Additive Manufacturing

Data-Driven Insights into Melt Pool Dynamics and Energy Absorption in Laser Manufacturing  

Abstract: 

Energy absorptivity is crucial in Additive Manufacturing (AM) because it directly influences how effectively the material absorbs laser energy, impacting the melting, fusion, and solidification of metal powders. This is why predicting the absorptivity of materials during AM processes has been a topic of interest for researchers. This study investigated the correlation between melt pool dynamics and energy absorptivity in laser-based manufacturing processes. The study explored a method of detecting the melt pool using the Attention U-Net-based Machine learning (ML) model. Furthermore, a correlation between real-time energy absorptivity and melt pool and keyhole features like depth and width has been established. The data used for this study is taken from a publicly available NIST Data Repository of  2022 Asynchronous AM-Bench challenge. The primary goal of this study was to understand how melt pool characteristics affect energy absorption for both spot and scan lasers, offering insights into how to optimize energy absorptivity through better control of the melt pool dynamics. This  is  done  by  predicting  the  melt  pool  and  keyhole  features  like  depth  and  height  and  predicting  time  dependent  absorption  using  these  features. The results demonstrate significant potential for using ML models to improve energy absorption by controlling melt pool features.