Automatic Parametric Modeling Technique for Structural Design Standardization

The challenge in designing new three-dimensional (3D) shapes is that considerable manual effort required for repeated and verification to satisfy the specifications. Extracting standardizing parametric design rules from a designed 3D shape dataset will be useful drastically reducing time. This paper describes framework of automatically generating models, rules, order standardize designs structural systems. In this framework, whose dimensions have relationships with target extracted using retrieval method. method an improvement conventional similar After extraction dataset, corresponding parts between collected are identified models generated. has two improvements over uses images captured multiple directions. First, converted unit by stretching or shrinking along each axis. Second, special artificial intelligence (AI) model learns relationships. effectiveness proposed evaluated data various shapes. Although accuracy rate similar-shape based on was 62% method, AIs, it improved 99% Moreover, were identified, been demonstrated.