Logic-Informed Graph Neural Networks for Structural Form-Finding

• Published in: Advanced engineering informatics 2024-08, Vol.61, p.102510, Article 102510
• Main Authors: Bleker, Lazlo, Tam, Kam-Ming Mark, D’Acunto, Pierluigi
• Format: Article
• Language: English
• Subjects: Combinatorial Equilibrium Modeling; Graph Neural Networks; Machine Learning; Structural Form-Finding
• ISSN: 1474-0346
• DOI: 10.1016/j.aei.2024.102510

Computational form-finding methods hold great potential concerning resource-efficient structural design. The Combinatorial Equilibrium Modeling (CEM), an equilibrium-based form-finding method based on graphic statics and graph theory, allows the design of cross-typological tension–compression structures starting from an input topology diagram in the form of a graph. This paper presents a novel Logic-Informed Graph Neural Network (LIGNN) that integrates the validity conditions of CEM topology diagrams into the learning process through semantic loss terms. A Primary-LIGNN (P-LIGNN) and a Modification-LIGNN (M-LIGNN) are introduced and incorporated together with the CEM into a general form-finding-based computational structural design workflow that transforms input topologies into parametric models of equilibrium structures. An implementation of this computational design workflow for the conceptual design of pedestrian bridge structures is made, and presented through a case study, for which a synthetic training dataset of topology diagrams for the LIGNNs has been developed. [Display omitted] •Logic-Informed Graph Neural Networks (LIGNN) implementing discrete logic-based rules•New synthetic dataset of graphs for Combinatorial Equilibrium Modeling (CEM)•Semantic loss function formalizing the CEM validity conditions for graph labeling•Machine Learning-assisted design workflow based on Structural Form-Finding•Case study of a human–machine collaborative design of a pedestrian bridge structure