Prediction of tool wear length in micro-EDM process using an artificial neural network

Abstract

Consideration of tool wear (TR) is the most significant parameter for machining in the micro-EDM process. Frequent tool wear is a major problem in the electric discharge machining process. This has a negative impact on the geometrical correctness of the workpiece's machined features. An accurate prediction can provide crucial information for a manufacturer to develop accurate strategies for machining parts. In this paper, an artificial neural network model has been developed using a backpropagation algorithm. The developed model has been used to predict the tool wear of the electrode used in the Micro-EDM process, which is one of the relevant and challenging areas which can be addressed by the application of advanced management and computation theory. Further, alternative predicting models have been developed using time series and regression techniques. A comparative study has been performed for these three methods. It was observed that the artificial neural network-based method provides better results in comparison with time series and regression-based models with the highest coefficient of determination (R2) value.