Abstract

The three explicit nonlinear Finite Element (FE) codes, LS-DYNA, PAM-CRASH, and RADIOSS are widely used by the automotive companies and government for vehicle and roadside hardware crash simulations. As simulation results from these commercial codes gain wider acceptance for crashworthiness evaluations, the codes also are becoming more sophisticated, including advanced features and options for the user than ever before. While these advanced modeling and simulation features are beneficial to users, they also raise a question regarding the trade off between the FE accuracy and the associated computational time and cost required.

The main objective of this paper is to investigate the effects of the major modeling and simulation features of the commercial FE codes and its direct impact on the simulation accuracy and computational efficiency when solving crash problems. The four major parameters considered in this study are element size/mesh density, shell element formulations, hourglass control form, and integration time step strategy. An intensive parametric study is conducted through FE simulations for quasi-static crush of a simple box beam. The nonlinear, dynamic, explicit FE code, LS-DYNA has been used to perform the parametric study. The accuracy of the FE simulations are measured by comparing the simulation results with experimental test results conducted, in the literature, for the same box beam.

Based on the results of the parametric study, final recommendations and guidelines for choosing the major simulation parameters are offered, to compromise between the FE simulation accuracy and the associated computational time and cost.

This content is only available via PDF.
You do not currently have access to this content.