Machining Feature Recognition for Cast Then Machined Parts

Abstract We present a method to recognize machining features for the domain of cast-then-machined parts. Non-interacting volumetric machining features are recognized through a face pattern based recognition approach, and are filtered out of the part model. From the filtered part model and the specification of part surfaces as being cast or machined, we systematically generate the surface machining features and the starting workpiece, which represents the casting output in sufficient detail to support machining process planning. By subtracting the filtered part from its starting workpiece, we obtain the removal volume that is to be realized through machining operations. We apply the feature recognition method using Alternating Sum of Volumes With Partitioning (ASVP) Decomposition to decompose this removal volume into volumetric machining features.