| The laser powder-directed energy deposition process is non-linear, and the dynamic response of the system changes with each layer. Current systems require considerable manual tuning and process knowledge to achieve quality production. To take advantage of the technology, it is necessary to develop in-situ process monitoring and closed-loop control. In this research a single-input, single-output fuzzy self-tuning PI controller was developed to ensure a constant melt pool diameter during the build. The laser power was selected as the control output and the diameter as the control input variable. A thin wall part was built with ten layers while varying the laser power using a pseudo random binary sequence. The controller was designed, and its performance simulated using MATLAB/SIMULINK® and compared to a simple PI controller. The results suggest that the fuzzy self-tuning proportional and integral controller will result in a more consistent melt pool diameter during the build process. Key Words: additive manufacturing, directed energy deposition, LENS®, fuzzy logic, PI control, self-tuning, system identification, particle swarm, optimization. |