A hybrid voxel-polygon framework for high-speed, high-accuracy five-axis machining simulations
摘要
In this study, a fast and highly accurate five-axis milling simulator that combines polygonal and voxel representations was developed. The simulator calculates the workpiece shape and cutting volume during machining. The workpiece was depicted by a voxel model containing two-dimensional stacked polygons. By representing the entire workpiece in voxels at the macroscale and locally supplementing shape information with polygons, this method achieves high spatial resolution with reduced computational costs. Furthermore, a tool polygon generation method suitable for Vatti clipping is proposed by rigorously classifying and formulating the ball-end mill cross-section under five-axis orientations into three patterns—circle, rectangle plus arc, and elliptical arc plus arc—and by approximating elliptical arcs with multiple circular arcs under controlled error. A case study on simple shape machining showed that the simulation time is correlated with the number of polygon vertices processed by Vatti clipping. In a case study on impeller machining, it was verified that simulating with a voxel size of 0.5 mm could be completed in the same time as actual machining at a spindle speed of 20,000 min− 1. Moreover, although reducing the pitch between layers improved the shape accuracy and cutting volume accuracy, the computation time increased. Additionally, the simulated workpiece surface reproduced geometric surface textures at the tool-mark scale with high accuracy.