Conference paper Open Access
Boussuge, Flavien; Tierney, Christopher M.; Robinson, Trevor T.; Armstrong, Cecil G.
Generating fit-for-purpose CAD models from complex assemblies is time consuming for analysts. Tedious tasks include to identify and isolate the components of interest for the analysis, remove duplicate components, or correct inconsistent components’ interfaces are common for large assemblies during the product development process. In this paper a new approach to help engineers analyse the consistency of CAD assembly models is proposed. The method utilises a tensor factorisation technique developed for relational machine learning and applies it on B-Rep topological and geometrical relations. The generated decomposition is used to identify which entities in the assembly are similar (within a threshold) to a selected input entity. The factorisation model regards globally all input relationships, e.g. the connections between components, to identify similar entities based on their relationships in the relational domain. It is shown that a hierarchical clustering method can group entities based on the similarities of their attributes and relationships.
|All versions||This version|
|Data volume||1.0 GB||1.0 GB|