Optimisation of machining parameters of AISI 304L stainless steel with the least error method using Taguchi, RSM, and ANN

Abstract

This study has been conducted on the machining of AISI 304 L stainless steels, which are difficult to machine, have chip-breaking problems, and therefore cause premature wear. A full factorial experiment was designed, and the experiments were conducted at five different inserts, five different cutting speeds, two different feeds, and three different cutting depths. As a result of the experiments carried out, the flank wear and surface roughness values were measured. Utilising Taguchi, Response Surface Methodology (RSM), and Artificial Neural Networks (ANN), interpolation estimates were obtained for flank wear and surface roughness, followed by optimisation investigations. The method with the least error was selected by examining the estimation results. Different methods confirm each other, and simultaneously, the chosen method strengthens the estimation. Confirmation experiments were performed for the parameters giving the optimum value. The optimal outcomes in the range of cutting tips and parameters were obtained with the insert with Al2O3+TiCN coating at a cutting speed of 170 m/min, a feed of 0.13 mm/rev, and a depth of cut of 1.1 mm. It has been seen that the validation experiments agree with the actual and estimated values.