A multiscale modeling approach for fast prediction of part distortion in selective laser melting

Selective laser melting (SLM) is a powder bed based additive manufacturing process. It is widely used to make functional parts in a layer upon layer fashion. The severe temperature gradients produce large tensile residual stress which leads to part distortion and negatively affect product performance. Due to the complex coupling multi-scale mechanisms, it is a great challenge to predict part distortion since traditional modeling approaches demand an exceedingly long computational time. This study has developed a practical multi-scale modeling methodology for fast prediction of part distortion by integrating a micro-scale laser scan model, a meso-scale layer hatch model, and a macro-scale part model. A concept of equivalent heat source has been developed for the micro-scale laser scan model. Local residual stress field was predicted in the meso-scale layer hatch model using the equivalent heat source. The residual stress field was then imported to the macro-scale model to predict part distortion and residual stress. The predicted part distortions were validated with the experimental data with four different scanning strategies. (C) 2015 Elsevier B.V. All rights reserved.