Layer-by-layer model-based adaptive control for wire arc additive manufacturing of thin-wall structures

Abstract Improving the geometric accuracy of deposited component is essential for wider adoption wire arc additive manufacturing (WAAM) in industries. This paper introduces an online layer-by-layer controller that operates robustly under various welding conditions to improve deposition WAAM process. Two control strategies are proposed and evaluated this work: A PID algorithm a multi-input multi-output model-predictive (MPC) algorithm. After each layer deposition, geometry measured using laser scanner. These measurements compared against CAD model, errors then compensated by controller, which generates new set parameters next layer. The MPC algorithm, combined with linear autoregressive (ARX) modelling process, updates between successive layers minimizing cost function based on sequences input variables predicted responses. Weighting coefficients ARX model trained iteratively throughout performance designed architecture investigated through both simulation experiments. Results show real-time improved increasing complexity implemented algorithm: controlled fluctuations test were reduced 200% whilst maintaining within 3 mm limit conditions. In addition, adaptiveness strategy verified accurately controlling fabrication part complex geometry.