OPTIMIZATION OF MATERIAL REMOVAL RATE AND SURFACE ROUGHNESS IN MILLING OF 7075 ALUMINUM ALLOY USING THE ARTIFICIAL BEE COLONY ALGORITHM

Abstract

This article presents the optimization results of material removal rate and surface roughness in milling 7075 aluminum alloy parts using a 5-axis CNC milling machine and the Artificial Bee Colony (ABC) algorithm. The input parameters include feed rate (150–450 mm/min), cutting depth (0.1–0.4 mm), and spindle speed (3500–5500 rev/min), while the optimization objectives are material removal rate and surface roughness. The experimental data, obtained from 27 experiments, were processed using regression analysis and artificial neural network before being applied to the ABC algorithm. The optimization results were compared with those obtained using the Response Surface Methodology (RSM) and further validated through 9 experiments. The findings indicate that: (1) the regression-based data processing approach provides higher accuracy than the artificial neural network method; (2) the optimal value of spindle speed for all values of weight of optimization objectives are equal or close to 5500 rev/min, while that for cutting depth and feed rate depends on the weight of optimization objectives and optimization methods; and (3) the deviation between the predicted and experimental optimal values is less than 10% for both objective functions.