthiness metric, to the change in material properties throughout the vehicle.
||Definite: At-meeting proceedings