Predictive Modeling And Optimization Of Cutting Parameters In High Speed Hardened Turning Of Aisi D2 Steel Using Rsm, Ann And Desirability Function

High speed machining (HSM) is an attractive process for numerous applications due to its potential to increase production rates, reduce lead times, lower costs, and enhance part quality. In this study, high-speed turning operations on AISI D2 steel using a coated carbide cutting tool under dry conditions were conducted. The cutting parameters examined in this investigation were Vc, f, and ap, while the outputs measured were surface roughness (Ra), cutting temperature (T), and flank wear (VB). To obtain reliable and accurate results, a Taguchi L27 orthogonal array for the 27 experimental runs was employed as well as analysis of variance (ANOVA), response surface methodology (RSM), and artificial neural network (ANN) to develop a constitutive relationship between prediction responses and the cutting parameters. The ANOVA results showed that Vc had a significant effect on T (36.81%) and VB (27.58%), while f had a considerable influence on Ra (24.21%). Additionally, nonlinear prediction models were created for each measured output and their accuracy was evaluated using three statistical indices: coefficient of determination (R2), mean absolute percentage error (MAPE), and root mean square error (RMSE). Finally, multi-objective optimization was successfully carried out using the desirability function (DF) approach to propose an optimal set of cutting parameters that simultaneously minimized Ra, T, and VB. The optimized cutting parameters were Vc = 477.28 m/min, f = 0.08 rev/min, and ap = 0.8 mm, resulting in Ra = 1.23 μm, T = 129.9∘C, and VB = 0.049 mm.