Analysis and Prediction of Process Parameters During Laser Deposition Manufacturing Based on Melt Pool Monitoring

During laser deposition manufacturing (LDM) process, melt pool width which is greatly influenced by process parameters is essential for the forming tracks geometry. In this paper, the melt pool geometry evolution was monitored by a CCD camera, and a method of applying Kalman filtering for the melt pool width detection during LDM process was presented to obtain accurate values. Orthogonal experimental design and multiple regression analysis were used to establish an empirical model describing the correlation between the melt pool width and three main process parameters (laser power, scanning speed, and powder feeding rate). And the developed model was verified experimentally. Finally, particle swarm optimization (PSO) was implemented for prediction of process parameters during the buildup of a thin wall. The results indicate that process parameters analysis and prediction for LDM process could make it possible to acquire an efficient process for the forming tracks geometry control.