3-Dimensional heat transfer modeling for laser powder bed fusion additive manufacturing using parallel computing and adaptive mesh

In this article, an innovative 3-dimensional (3D) heat-transfer finite element parallel-computing model with adaptive mesh capability, named LPBFSim, for Laser Powder Bed Fusion (LPBF) additive manufacturing is introduced for high precision prediction of melt pool dimensions and single-layer part-level process simulations. Numerical modeling can significantly reduce the expense of the sole deployment of trial-and-error experiments for achieving optimal process parameters. The previously developed single-track model [1] is highly accurate to predict melt pool dimensions for different combinations of process parameters. However, without using parallel computing and an adaptive mesh, it is very challenging to scale it up to a multi-track model or even a part-level model due to the high computational cost. The proposed model is parallelized to be able to run on a cluster and aimed to solve this diffi culty while keeping all the accuracy from the previous version. Single-track, multi-track, and single-layer part-level models have been implemented to demonstrate its efficiency and accuracy.

Automated summary

This article introduces an innovative 3D heat-transfer finite element parallel-computing model with adaptive mesh capability called LPBFSim, which is used for high precision prediction and process simulation in Laser Powder Bed Fusion (LPBF) additive manufacturing. The model is parallelized to overcome the high computational cost of scaling up from single-track to multi-track or part-level models, while maintaining accuracy. Single-track, multi-track, and single-layer part-level models have been implemented to demonstrate the efficiency and accuracy of the proposed model.