Mesh regularization through predictive scaling factors based on material model parameter estimation for debris impact test on titanium alloy (TI-6Al-4V) plate.

Abstract

The mesh Size is a crucial element of the FEA simulation. This study focuses on mesh regularization, which is the process of removing mesh dependency from FEA simulations. These techniques are commonly used to match experimental data and reduce computational time. In this study, mesh regularization was approached using a material failure criterion based on the element size. However, the current approach to failure scaling based on the element size is a trial-and-error process. To address this, four different analytical models were formulated to predict the scaling factors based on the failure criterion vs. mesh size. The material model-based scaling factors were determined for the tensile tests using 11 different mesh sizes. These predictive scaling factors were then compared with the scaling factor obtained using the conventional tensile testing method. The findings of this study show that using the scaling factors found through the experimental method eliminates errors in the tensile test model, and the predicted scaling factor significantly reduces the error in force-displacement for different mesh sizes compared to the base mesh size. Additionally, in the debris impact test, the predictive scaling factors reduced the deviation in residual velocity to 1.67 % from the initial 17.81 %, and the maximum deviation in energy absorption was reduced from 48.40 % to 8.10 % without scaling factors. This study introduces a new method for predicting material-model-based scaling factors for mesh regularization and discusses the applicability of this method to various problems.