Optimization of Process Parameters for Residual Stress Based on GA-BP-FA Algorithm
A GA-BP-FA multi-objective optimization model for residual stress control in multi-axis milling GH4169 is proposed.Genetic algorithm(GA)is used to optimize the initial weight and threshold of BP neu-ral network,which improves the rate of convergence and prediction accuracy of the model.A method for pre-dicting residual stress in multi-axis milling using the GA-BP model is proposed,and the influence of process parameters on residual stress is analyzed.In order to obtain the minimum residual tensile stress/maximum re-sidual compressive stress,the milling process parameters are optimized based on firefly algorithm(FA).The research results indicate that the established GA-BP prediction model for residual stress has high prediction accuracy,with average errors of 18.9%and 16.7%in both directions,respectively.After optimization by the FA algorithm,the optimal process parameter within the selected parameter threshold is the cutting inclination angle Φ=85°,cutting speed v=20 m/min,feed rate per tooth fz=0.01 mm/z.The optimal residual compres-sive stress is-463.43 MPa in the σx direction and-686.93 MPa in the σy direction.
NETL
NSTL
万方数据
