Prediction and Optimization of Surface Roughness in Electrical Discharge Machining Using Machine Learning Techniques

The performance of machined components in precision manufacturing depends on the ability to predict and control surface roughness. Because it uses a non-contact thermal erosion technique, Electrical Discharge Machining (EDM) is frequently used to shape materials that are difficult to machine. Yet, the useful intricate relationship between surface characteristics and machining parameters makes it challenging to achieve the ideal surface finish. is difficult due to the complex relationship between machining parameters and surface characteristics. This research presents a machine learning-based model to predict and optimize surface roughness (Ra) in EDM. This study employs Support Vector Regression (SVR) in conjunction with a metaheuristic optimization technique to modify hyperparameters and enhance prediction accuracy, drawing inspiration from the approach utilized in "Prediction and Optimization of Surface Roughness for Laser-Assisted Machining SiC Ceramics Based on Improved Support Vector Regression." The model is trained and validated with experimental EDM datasets that include discharge current, pulse-on time, pulse-off time, and voltage as input features. The new hybrid optimization model shows better prediction accuracy than traditional regression methods and standalone SVR, achieving a coefficient of determination (R²) greater than 0.98 and an average prediction error below 3%. This system provides a reliable, data-driven way to optimize EDM parameters for the best possible surface finish.