A Convolutional Neural Network for Prediction of Laser Power Using Melt-Pool Images in Laser Powder Bed Fusion
Ohyung Kwon, Hyung Giun Kim, Wonrae Kim, Gun-Hee Kim, Kangil Kim
Abstract
In laser powder bed fusion, a convolutional neural network could build a good regression model to predict a laser power value from a melt-pool image. To empirically validate it, we used the acquired image data from a monitoring system inside metal additive manufacturing equipment and optimally configured a convolutional network by the grid search of hyper-parameters. The proposed network showed only 0.12 % of test images were out of the criterion for judging the predicted laser power value to be reliable and showed more accurate results than deep feed-forward neural network in the prediction of laser power states unseen in training steps. We expect that the proposed model could be utilized to discover the problematic position in additive-manufactured layers causing defects during a process.