Cutting Forces, Surface Roughness and Tool Wear Quality Assessment Using ANN and PSO Approach During Machining of MDN431 with TiN/AlN-Coated Cutting Tool

Abstract

The aim of this study was to improve the life and performance of tungsten carbide turning tool inserts coated with TiN/AlN multilayer thin films using physical vapor deposition technique. Quality characteristics of the coating are evaluated using Calo and VDI 3198 tests. Thickness of the coating is found to be $$3.651 \,\upmu \hbox {m}$$ with adhesion quality of HF1. The performance of coated tool inserts is evaluated using cutting speed (59–118 m/min), feed rate (0.062–0.125 mm/rev) and depth of cut (0.2–0.4 mm) as process parameters in turning MDN431 steel. Experimental investigation has been carried out based on full factorial design, and regression analysis was used to analyze and build the mathematical models for cutting force and surface roughness. Multi-objective optimization of the process parameters has been done with the combination of desirability approach and MOPSO technique. Optimum machining condition for least cutting force and optimum surface roughness is found to be $${V}_{\mathrm{c}} =59\, \hbox {m/min}$$, $${f}=0.063\,\hbox {mm/rev}$$ and $${a}_{{p}} =0.2\,\hbox {mm}$$. Cutting force and surface roughness are reduced by 9% in TiN/AlN-coated tools compared with the uncoated tool. To improve the CoD and capability of predictive regression models, ANN modeling has been adopted. ANN trained model and mathematical regression models are used to predict the results and predict the responses, which follow the experimental data with minimum absolute error. The predicted results are validated using ANN and regression analysis found with minimum error, and developed models are adequate for further usage. Tool wear was reduced by 105% in TiN/AlN-coated tools compared with the uncoated tool.