Improving Texture Prediction by Increasing Mesh Resolution in Submodel: A Crystal Plasticity FE Study and Experiment Verification

Abstract

Crystal plasticity finite element simulations require tremendous computation time and, accordingly, coarse mesh is generally used. To improve the texture prediction, Submodelling was applied to feature grains in this study. A simulation of the Wholemodel (whole sample) was firstly carried out to obtain the global texture, and then a smaller region from the Wholemodel was selected, reconstructed and finely meshed in the Submodel. The movement on the selected region boundary, obtained from the Wholemodel, was used to deform the Submodel. The Submodel reproduced the predictions in the Wholemodel, and the texture prediction, especially at micro-scale, was greatly enhanced in the Submodel due to the fine mesh. This significant drop in the Submodel computation time marks an ~85% decrease compared to the Wholemodel.