Inverse Identification Research For Dynamic Constitutive Parameters Of Thread Material Based On MSARS Algorithm

CST energy absorbing device is a kind of passive safety device which uses the principle of metal cutting to cut the thread,while collision happens. it can obtain smooth and continuous mechanical property and absorb the collision energy through cutting the running thread,simultaneously,reduce the damage to cars and guarantee the safety of passengers. The researches on this device is currently focus on finite element modeling, after analyzing the date obtained by the numerical simulation test, and then make some improvements according to the specific requirements for further research. Therefore, choosing a right constitutive model and its parameters to build a credible finite element model is the key to the reliability of the date from numerical simulation test and the feasibility of the corresponding improvements. The thesis aimed at providing a correct and reliable methodology to inverse identify the material parameters of CST energy absorbing device,and based on the method of model comparison , a systematic inverse identification research for Johnson-Cook constitutive parameters of thread material was conducted. The thesis includes following:1. According to the implicit problems in inverse identification of material parameters,after analyzing the advantage and disadvantage of random search method and surrogate model for optimization method,a inverse methodology for global optimal solution which optimized surrogate model to accelerate random search was proposed,referred to as the MSARS algorithm. For the efficiency improvement of parameters inverses identification,some aided program was compiled by Fortran 90 programming language,and combined with Hyperstudy software, the MSARS was able to be acted by programs. Finally, the superiority of this algorithm was verified by an example.2. Mat_Plastic_Kinematic model which was generally thought to describe the mild steel's dynamic mechanical properties correctly was used to act as a validation test model in this thesis. Based on the method of model comparison, the MSARS algorithm was adopted to validate and optimize the Johnson-Cook model parameters, and then an optimal Johnson-Cook model parameter which was able to describe thread's dynamic mechanical properties correctly was finally attained.3. In order to investigate the robustness of the thread material parameters inverse identified,the impact of uncertain factors(the error of collision angle,collision mass, collision speed) and certain factors(different shapes of section) on parameters inverse identification was systematically researched.