Prediction and analysis of radial overcut in holes drilled by electrochemical machining process

Radial overcut predictive models using multiple regression analysis, artificial neural network and co-active neuro-fuzzy inference system are developed to predict the radial overcut during electrochemical drilling with vacuum extraction of electrolyte. Four process parameters, electrolyte concentration, voltage, initial machining gap and tool feed rate, are selected to develop the models. The comparison between the results of the presented models shows that the artificial neural network and co-active neuro-fuzzy inference system models can predict the radial overcut with an average relative error of nearly 5%. Main effect and interaction plots are generated to study the effects of process parameters on the radial overcut. The analysis shows that the voltage, electrolyte concentration and tool feed rate have significant effect on radial overcut, respectively, while initial machining gap has a little effect. It is also found that the increase of the voltage and electrolyte concentration increases the radial overcut and the increase of the tool feed rate decreases the radial overcut.