Abstract:  Aiming at the problem of large deformation in the local area of annular thinwalled aluminum alloy parts, a multi-island genetic algorithm was proposed to optimize the milling parameters of thin-walled parts. Through the finite element analysis software, a threedimensional milling simulation was performed on this area to obtain the milling force. Based on the Isight platform, the milling force was taken as the optimization target, and the three milling parameters of spindle speed, radial depth of cut and axial depth of cut were used as the optimization parameters. The optimal Latin hypercube test method was used to design the sample points, the response surface approximate model method was used to fit the approximate model, and the approximate model was optimized by the multi-island genetic algorithm. The research results show that the response surface approximation model accurately fits the functional relationship between the milling parameters and the milling force, and combining with the optimization of the multi-island genetic algorithm, the milling force is reduced by 38.2%, which effectively reduces the milling force in the semi-finishing stage. The deformation of the annular thin-walled aluminum alloy parts is significantly reduced, which verifies the feasibility of the multi-island genetic algorithm for optimizing the milling parameters.

Key words:  thin-walled parts;processing deformation;multi-island genetic algorithm;optimizing Latin hypercube;approximate model;Isight

HAN Jun, CAO Long-kai, XU Rui, et al. Optimization of milling parameters of annular thin-walled aluminum alloy parts based on multi-island genetic algorithm［J］.Journal of Mechanical & Electrical Engineering, 2022,39(1):100-106.