An intelligent hybrid optimization approach to improve the end milling performance of Incoloy 925 based on ANN-NSGA-II-ETOPSIS

To improve the machinability of Ni-based superalloy, optimization of tool nose radius and cutting parameters is one of the effective ways of dry machining, which leads to green manufacturing. However, combined investigation and optimization of tool nose radius and cutting process parameters concerned with cutting force, surface finish, and microhardness of machined part still need to be completed. In this study, we propose an intelligent hybrid approach to improve the machinability of Incoloy 925. Analysis of variance (ANOVA) is used to investigate the combined effects of the radius of inserts and process parameters on the machining performance. Response Surface Methodology (RSM) and Artificial Neural Network (ANN) are applied to develop the mathematical models between cutting parameters and technological responses, and their prediction results are compared. Non-dominated Sorting Genetic Algorithm (NSGA-II)-based on ANN models are used to get the optimal set of parameters. In order to get the best optimal solution from the optimal set of parameters, the Technique for Order Performance by Similarity to Ideal Solution (TOPSIS) coupled with Entropy (E) is used. ‘ANN-NSGA-II-ETOPSIS’ predicts the best possible efficient results near to experimental results of cutting force, surface roughness, and micro-hardness with an accuracy of 95.66%, 91.94%, and 90.07%, respectively. Moreover, surface quality and tool wear at optimal machining parameters are better than that of low- and high-level machining parameters. This proposed model will be helpful for industry to select the best cutting parameters for multi-objective optimization in machining of Ni-based alloys.