A Comparative study in heat-assisted machining of Inconel 718 and Hastelloy-276 using machine learning techniques

Abstract

Machining parameters have an important role in material removal, tool wear, and surface finish. The manufacturers need to obtain optimal operating parameters with a minimum set of experiments and minimize the simulations to reduce machining set up costs. Due to more demanding manufacturing systems, the focus is on applying machine learning (ML) algorithms in heat-assisted turning in which critical process parameters are predicted. Based on the predictions, the machining parameters can be altered to avoid essential conditions of the process. The experimental results of Inconel 718 and Hastelloy C-276 are analyzed to compare the surface roughness and flank wear using ANN and SVR methods of machine learning. The performance of the models in predicting the parameters is presented to improve further the machining process. Finally, ANN showed a more accurate and effective method in predicting the heat assisted turning responses for the selected machining parameters.