Sensor-Based In-situ Process Control Of Robotic Wire Arc Additive Manufacturing Integrated With Reinforcement Learning

Wire and Arc Additive Manufacturing (WAAM) is a manufacturing technique capable of fabricating large-scale metallic components in a layer-by-layer fashion. As an emerging technology, there still exist numerous challenges that need to be overcome to ensure the geometrical accuracy of the part produced. With an increasing number of deposited layers, geometrical errors often accumulate in height and the accumulated heat becomes significant, leading to the slumping of the beads. The quality of the part can be enhanced through in-situ real-time feedback control. However, as the WAAM process is a time-variant process that is highly non-linear and multi-dimensional, it is difficult to model the process relating the process parameters to the final quality of the produced part. To address this challenge, a sensor-based in-situ process control framework integrated with reinforcement learning (RL) artificial intelligence (AI) is proposed to iteratively learn the impacts of various process parameters to finally control the output geometry of a single-bead multi-layer part. The proposed control frameworks are then implemented and simulated on a robotic large-scale WAAM system.