Towards Geometry Prediction in Additive Manufacturing by Considering Variation in the Inherent Strain

Paper i proceeding, 2022

The inherent strain method is commonly employed to predict part distortion in the additive manufacturing (AM) process simulations. However, mean inherent strain values are considered which could hinder the prediction accuracy. Therefore in this paper, variation in inherent strain values are considered to predict the geometric distortion. In the first step, a five layer mesoscale thermo-mechanical model is employed to estimate the varying inherent strain values in each of the five layers. This serves as an input to the inherent strain method to predict geometric distortion at the part level. A comparison between mean versus varying inherent strain approach is shown to highlight the differences in geometric distortion prediction accuracy.