Abstract
Manufacturing industry tends to high-mix low-volume production in recent years. Therefore, it is strongly required to develop a computer aided process planning (CAPP) system to shorten the preparation time and to generate NC program. Feature recognition has been treated as a key technology to develop a CAPP system. In authors’ previous study, novel machining features for multi-tasking machine tools have been proposed. The machining features can correspond to several alternative machining methods. However, complex target shapes of practical mechanical parts have not been considered. In order to solve this problem, special shapes such as chamfer part and freeform surface are firstly approximated and machining features such as cylinder or cuboid are recognized from the converted simple shapes in this study. From the results of conducted case study, it is found that the proposed feature recognition method has a potential to deal with complex target shapes of practical mechanical parts.
