Predictive model development in dry turning of Nimonic C263 by artificial neural networks

In the present article, a predictive model using ANN is developed to predict the machining attributes like cutting force, temperature at cutting edge, surface roughness and flank wear in turning of Nimonic C263 alloy using cubic boron nitride (CBN). The experimental results and developed model was found to be less percentage error. The average percentage error for like cutting force, temperature at cutting edge, surface roughness and flank wear among experimental and predicted values were in the range of 1.57%, 1.07%, 1.48%, and 3.55% respectively. The predictive model would be useful to predict as well to forecast the machining attributes prior to the experiments. The model with 9-4-1 architecture is found to be best to predict the machining attributes.Copyright (c) 2021 Elsevier Ltd. All rights reserved. Selection and peer-review under responsibility of the scientific committee of the International Conference Virtual Conference on Technological Advancements in Mechanical Engineering