Predicting depth of cut in vibration-assisted EDM cutting on titanium alloy using adaptive neuro fuzzy inference system

In vibration assisted electrical discharge machining (VEDM) process, the depth of cut (DoC) affects the quality of the machined specimens. It is very important to predict this parameter for the better processing of specimens. In this investigation, Adaptive Neuro Fuzzy Inference System (ANFIS) was used to predict the depth of cut on a titanium alloy workpiece during VEDM. Three machining process factors, such as peak current, pulse-on time, and vibrational frequency, were used for training the model, due to the importance on depth of cut. The accuracy of the ANFIS model was assessed by comparing the predicted values to the measured ones. The findings indicated that all three parameters were statistically significant, with the most important being pulse-on time, followed by vibration frequency and peak current. The ANFIS model was proven superior to both linear and nonlinear regression models, as well as regression tree model, as it exhibited considerably lower MSE and MAPE values and it was shown that it can predict the depth of cut with a 97.3% level of accuracy.

Automated summary

In vibration assisted electrical discharge machining (VEDM), the depth of cut (DoC) has a significant impact on the quality of the machined specimens. This study utilized an Adaptive Neuro Fuzzy Inference System (ANFIS) to predict the DoC in VEDM of a titanium alloy workpiece. Three machining process factors, including peak current, pulse-on time, and vibrational frequency, were employed to train the ANFIS model, with pulse-on time being the most influential factor. The ANFIS model outperformed linear and nonlinear regression models, as well as the regression tree model, achieving a 97.3% accuracy in predicting the DoC.