Computational Techniques to Improve Efficiency and Accuracy for High Performance Machining of Polyhedral Models

Abstract

The objective of this research is to investigate the computational techniques to improve the machining efficiency and accuracy for machining polyhedral models in CAD/CAM systems. High performance machining can be achieved by considering part surface geometry and physical property of machining processes in roughing, finishing and clean-up machining systematically.

To improve the machining accuracy, high accurate tool path interpolations are critical for CAD/CAM systems. By considering the maximal error conditions, explicit solutions are proposed to calculate the exact maximal interpolation error for interpolating 3D cubic polynomial curves and 2D planar offset curves. Using the explicit solutions of finding the maximal interpolation errors, an adaptive interpolation method is proposed for smooth curve interpolation for tool-path generation and numerical control machining.

To improve the tool-path generation efficiency for machining polyhedral models, an inverse cutting profile method is proposed for constructing Generalized Cutter Location (GCL) surfaces. The GCL-surfaces consist of facet-CL-surface, edge-CL-surface and vertex-CL-surface. GCL-surface can be calculated efficiently from the developed inverse cutting profile and the derived instantaneous offset vector of the cutting tool.

The finishing and clean-up machining can significantly improve the machining efficiency by using multiple-size tools. To shorten the total machining time, multiple tools are used to machine various regions with different geometric characteristics. A contraction tool method is proposed to identify the clean-up regions and to construct the strip-parallel clean-up tool paths. Using the proposed contraction tool method, clean-up tool paths can be successfully generated for improving machining efficiency.

By applying High-Speed Machining (HSM), Machining efficiency can be greatly improved due to the high material removal rate in HSM. Chattering is one of the most critical problems encountered in high-speed machining. To overcome the chattering in high speed machining, a time-domain computational modeling technique is proposed for chatter prediction. The time-domain computational modeling of chatter prediction is based on the trochoidal link-list surface models and Poincarè plot chatter criterion techniques. Compared with the traditional analytic solutions, the proposed time-domain chatter prediction solution can successfully generate the instant cutting information and can efficiently deal with non-linearity cutting in high-speed machining.

The presented techniques have been implemented and tested for feasibility. Computed implementation and practical examples are presented in this paper. The results show that the developed computational techniques can significantly improve the machining performance of complex polyhedral models. The presented techniques and detailed algorithms can be used in CAD/CAM/CNC systems for high performance machining.