A Taguchi-based optimization on EDM parameters for enhancing surface roughness in ductile cast iron: Artificial Neural Networks (ANN)-validated approach
Abstract
Abstract This study uses the Taguchi approach to conduct a thorough examination of the Electrical Discharge Machining (EDM) process for optimizing surface roughness (R a ) on ductile cast iron. The effects of major EDM parameters—peak current, pulse-on time, pulse-off time, and jet pressure—were examined using solid copper electrodes. Experimental results showed that using a lower peak current of 5 A, a pulse-on time of 6 µs, and a jet pressure of 10 kg/cm 2 reduced surface roughness to 2.157 µm. A higher peak current of 15 A, pulse-off duration of 5 µs, and jet pressure of 20 kg/cm 2 resulted in a maximum surface roughness of 4.853 µm. The ANOVA findings showed that current was the most relevant parameter, accounting for 62.81% of the total variation in surface roughness, followed by jet pressure (11.69%). The study also used an artificial neural network (ANN) model to predict surface roughness, which yielded a high correlation coefficient (R = 0.95773), verifying the experimental results and displaying great predictive ability. These findings emphasize the importance of current and jet pressure in determining surface finish during the EDM process. The findings contribute to improving the precision and efficiency of EDM, presenting substantial potential for applications in industries that need high-quality machining of sophisticated materials, such as aerospace, automotive, and heavy engineering sectors.