Pulse shape analysis and data reduction of real-life frontal crashes with modern passenger cars

Abstract

The increased use of computer simulations such as finite element modelling for evaluating passive safety applications has made it possible to simplify and parameterise complex physical processes. Crash pulses derived from laboratory tests have been used in many studies to evaluate and optimise passive safety systems such as airbags and seat belts. However, a laboratory crash pulse will only be representative of the acceleration time history of a specific car crashing into a barrier at a specified velocity. To be able to optimise passive safety systems for the wide variety of scenarios experienced during real-life crashes, there is a need to study and characterise this variation. In this study, crash pulses from real-life crashes as recorded by event data recorders were parameterised, and the influence of vehicle and crash variables was analysed. The pulse parameterisation was carried out using eigenvalue analysis and the influence that vehicle and crash variables had on the pulse shape was determined with multiple linear regression. It was shown that the change in velocity, the subject vehicle mass, and the properties of the collision partner were the variables that had the greatest effect on the shape of the crash pulse. The results of this study can be used to create artificial real-life pulses with different crash parameters. This in turn can be used for stochastic computer simulation studies with the intention of optimising passive safety systems that are robust to the wide variation in real-life crashes.