Influence of Submodel Size and Evaluated Functions on the Optimization Process of Crashworthiness Structures
Optimization of crashworthy structures is an influential aspect during the development of a vehicle body. This paper deals with the structural optimization of large crashworthy systems with modifications to its shape and topology using a submodel technique. A submodel is a region of interest cut out from the large system which is to be analyzed in detail. The main advantage of the submodel technique is the smaller size of the submodel, which leads to a reduction of computation time, the usage of less disk space and the possibility to use finer mesh for higher resolution of critical areas. The development of large systems can be carried out using optimization in different levels. For example, level 1 is the large system and level 2 is a submodel of the large system. Both levels are coupled together with help of interface functions to form a multilevel optimization process. The influence of the size of submodel and the update of the submodel boundary conditions during the optimization process have been studied. The study is done with two different application examples. The first example is a cantilever beam impacted by a solid sphere. Optimization functions are the intrusion of the sphere in the structure, the acceleration of the sphere and the maximum contact force. The design variables are the position, the orientation and the size of beads. The position of the submodel including a bead is optimized in the level 1 and the orientation and size are optimized in level 2. The second example is a rocker beam in a pole impact load case. Here, the optimization functions are the intrusion of the pole and the mass. The optimization process is done with the “Graph and Heuristic based Topology optimization approach (GHT)” [1]. The design variables manage the topology, the shape and the sizes of the single components of the rocker. For both examples, different possible optimization functions are used and discussed.
https://www.dynamore.de/de/download/papers/2016-ls-dyna-forum/Papers%202016/montag-10.10.16/crash/influence-of-submodel-size-and-evaluated-functions-on-the-optimization-process-of-crashworthiness-structures/view
https://www.dynamore.de/@@site-logo/DYNAmore_Logo_Ansys.svg
Influence of Submodel Size and Evaluated Functions on the Optimization Process of Crashworthiness Structures
Optimization of crashworthy structures is an influential aspect during the development of a vehicle body. This paper deals with the structural optimization of large crashworthy systems with modifications to its shape and topology using a submodel technique. A submodel is a region of interest cut out from the large system which is to be analyzed in detail. The main advantage of the submodel technique is the smaller size of the submodel, which leads to a reduction of computation time, the usage of less disk space and the possibility to use finer mesh for higher resolution of critical areas. The development of large systems can be carried out using optimization in different levels. For example, level 1 is the large system and level 2 is a submodel of the large system. Both levels are coupled together with help of interface functions to form a multilevel optimization process. The influence of the size of submodel and the update of the submodel boundary conditions during the optimization process have been studied. The study is done with two different application examples. The first example is a cantilever beam impacted by a solid sphere. Optimization functions are the intrusion of the sphere in the structure, the acceleration of the sphere and the maximum contact force. The design variables are the position, the orientation and the size of beads. The position of the submodel including a bead is optimized in the level 1 and the orientation and size are optimized in level 2. The second example is a rocker beam in a pole impact load case. Here, the optimization functions are the intrusion of the pole and the mass. The optimization process is done with the “Graph and Heuristic based Topology optimization approach (GHT)” [1]. The design variables manage the topology, the shape and the sizes of the single components of the rocker. For both examples, different possible optimization functions are used and discussed.