Random Forest (RF) Assisted and Support Vector Machine (SVM) Algorithms for Performance Evaluation of EDM Interpretation

Hard and electrically conductive materials can be shaped and given detailed characteristics using the non-traditional machining technique known as electro-discharge machining (EDM). Artificial Intelligence and Machine Learning are being utilised to predict high precision operations in the intricate electrochemical discharge machining process. In the proposed work, during an electro-discharge chemical machining (EDM) operation, the support vector machine (SVM), a supervised learning system developed using a novel artificial intelligence paradigm, is used to anticipate three responses, including pulse time on, pulse time off, current, and servo feed. In comparison to linear and quadratic models, the SVM-based results demonstrate good agreement between the experimental and projected response values. Servo feed is the parameter of the four used in the EDM process that has been found to have the greatest impact on the outcomes under consideration. Based on the figures for accuracy, Reminder, correctness, and F1, RF has higher ratios across all four measurement scales, with respective values of 93.95, 92.62, 89.62, and 91.77%. In order to maintain high levels of precision in their job, RFs have the highest chance of determining the best machining condition and performance.