F2008-12-041
Robust Design Method for Restraint System in Multi Side Impact Test Configurations Using Multi-Objective Optimization
This paper describes the development of robust design method for restraint system considering unevenness in multi side impact test modes. Side impact vehicle safety standard, FMVSS214, has been upgraded in 2007 and will be effective from 2009. Pole impact tests have been added and the SID dummy used for measuring injury values in tests has been replaced with ES2re and SID2s dummies. As a result, the test modes of FMVSS214 increase from 1 to 3, and the total number of side impact test modes increase from 3 to 5 including the two rating tests which are IIHS side impact test and New Car Assessment Program (NCAP) test. Therefore, it becomes a difficult issue to develop a restraint system to meet the injury values requirement in all test modes. On the other hand, it is well-known that injury values have large unevenness due to variations in parts and test conditions. Many optimization methods for development of restraint system considering the unevenness have been reported. However, there have been no reports on multi objective optimization method considering unevenness for multi side impact modes. The reason for this is thought to be the need of a detailed FE model for side impact modes, because parts in the cabin come in direct contact with the dummy and they are seriously deformed in side impact test. In addition, since the multi objective optimization considering unevenness requires the large number of calculations, the usage of FE model needs huge computational efforts and time, and it is not suitable for the vehicle development process. The authors have developed a method for multi objective optimization considering the unevenness, based on response surface methodology, using a detailed side impact occupant injury FE model, which requires less computational effort and time. The developed method has been applied to the FMVSS214 pole side impact test and IIHS side impact test. The computational efforts and period are reduced one tenth compared current method and it has been revealed from the results that the feasible design area that can meet the injury values requirement of all test modes, considering unevenness is very narrow. Therefore, it is difficult to derive the optimal design using currently existing optimization methods in each mode. Results of crash test, which incorporated the optimal design obtained from the developed method show that the developed method is effective for designing the restraint system in multi side impact test modes.
Poster presentation: Simulation and testing