The crashworthiness of auto-body structures is an indispensable issue in the automotive industry. The automotive industry has made an effort to develop light-weight auto-body structures to increase the fuel efficiency and to satisfy the emission gas regulation of vehicles. Since the weight reduction of an auto-body should not obstruct the safety improvement, the crash analysis has to be accurately conducted for efficient reducing of the auto-body weight. The vehicle body structures are generally composed of various members such as frames, stamped panels and deep-drawn parts from sheet metals. The strength of sheet metals depends on the rate of deformation and the dynamic behavior of sheet metals is a key to investigate the impact characteristics of the structure. As the dynamic behavior of a material is different from the static one due to the inertia effect and the stress wave propagation, an adequate experimental technique has to be developed to obtain the dynamic response for the corresponding level of the strain rate. To acquire the high strain-rate material properties of sheet metals, a tension type split Hopkinson bar apparatus specially designed for sheet metals was used. The tension split Hopkinson bar inevitably causes some errors in the strain at grips for the plate type specimens, since the grip and specimen disturb the one-dimensional wave propagation in bars. Validation of experiment is carried out with the error analysis that is estimated by comparing the waves acquired from experiments with the one from the Pochhammer-Chree solution. The optimum geometry of the specimen is determined to minimize the error from the loading equilibrium.
The dynamic response of sheet metals at the high strain-rate is obtained from the tensile split Hopkinson bar test using plate type specimens. Experimental results from both quasi-static and dynamic tensile tests with the tensile split Hopkinson bar apparatus are interpolated to construct the Johnson-Cook and a modified Johnson-Cook equation as the constitutive relation that should be applied to simulation of the dynamic behavior of auto-body structures. The constitutive relation obtained is applied to simulation of thin-walled structures by an elasto-plastic explicit finite element method with shell elements incorporating with a contact algorithm. The present algorithm adopts the plastic predictor-elastic corrector (PPEC) scheme in stress integration in order to accurately keep track of the stress-strain relation for the rate-dependent model. To verify the finite element code developed with the dynamic constitutive relation, the dynamic crash experiment is conducted with square tubes made from the sheet metals. Comparison between the experimental and analysis results show that the dynamic mean crash loads of square tubes calculated with dynamic constitutive model are more accurate than the one with quasi-static model.
To evaluate the crashworthiness of a car, the dynamic response of auto-body components has to be correctly simulated for various loading conditions. The crash analysis is performed for auto-body structures such as a hood and frontal frames of an automobile. The finite element model of an automobile is adopted from the web site of NHTSA (National Highway Traffic Safety Administration). The vehicle crashworthiness is greatly affected by side members and s-rails that are the energy absorbing structures with the axial and bending collapse mode. These structures are designed to effectively dissipate the vehicle kinetic energy in terms of plastic deformation in favor of the safety of passengers. Simulation is carried out with the dynamic and the quasi-static model of sheet metals in order to compare the results from both models. Results show remarkable difference in the reaction force and impact energy absorption. The reason is considered mainly due to the strain rate effect. Simulation results also provide the deformed shape and the deformation energy in order to predict and evaluate the crashworthiness of a car.
