A study on experimental benchmarks and simulation results in sheet metal forming

Abstract

The introduction of new materials brought new challenges to sheet metal forming processes. The behaviour observed with conventional steels may not be applied when using high-strength steels or aluminium alloys. Therefore, numerical codes need to model correctly this different behaviour and different constitutive equations exist to model with greater accuracy the material behaviour. As a consequence, numerical validation demands higher experimental reliability Such reliability is nowadays one of the main points in experiments, since numerical modelling has arrived to a higher quality level of results. Therefore, new numerical developments need a higher level of experimental accuracy in order that validation with experiments may take place. A way to achieve such reliability is to select a benchmark model, define a standardization procedure to perform experiments, define a correct method for measurement of selected results and perform the benchmark in several different institutions. This method has been used for several times, e.g., in NUMISHEET conferences. However the corresponding experimental accuracy of results has been behind expectations, especially with increasing similarity of numerical results obtained by different codes. This paper presents proposed benchmarks to be used in sheet metal forming, which has been object of an international project. Each benchmark intends to model fundamental behaviour of sheet metal parts. Several different partners have performed experiments with a defined and detailed standardization procedure and results have been compared. Also, different materials, steels and aluminium alloys, have been object of these experiments. Some results of these benchmarks are presented and subsequent selected studies are shown, which were performed to understand and explain some scatter in results. Finally, it is presented a numerical simulation of selected benchmarks and corresponding results are compared with experiments.