Open AccessA study on die sinking EDM of Nimonic C-263 super alloy : an intelligent approach to predict the process parameters using ANN
Author(s) -
Rama Bhadri Raju Chekuri,
R. Murthy Kalluri,
S. Rajesh,
Jamaleswara Kumar Palakollu
Publication year - 2017
Publication title -
international journal of engineering and technology
Language(s) - English
Resource type - Journals
ISSN - 2227-524X
DOI - 10.14419/ijet.v7i1.1.10796
Subject(s) - nimonic , electrical discharge machining , artificial neural network , machining , taguchi methods , surface roughness , materials science , linear regression , superalloy , computer science , alloy , mechanical engineering , artificial intelligence , metallurgy , engineering , machine learning , composite material
In current study, machining characteristics of Nimonic C-263 are analysed by TAGUCHI and modelled using Artificial Neural Networks (ANN). The response parameters under consideration are Material Erosion Rate (MER), Electrode Wear Rate (EWR), Surface Roughness (SR) and Dimensional Overcut (DOC). A regression mathematical model is also developed to verify the capabilities of ANN. The modelling of ANN includes identifying appropriate combination of hidden layers and number of neurons in each hidden layer. Study on machining characteristics revealed, peak current as the most influential process parameters affecting all the responses; followed by Pulse on-time. A contrary effect is observed for Pulse off-time. A rare process parameter named flushing pressure showed negligible influence on responses. Among various ANN architectures, 6-6 architecture is noted to possess phenomenal prediction accuracy of 99.71% compared to 93.55% of regression analysis.