Abstract
This study develops a neural network based methodology for recognizing manufacturing feature (MF) using the boundary representation (B-rep) information. The methodology is capable of recognizing not only basic MF (including standard MF and non-standard MF) but also interacting MF. Firstly, both the edges' convex concave attribute and the edge's position (whether an edges belongs to the inner edge loop or outer edge loop), which reflects the edges' characteristics and the relationship between the bottom profile and the adjacent faces, were analyzed to defined the input vector for the neural network. Based on this, a BP neural network with a single hidden layer which contained 10 neurons was obtained. The basic MF is divided into additive MF (AMF), throuth subtractive closed MF (through SCMF), blind subtractive closed MF (blind SCMF), subtractive opened MF (SOMF) four groups, and can be recognized easily using the neural network has been trained. For the interacting MF, a basic MF is recognized firstly, the bottom profile is updated when a MF interactive with the recognized basic MF is identified. The height or depth of the MF is determined by the height of the lowest adjacent face. The MF has been recognized will be added to the part, the following recognition will be carried on based on the new part model.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call