Cutting parameters play an important role in the milling process. Proper scheduling of cutting parameters will improve the performance of a five-axis machine tool while an unoptimized one may lead to the damage of machine tool under extreme conditions. In the context of five-axis milling, the cutting parameters optimization has become a necessary procedure to realize the entire potential of the expensive and sophisticated machine.

Short tool service life is always a significant concern when milling hard materials such as Ni-based superalloy. While most researchers focus on improving tool material and surface coating technology, in this thesis tool wear is alleviated by choosing optimized cutting parameters for five-axis milling. In this thesis, aiming at averaging the tool wea...[ Read more ]

Cutting parameters play an important role in the milling process. Proper scheduling of cutting parameters will improve the performance of a five-axis machine tool while an unoptimized one may lead to the damage of machine tool under extreme conditions. In the context of five-axis milling, the cutting parameters optimization has become a necessary procedure to realize the entire potential of the expensive and sophisticated machine.

Short tool service life is always a significant concern when milling hard materials such as Ni-based superalloy. While most researchers focus on improving tool material and surface coating technology, in this thesis tool wear is alleviated by choosing optimized cutting parameters for five-axis milling. In this thesis, aiming at averaging the tool wear on the entire cutting edge and hence prolonging the tool service life, a study is reported on how to generate a multi-layer toolpath with different optimized tool lead angles and cutting depth for multi-axis milling of an arbitrary freeform surface from an initial raw stock. The optimized cutting parameters are guaranteed to be free of chatter which is well known for its detrimental effect on the cutting edge. In this study, the chatter stability lobe diagram is experimentally generated. It reveals the relationship between the lead angle and the cutting depth. Under the constraint of chatter stability lobe diagram, the machining toolpath is generated by selecting a proper pair of the best lead angle and cutting depth along the toolpath. While the proposed algorithm currently is restricted to the iso-planar type of toolpath, it can be adapted to other types of milling. The physical cutting experiments performed in this thesis have convincingly confirmed the advantage of the proposed machining strategy as compared to the conventional constant lead angle and constant cutting depth strategy – in the tests the maximum wear on the cutting edge is reduced by as much as 39%.

While the cutting parameters can be effectively optimized by the state-of-the-art algorithms, the efficiency of optimization is not satisfying yet. Taking the feed rate optimization as an example, the most conservative strategy requires the deformation prediction of the in-process workpiece (IPW) on all CC points to obtain the limit of tolerable feed rate. However, the current conventional method of predicting deformation on consecutive CC points is too tedious and not flexible enough to support the demand for massive calculation. The statistic and dynamic characteristic of the in-process workpiece is an important basis of constraint calculation in various aspects of tool path planning and cutting parameter optimization. Therefore a new methodology should be developed to improve the efficiency of predicting the deformation or eigenmode of the in-process workpiece. In this thesis, the voxel model and the finite cell method are introduced to propose a new methodology of predicting the IPW deformation on consecutive CC points. The methodology consists of three major components: the fast updating of the voxel model, the fast cutting force calculation method and the partial updating technique of calculating the stiffness matrix in the finite cell method. The proposed method performs well in efficiency with only 5 seconds of deformation analysis per CC point and acceptable accuracy.