Algorithms for efficient process-planning and automated tool sequence selection for 2.5-axis and 3-axis rough milling operations.

摘要

The cost of manufacturing by milling is strongly influenced by the tool sequences selected to mill out removal volumes. This thesis develops algorithms that automate and optimize this process for 2.5 axis and 3 axis rough milling.; The problem of selecting tool sequences has been formulated as a shortest path search in a single-source single-sink directed acylic graph under certain assumptions. The nodes in the graph represent the state of the stock after a particular tool is done machining. The weights of the edges represent the cost of machining.; Using a sequence of tools to machine a pocket necessitates the decomposition of the pocket into sub-pockets for each tool in the sequence. An associated problem of pocket decomposition is open-edge covering where the boundary between sub-pockets assigned to different tools has to be traversed over by a tool for complete machining. Algorithms have been developed for extending sub-pockets to cover inter sub-pocket boundaries.; Tool holder collision becomes a serious issue when machining parts that have nested pockets. Algorithms have been developed in this thesis to incorporate the tool holder collision into the optimal tool sequence selection problem. If tool holder collisions exist, these algorithms eliminate the generation of redundant data and thus reduce the complexity of building the tool sequence graph.; A cost model has been developed for accounting not only machining costs but also tool wear costs. The machining costs are derived from the generation of actual tool paths. The cost model can be tuned to either optimize total cost or time. A concise schema for a tool database has been developed. This tool database not only captures tool geometry but also associated cutting parameters depending on the work piece material and the type of cutting use.; Air-time is the non productive time spent by a tool in traversing from one machining region to another. This time can be significant, especially when multiple tools are used to machine a part. Smaller tools in particular may have to traverse over a wide area to reach several disconnected regions. The problem of minimizing air-time has been formulated as a variant of the standard traveling salesman problem, called the sequential ordering problem.