Title: Application of a general regression neural network for predicting radial overcut in electrical discharge machining of AISI D2 tool steel
Authors: M.K. Pradhan; Raja Das
Addresses: Department of Mechanical Engineering, Maulana Azad National Institute of Technology, Bhopal, India ' School of Advanced Sciences, VIT University, Vellore, Tamil Nadu, India
Abstract: In the present research, an artificial neural network (ANN) model has been presented to investigate the variation of the radial overcut, in the electrical discharge machining (EDM). The radial overcut significantly influences the precision and accuracy of the product. Due to the nonlinear, complex relationship amongst the electrode wear, the electrode diameter, EDM parameters, and the machine positioning accuracy the final work piece dimensions are difficult to predict. A full factorial design was used to conduct the experiments the obtained data were segregated in three parts to train the network, to testing for convergence and finally to validate the model. The mean square error convergence criteria, both in training and testing, came out very well. The developed models are found to approximate the responses quite accurately. Results reveal that the proposed models can be employed successfully in the prediction of the stochastic and complex EDM process.
Keywords: electrical discharge machining; EDM; artificial neural networks; ANNs; general regression neural networks; GRNNs; radial overcut; electro-discharge machining; tool steel; electrode wear; electrode diameter; positioning accuracy; modelling.
International Journal of Machining and Machinability of Materials, 2015 Vol.17 No.3/4, pp.355 - 369
Received: 05 Feb 2015
Accepted: 21 Mar 2015
Published online: 27 Sep 2015 *