The dynamics of cutting operation necessitate the use of a reliable predictive model for accurate prediction of machining parameters, namely, cutting force (CF) and surface roughness (SR). This study predicts the cutting force and surface roughness of Al 6065 T6 during turning operation using a combined approach, namely, surface response methodology (RSM), machine learning (ML), and computer-aided simulation-based approach. The RSM was conducted in the Design Expert 2022 environment producing 20 experimental trials. The computer-aided modelling of the turning process was done in the complete Abaqus environment (CAE) while the ML technique was carried out in the Orange software environment using six ML models. To validate the experimental trials, turning operation was carried out on a centre lathe (type CTX 310 eco DMG) using carbide turning inserts as the cutting tool. The process parameters and the measured responses serve as the input into the ML model. The values of the process parameters that produced the least SR (1.02 휇m) are cutting speed (125 m/min), feed rate (0.4 mm/rev), and depth of cut (0.75 mm). The simulation result shows a slight variation in the strain profile of the workpiece from a minimum value of−2.334e−03 to a maximum value of+8.43e04, which indicates slight SR. The
feed rate (FR) had the highest influence on the magnitude of the CF and SR, followed by the cutting speed (CS), while the depth of cut (DoC) had the least influence. The outcome of this study demonstrates the feasibility of deploying an integrated approach for investigating the critical machining parameters such as CF and SR.
Keywords: Aluminium alloy · Cutting force · FEA · RSM · Surface roughness · Turning operation
