Electrochemical machining parameter optimization and prediction of performance using artificial neural network

Electrochemical micromachining is promising technique used to machine metal matrix composites and hard to cut materials. In this research mixed L Orthogonal array experiments are used to along with 30 vol% of ethylene glycol mixed sodium nitrate electrolyte. The tool electrode is coated with ceramic coating whose diameter is 360 µm and Al7075 + 10 vol%BC metal matrix composites of thickness 500 µm is used as a workpiece. The experiments are optimized using grey relational analysis and most significant factor is found using Analysis of variance (ANOVA). As per the Grey relational Analysis (GRA) the optimal combination is 30 vol% of ethylene glycol,voltage of 9 V,duty cycle of 70% and electrolyte concentration of 35 g/L. As per the ANOVA, the most promising factor is electrolyte concentration which shows 46.36%. The Artificial Neural Network (ANN) model prediction value is 0.1675 and 1.1400 which is very close to GRA optimized values of machining rate and surface corrosion factors, ie 0.1667 and 1.1395 respectively.