Iterative Learning Control for Melt Pool Depth Consistency in Directed Energy Deposition

Porosity is a critical defect affecting the quality of parts produced by the Directed Energy Deposition (DED) process, with its formation significantly influenced by the melt pool depth (MPD). Despite extensive research on controlling melt pool dimensions and temperature, advancements in MPD control are stymied by the inability to directly observe it during the DED process. Here, we present an innovative Iterative Learning Control (ILC) framework for enhancing MPD consistency in DED. Leveraging a simulation-based process design, our research showcases the capability of ILC in achieving precise control over MPD, reducing MPD variance between layers. Through comparative analyses, the proposed approach is evidenced to outperform traditional MPD control strategies.