Crash structure development is a key element in the design of vehicle bodies. The focus here is on occupant protection and the protection of critical components such as the fuel supply. Crash simulation based structural optimization can be a tool to support the engineers involved. The problem with optimizing crash problems is the long computing times of crash simulation. Hundreds of simulations may be necessary in the course of an optimization, which leads to enormous resource requirements and thus to high costs. By reducing the computational model, the resource requirements can be limited. In this dissertation different approaches to model reduction are realized. Components or elements are removed from the model and replaced by suitable boundary conditions. For example, a substitute mass is attached to the remaining model to compensate for inertia and missing mass. To ensure that the reduced model corresponds to the original model in its characteristics, a validation is carried out. Reduced models can be used e.g. in case studies or optimizations. Due to the reduced resource requirements, more simulations can be carried out in the same time, which leads to better results. Examples are used to apply the developed methodology in practice. Two vehicle optimizations serve as benchmarks.

Complete dissertation as PDF