Multiobjective optimization of cutting parameters in Ti-6Al-4V milling process using nondominated sorting genetic algorithm-II

The present article established empirical models of tool life, residual stress, and surface roughness according to titanium alloy milling parameters. The empirical models were utilized for optimization of production cost and surface quality. As the effects of milling parameters on production cost and surface quality are conflicting in nature, the multiobjective optimization problem in titanium alloy milling was proposed. Considering different industrial demands, two optimization objectives were established, optimization objective I aims to minimize the production time per piece and the number of consumed tools, the objective II aims to minimize the production time per piece, surface roughness, and absolute value of residual stress. In addition, the coupling of the two optimization objectives is considered to optimize the production cost and surface quality simultaneously. Nondominated sorting genetic algorithm-II (NSGA-II) was adopted to solve the multiobjective optimization problem and the optimized results were obtained by the Pareto-optimal solutions. These pareto-optimal solutions were used to conduct the verification experiments. Comparison of experimental and optimized results shows that the relative errors of tool life, surface roughness, and residual stress are less than 5, 7, and 5 %, respectively. The achieved results can provide the beneficial guidelines for the engineering applications depending upon industrial demands.