NC process information mining based optimization method of roughing tool sequence selection for pocket features

The appropriate and intelligent selection of roughing tool sequence for pocket features is essential to improve the efficiency of NC machining. However, there exist few researches exploring how to discover and utilize the valuable NC process information imbedded in 3D CAD models. In this paper, a NC process information mining based optimization method of roughing tool sequence selection for pocket features is presented. Firstly, the medial axis transform (MAT) is introduced to represent the tool paths of a pocket feature, and corresponding parameters of MAT are calculated. Secondly, considering the restrictions of tool movements during machining, the mapping mechanism between geometry and NC process is elaborated based on MAT to reveal the association of 3D CAD models and NC machining. The deeper information for NC process planning is mined to reflect the machining procedure feasibly. Then, the multi-objective optimization model is constructed by considering material removal amount and cutting consistency synthetically. Moreover, the precedence and distances between roughing tools are formulated in a Candidate Tools Graph (CTG). Finally, a hybrid ant colony algorithm (ACA) and simulated annealing (SA) approach based on CTG is proposed to generate the global optimal roughing tool sequence. In the experiment, various pocket features are conducted to illustrate the application effectiveness of the proposed method. The experimental results show that our method can achieve highly satisfactory results and outperforms other approaches.