Artificial neural network based prediction of responses on eglin steel using electrical discharge machining process. (2020)

• Record Type: Journal Article
• Authors: Martin Sahayaraj, J.; Arravind, R.; Subramanian, P.; Marichamy, S.; Stalin, B.
• Is Part Of: Materials today. Volume 33:Part 7(2020)
• Page Start: 4417
• Page End: 4419
• Publication Date: 2020
• Subjects: Electrical discharge machining process -- Artificial neural network -- Eglin steel -- Radial basis function -- Analysis of variance; Materials science -- Congresses -- Periodicals; 620.1
• DOI: 10.1016/j.matpr.2020.07.664
• Languages: English
• ISSNs: 2214-7853

Abstract

Abstract: The contribution of steel in metal industries is rapidly increasing due to its excellent properties. Eglin steel is one of the high strength steel which is used in shield, aerospace, bridges and commercial purposes. In this work, the Eglin steel is machined by Electrical Discharge Machining (EDM) process due to its high hardness. Artificial Neural Network (ANN) is used to forecast the outcome, such as Material Removal or Material Deletion Rate (MDR). Radial Basis Function (RBF) is used to develop the ANN model which is used to predict the responses. The optimum architecture is obtained through MATLAB by control the neurons and hidden layers. The errors are determined through the comparison of experimental results and ANN predicted results. The material properties and the contribution of each parameter are also discussed through analysis of variance.