Multi-Objective Optimization of the Surface Grinding Process for Heat-Treated Steel

Abstract

This paper effectively integrates Taguchi, Response Surface Methodology (RSM), and Genetic Algorithm (GA) approaches for both single and multi-objective optimization, delivering low-cost, high-effectiveness solutions for grinding. It examines the effects of three factors—depth of cut in coarse grinding, depth of cut in fine grinding, and the number of spark-outs—on three objectives: surface roughness, grinding time, and the deviation between the desired and actual grinding depth for SKD61 steel. Through in-depth analysis, the paper describes the impact of these factors and their interactions with the responses. It also proposes the optimal parameter setup for each objective. The optimal grinding time is achieved at 950 seconds with a coarse grinding depth of 0.007 mm, fine grinding depth of 0.004 mm, and zero spark-out. The minimal deviation and surface roughness were obtained at 0 mm and 0.144 µm, respectively, using the optimal setup of a coarse grinding depth of 0.004 mm, fine grinding depth of 0.001 mm, and 10 spark-outs. By applying GA, the paper provides a Pareto solution set, offering multiple combinations of optimal factors for minimizing grinding time, deviation, and surface roughness. These solutions serve as useful references for users seeking the best trade-offs in multi-objective scenarios. This paper contributes to improving customer satisfaction by enhancing the quality and efficiency of the machining process while reducing production costs for grinding machines. Its methodology can also be applied to other optimization fields.