Multivariable predictive control of laser-aided powder deposition processes

This paper derives and illustrates a multivariable predictive control scheme for laser-aided powder deposition (LAPD) processes utilizing a mobile co-axial laser and powder nozzle. First, a control-oriented multi-input-multi-output (MIMO) process model is adopted that captures the coupled nonlinear dynamics of deposited layer height and average melting pool temperature, with laser power and scanning speed as process inputs. Then, a nonlinear model predictive control (NMPC) scheme is devised in a transformed spatial coordinate system for simultaneous online control of deposition height and melting pool temperature. The conditions that ensure the closed-loop stability of the proposed control scheme are also provided. The effectiveness of the approach as well as the clear trade-offs in the multivariable control is illustrated via a case study on a single layer deposition process.