Approaches to analysing scatter in forming simulations: from fundamental to pragmatic

Abstract

Including scatter in material properties has received attention in the last decades but, it is not fully mature yet. Industry has cooperated with academia to progress the topic, but at the same time devised simpler ways to start analysis anyway. Robustness analysis is a pre-requisite for process optimisation. Tata Steel has been in material modelling for Sheet Metal Forming for many years and has developed its own yield locus model to enable better customer support. The accuracy of this Vegter model comes at a price, however: more parameters to be measured. Wiebenga devised a robustness analysis but it requires more input than is usually available. At the same time leading software vendors introduced options to evaluate statistical behaviour. Obviously, process optimisation is only useful when accurate models can be employed, so in this paper we will focus on how the scatter in material properties translates in scatter in yield locus for the BBC2005 model vs. Vegter 2017. Automated statistical analysis was not available for Vegter 2017. So, finally, to enable estimation of scatter in forming processes in the short term with Vegter individual simulations were done on a selected number of input datasets. This is shown to give markedly different results to BBC2005.