INTELLIGENT HYBRID ANFIS-GRG METHOD FOR MACHINABILITY ANALYSIS OF METAL MATRIX COMPOSITES IN WIRE ELECTRICAL DISCHARGE MACHINING

Abstract

The present investigation entails the integrated optimization and modelling of the Wire Electro-Discharge Machining (WEDM) process for the stir-cast AA7050 alloy reinforced with graphite particles, forming metal matrix composites. Because of the composite's high strength and heterogeneous nature with a very fine microstructure, machining parameters must be carefully controlled for an improved process capability. A Taguchi-based design is practically followed so that important process variables-pulse-on time, pulse-off time, and peak current are effectively evaluated for their effects on important performance measures such as material removal rate (MRR), surface roughness (SR), and Dimensional Deviation (DD), and form/orientation tolerances. Grey Relational Analysis (GRA) techniques optimize the multi-objectives to analyse the best possible settings that will maximize overall performance. For modelling and prediction of the process behaviour, an Adaptive Neuro-Fuzzy Inference System (ANFIS) was developed with the Grey Relational Grade (GRG) as the output response. The ANFIS model is developed to capture the nonlinear interactions between the input variables and has a very high prediction accuracy, proving its usefulness for modelling the WEDM process. The synergistic application of Taguchi, Grey, and ANFIS modelling is a robust structure for improving the machinability of AA7050-graphite composites and aiding intelligent decision-making in advanced manufacturing applications.