Estimation of Machining Performances using GMDH and ANN in Wire EDM of Cu-1Cr-0.1Zr Alloy

Abstract

Wire cut Electrical Discharge Machining (WEDM) is a special form of EDM process in which electrode is a continuously moving conductive wire. The material removal is by controlled erosion through a series of repetitive sparks between workpiece and wire electrode. WEDM is a specialized thermo electrical machining process capable of accurately machining parts with varying hardness or complex shapes. Present study outlines the estimation of machining performances in the wire electric discharge machining of Cu-1Cr-0.1Zr alloy using Group Method of Data Handling (GMDH) technique and Artificial Neural Network (ANN). Cu-1Cr-0.1Zr alloy was machined using different process parameters based on Taguchi’s L27 standard orthogonal array. Parameters such as pulse-on time, pulse-off time and current were varied. The response variables measured for the analysis are Dimensional Error (DE), Surface Roughness (SR), Volumetric Material Removal Rate (VMRR) and Electrode Wear (EW). Machining performances have been compared using sophisticated mathematical models viz., GMDH and ANN. The GMDH algorithm is designed to learn the process by training the algorithm with the experimental data. Different GMDH models can be obtained by varying the percentage of data in the training set and the best model can be selected from these, viz., 50%, 62.5% & 75%. Three different criterion functions, viz., Root Mean Square (Regularity or RMS) criterion, Unbiased criterion and Combined criterion were considered for estimation. Machining performances is predicted for 70% of data in training set using ANN. Estimation and comparison of machining performances were carried out using GMDH and techniques.