Forecasting the Material Removal Rate of Inconel 718 Alloy in Electrochemical Machining through Machine Learning Approaches

The Material Removal Rate (MRR) is a vital aspect of Electro-Chemical Machining (ECM), an engineering manufacturing method that depends on electrochemical reactions. The MRR is dependent on factors such as current, voltage, electrolyte concentration, and machining time. To investigate the effect of MRR on Inconel 718 super-alloy, experiments were conducted using stainless steel tool under different independent machining conditions. Machine Learning (ML) approaches could be utilized to predict machining outcomes based on specific input parameters. In this research, ML techniques were applied to ECM by developing models using multiple linear regression, Random Forest, K-Nearest Neighbors (KNN), and Xtreme gradient boosting algorithms. These models aimed to establish the association among the collaborative impacts of the electrolytic solution, volts, amps, and feed rate on MRR. Additionally, the study seeks to recognize the best ML technique for forecasting the MRR of Inconel 718 alloy during ECM utilizing a regression approach. The outcomes indicated that the Xtreme gradient boosting algorithm achieved the highest forecasting performance, with an accuracy of 99.42%. This was followed by the KNN model in terms of predictive accuracy.