Adaptive neuro-fuzzy inference system based modeling of recast layer thickness during laser trepanning of Inconel-718 sheet

Abstract

Re-solidification of molten material during the laser trepanning of Inconel-718 is a major hindrance in achieving good quality drill with high precision and accuracy. Re-solidification affects the performance of the drilled hole. Many researchers have tried for the optimization of laser trepan drilling in order to improve the drilled hole quality characteristics. But till now, limited work has been reported in concern with recast layer formation in laser trepan drilling of Inconel-718. This paper experimentally investigated the recast layer formation during laser trepan drilling followed by the prediction of the recast layer formation using the adaptive neuro-fuzzy inference system (ANFIS). Experiments are performed on 1.4-mm-thick Inconel-718 sheet using pulsed Nd: YAG laser. Recast layer thickness has been measured for each experiment followed by the ANFIS-based prediction of recast layer. Moreover, the effect of different input parameters on the recast layer has also been discussed.