Impact of change in machining time, MRR in WEDM modeled by ANN-RSM

Abstract

A combination of nickel and chromium alloy named as Inconel 800 and zinc coated brass wire is used as a workpiece and wire electrode respectively for this experimental analysis designed by Taguchi orthogonal array L18. The experimental analysis is carried out under the highest, medium and lowest rate of dielectric fluid flushing through both the upper and lower nozzle placed nearer to the metal cutting edge. The flushing through the nozzles is in litters per minutes and the process parameters such as pulse on time, pulse off time, spark gap voltage, peak current were varies during the experimental phase and wire tension, wire feed, water pressure, peak voltage, servo feed were kept constant. For checking the correctness of the input parameters a digital storage oscilloscope (Agilent 3000) is used to find out the pulse on and pulse off time signals and the signals were validated with respect to their actual units by analyzing the time signals. The status of the influence and the search for best possible responses of the machining parameters for minimum machining time and increase in MRR, is determined by using analysis of variance (ANOVA). Finally an analytical model has been designed with the help of artificial neural network (ANN) and Response surface methodology (RSM)