Robust Design Optimization in forming process simulation

Today, FE-simulation of forming process has become an integral part for assessing and evaluating forming processes. The optimization, i.e. improvement of product characteristics, has been an integral part of forming simulation based virtual product development for several years now. On the other hand, the robustness of forming processes is becoming more and more focused on recently. Therefore the introduction of numerical robustness evaluation and the development of suitable measurements of robustness are integrated into the virtual product development right now. In fact, robustness is an additional demand on optimized forming processes. Therefore; a process is necessary of optimizing and at the same time securing the robustness. That process is called Robust Design Optimization (RDO). The optimization and robustness evaluation are either performed consecutively or simultaneously, and several methods are available for this. In the following, existing methods shall shortly be introduced and discussed from a practical point of view regarding their appliance. From our experience of introducing optimization and robustness evaluation methodology in virtual product development processes, it is recommended to start with a consecutive approach of using sensitivity analysis, robustness evaluation and deterministic optimization to solve the problem or to achieve the knowledge which is necessary to define a Robust Design Optimization task. This procedure will be demonstrated at a practical application. Of course, the final dream of virtual product development is an automatic Robust Design Optimization procedure with dealing simultaneously with optimization and reliability domain. Therefore, methods for a simultaneous performance of the optimization and the robustness evaluation will be introduced and their potentials will be discussed. In the paper the approach is discussed at a practical application. For the forming simulation, LSDYNA is used. For the sensitivity study, the robustness evaluation and the optimization optiSLang, a general purpose parametric optimization and reliability software package [1] is used. Especially for forming simulation, Dynardo developed Statistics_on_Structure (SoS), a statistical post processor. The statistical measures on the FE structures serve as discussion basis for the identification of critical areas and as a basis for evaluating the robustness. In addition, this type of representation leads to a high acceptance of the results in the production departments. The practical application of robust design optimization in forming simulation shows a high degree of nonlinearity in the optimization domain. To ensure robustness it was not possible to identify a constant safety distance which means a deterministic design with a maximum FLD_crack value of 0.7 …