Improved Prediction of Hybrid-III Injury Values Using Advanced Multibody Techniques and Objective Rating

Virtual testing has become an essential part of the vehicle design process. Increasingly stringent safety norms call for accurate prediction of injury values. Numerical techniques such as DOE, optimisation and stochastics yield increasing numbers of simulations and this requires CPU efficient and robust solutions. In order to face these demands a new Hybrid-III 50^th %ile crash test dummy model has been developed in MADYMO.

Advanced multibody techniques have been used to obtain fast computation times with the geometry and potential accuracy of CPU intensive finite element models. So-called facet surfaces have been used in combination with flexible bodies and rigid bodies. The MADYMO contact algorithm has been enhanced with options to separately describe the non-linear compliance of two contacting objects such as a dummy and a seat. The principles and pitfalls of the facet methodology in relation to typical automotive applications is described, resulting in modelling guidelines and recommendations.

The available set of component and full dummy validations has been extended with load cases representing the latest restraint system designs and test procedures. A systematic validation has been performed using objective rating techniques. Rating is used in order to show the predictivity increase of the new facet model. Special attention has been paid to different rating criteria.