MODELING OF WEDM PROCESS FOR COMPLEX SHAPE USING MULTILAYER PERCEPTRON AND RESPONSE SURFACE METHODOLOGY

V. L. Bhambere, S. S. Khandare

Abstract: Wire cut electric discharge machining is the present days requirement in manufacturing the intricate and complex shape parts as required in the modern industrial products. The process being complex in nature for control over the machining process parameters. In the present study Multilayer perceptron (MLP) model is developed to predict the Material removal rate. The mathematical regression model is also developed by using Response Surface Methodology (RSM) to establish the relationshIp between the MRR and input process parameters like pulse on time (Ton), Pulse off time (Toff), Peak current (IP) and Servo voltage (SV). The predicted value by using MLP and RSM were compared with the experimental values. The average percentage errors are found to be 1.29 and -0.36527 for MLP and RSM respectively. It is observed that the predicted values with RSM are closer to experimental values as compared to MLP model

Keywords: WEDM, Response Surface Methodology, Neural Network, Multi Layer Perception, Material Removal Rate, Surface Roughness

DOI: https://doi.org/10.15623/ijret.2015.0409024