Tribological analysis of laser deposited SS316L/Co27Cr6Mo functionally graded materials using adaptive neuro-fuzzy inference system

Abstract

Image processing, power engineering, robotics, industrial automation etc., have all found successful uses for artificial intelligence (AI) techniques such as artificial neural networks (ANN) and neuro-fuzzy logic (FL). In this study, an adaptive neuro-fuzzy inference system (ANFIS) modelling of machine learning (ML) has been implemented to estimate the tribological properties of functionally graded materials (FGM). These FGMs were developed using a direct energy deposition (DED) technique of additive manufacturing (AM) from SS316L and Co27Cr6Mo alloys. The input data for this ANFIS modelling is acquired from the experiments done on FGM samples using the Pin on Disc (PoD) apparatus. The main objective of this work is to predict the tribological parameters of FGM samples by creating a data-driven predictive model called ANFIS. From the findings, the ANFIS was found to be the efficient method to estimate the wear rate of FGM samples.