Search Results

In the current highway vehicle accident environment a large percentage of fatalities and injuries occur in frontal offset crashes. Computational modeling is being used in support of crashworthiness studies at the National Highway Traffic Safety Administration (NHTSA) to develop an understanding of structural response in offset crash events. This paper presents a lumped-parameter approach to modeling offset impacts. Applications for offset models are discussed, including providing inputs to occupant simulations and as the basis of multipurpose frontal impact models. The role of discretization errors, nonuniaxial motions, and test data dependence and sensitivity in limiting model accuracy is discussed. Extensions to the modeling methodology that will better support offset events are identified.

The SISAME methodology is a system for extracting one-dimensional lumped parameter vehicle crash models from non-oblique crash test data, and for simulation of such models. Model extraction is based on constrained least squares optimization of an overdetermined system of target equations for the model parameters. The SISAME computer program performs extraction and simulation with a number of features that allow user control of the computations and outputs. Additional computer programs perform data assessment/correction and filtering. Experience has shown that the SISAME methodology can efficiently produce predictively useful models that accurately capture the motions of the actual crash event. The essential formulation of SISAME and some sample applications are presented in this paper.