Corner-milling process parameter optimization regarding H62 brass using Kriging model and improved particle swarm optimization algorithm

Single-pass corner milling is common in machining process. Machining quality and cost are the two most important goals for machining process. Machining quality is greatly affected by deformation and mechanical property. Cutting temperature has a very important effect on tool wear, mechanical property, deformation and machining precision. In this paper, cost per volume (CPV) and machining temperature are selected as objectives. To reduce the cost of research, AdvantEdge simulation software was used to simulating the machining process and examined for its accuracy. Comparing the result of AdvantEdge with physical experiments, it showed an average 8% error on CPV and a 6% error on temperature. Given this comparison, simulation data were used to train Kriging model as the surrogate model. Kriging model manifested high accuracy, showing mean error of 6% and 3% in terms of CPV and temperature on validation points. To solve the multi-objective problem, a K-means particle swarm optimization (PSO) was constructed, which outperformed traditional adaptive gird algor (AGA) by finding the better optimal solutions, meanwhile consuming lesser time. Optimization result showed that corner milling should have a low-level spindle speed to ensure low cost and low temperature. High feed rate, cutting depth and cutting width will result in low cost along with high temperature. Optimal parameters should be chosen from Pareto set according to different practical needs. This integrated method manifested its applicability in solving multi-objective problems.