Dynamic response analysis of a ball-end milling cutter and optimization of the machining parameters for a ruled surface.

Cutter displacements and deflections induce a significant amount of surface error in machined thin-walled parts and are a major obstacle to achieving higher productivity in a five-axis milling operation. In this study, a milling force model for the five-axis flank milling process was established and implemented by calculating and simulating the cutter dynamic displacement to investigate the dynamic response of ruled surface impeller blades in the milling process. A machining trajectory and related machining parameters were selected to calculate the dynamic milling forces in the X and Y directions for each analysis time using the calculation method. Using ANSYS software, the milling forces in three directions were input into the finite element model of the ball-end milling tool. The deformation was analyzed using post-processing for the ruled surface milling. Then, the milling parameters related to the dynamic displacement of the tool in the side milling blade were trained and optimized using a back-propagation neural network and particle swarm optimization algorithm during the side milling of the ruled surface. In addition, processing experiments were performed based on the optimized parameters. The blade roughness tests indicated that the surface quality was considerably improved after optimization, thus verifying the feasibility of the optimization method. [ABSTRACT FROM AUTHOR]