MACHINABILITY INVESTIGATION OF HIGH-STRENGTH 7068 ALUMINUM ALLOY: A POTENTIAL STUDY THROUGH EXPERIMENTATION, SPOTTED HYENA OPTIMIZATION AND ADVANCED MODELING APPROACHES

Abstract

This research emphasizes the machinability investigation on CNC turning of 7068 aluminum alloys. CVD-coated carbide tool was implemented for the [Formula: see text] full-factorial-based turning experiments in dry conditions. Machinability study includes the assessment of flank wear, cutting tool vibration, surface roughness, cutting temperature, chip reduction coefficient, and chip morphology. The selected tool performed well as very low wear (0.030–0.045[Formula: see text]mm) and low surface roughness (0.28–1.14[Formula: see text][Formula: see text]m) were found. All the input variables have significant impact on the flank wear, cutting tool vibration, cutting temperature, and chip reduction coefficient while for surface roughness, the effects of cutting speed and feed were significant at the 95% confidence level. Further, a novel optimization tool namely the spotted hyena optimizer (SHO) algorithm was utilized to get the optimal levels of input variables. Additionally, two different modeling tools namely multiple adaptive neuro-fuzzy inference system (MANFIS) and radial basis function neural network (RBFNN) were utilized for formulating the cutting response models. Further, the average of the absolute error was estimated for each model and compared. The MANFIS modeling tool exhibited a more close prediction of outputs as compared to RBFNN, as the obtained average absolute error for each response was lower with MANFIS.