Empirical Modelling of Mrr in Electrochemical- Mechanical Finishing of Bevel Gears

This paper presents the micro-finishing of bevel gears by electrochemical-mechanical finishing (ECMF) process under different experimental conditions of current, rotating speed of workpiece and electrolyte concentration. These different experimental conditions are planned on the basis of box behnken design (BBD) design of experiments (DOE) technique of response surface methodology (RSM). An empirical model has been developed using artificial neural network (ANN) with feed-forward backpropagation algorithm. The developed ANN model has been optimized and it is observed that the developed model is quite capable of predicting the material removal rate of the process for different sets of input process parameters.