NSGA-II in DNN for Interpretable AI Predictions of Melt Pool Geometry in Digital Twin Applications within IoT-Enabled Additive Manufacturing

Additive manufacturing (AM), such as Laser Powder Bed Fusion (LPBF), enables intricate fabrication of custom components via layer-by-layer deposition. This study focuses on improving precision in additive manufacturing (AM), specifically Laser Powder Bed Fusion (LPBF), by using regression machine learning, including deep neural networks (DNNs), to predict and optimize melt pool geometry (depth, width, length). Leveraging process parameters and material properties, the goal is to refine the geometry to meet design and performance criteria. Integrating digital twin and IoT technologies enables real-time monitoring and dynamic control. The use of NSGA-II optimization further enhances melt pool behavior analysis. This research aims to advance AM quality, reliability, and efficiency in digital manufacturing.