Predictive Modelling of Residual Stresses for Single Bead P420 Laser Cladding onto an AISI 1018 Substrate

摘要

Research correlating the process parameters, bead geometry, and induced residual stresses is essential for addressing geometric and functional quality issues for laser cladding based surface treatment and additive manufacturing solutions. However, analysis of variance methods cannot be readily applied, as the residual stresses values vary based on their position within the clad bead or substrate. Experimental data is used to calibrate single bead finite element models, which in turn, are used to seed an artificial neural network model and an adaptive neuro fuzzy interference system (ANFIS) model. Preliminary results indicate that the ANFIS approach has potential.