A macroscale thermal simulation strategy with track-scale resolution for laser powder bed fusion

Abstract

Numerical modelling is an effective tool to investigate the rapid temperature cycling in laser powder bed fusion. However, the high computational cost limits the modelling of part-scale components due to the significantly different scales between the highly local energy input region and the global field response region. This study presents a macroscale finite element thermal simulation strategy with a track-scale resolution by combining the recently developed characteristic time-based heat input (CTI) model with a currently proposed adaptive mesh re-mapping method. Due to the uniformly distributed power density along the scan direction in the CTI model, the optimized mesh strategy allows one element to capture the temperature solutions over the majority of a scan track except at the two ends. The adaptive mesh re-mapping method is used to map simulated results from a previous scan track model to a current scan track model with a new mesh configuration. This strategy has been implemented for a Ti-6Al-4V part with the dimensions of 20×40×3mm3. The predicted temperature histories agree well with the in-situ measurements at the points in both the printed part and powder bed. Three case studies demonstrate that our strategy can improve computational efficiency between 6 and 30 times.

Keywords

• Laser powder bed fusion
• FE thermal modelling
• Characteristic time-based heat input model
• Adaptive mesh re-mapping
• Computational efficiency