Applications of Genetic Algorithms in Process-Planning: Tool Sequence Selection for 2.5D Pocket Machining

摘要

Rapid tool change mechanisms in modern CNC machines have enabled the use of multiple tools (sequence of tools) to machine a given pocket. Larger diameter tools that have higher material removal rates are used to clear large open spaces, smaller tools are used for clean up. The challenge lies in selecting that particular combination of tools that minimizes total cost. Previously, we developed algorithms based on network optimization to find the best tool sequence given a list of cutters, cutting parameters and pocket geometry. The formulation was based on certain assumptions that did not account for tool holder geometry. It also required the evaluation of all possible tool-pair combinations for a given tool set. This can get time consuming if the tool set is large. In this paper, we present a genetic algorithm based method to select optimal tool sequences. The algorithm was implemented and bench marked against the graph algorithm. We have found that the GA based method is able to find a near optimal tool sequence without evaluating up to 30% of all possible tool-pairs.