Rationalization of freeform space-frame structures: Reducing variability in the joints. (March 2020)

• Record Type: Journal Article
• Authors: Koronaki, Antiopi; Shepherd, Paul; Evernden, Mark
• Is Part Of: International journal of architectural computing. Volume 18:Number 1(2020)
• Page Start: 84
• Page End: 99
• Publication Date: 2020-03
• Subjects: Geometry optimization -- space-frame structures -- joint -- fabrication process -- construction -- cost -- clustering -- control surface; Architecture -- Data processing -- Periodicals; Architecture -- Informatique -- Périodiques; Virtual reality in architecture -- Periodicals; Computer-aided design -- Periodicals; Architecture -- Data processing; Periodicals; 720.2840285536
• DOI: 10.1177/1478077119894881
• Languages: English
• ISSNs: 1478-0771

Abstract

In recent years, the application of space-frame structures on large-scale freeform designs has significantly increased due to their lightweight configuration and the freedom of design they offer. However, this has introduced a level of complexity into their construction, as doubly curved designs require non-uniform configurations. This article proposes a novel computational workflow that reduces the construction complexity of freeform space-frame structures, by minimizing variability in their joints. Space-frame joints are evaluated according to their geometry and clustered for production in compliance with the tolerance requirements of the selected fabrication process. This provides a direct insight into the level of customization required and the associated construction complexity. A subsequent geometry optimization of the space-frame's depth minimizes the number of different joint groups required. The variables of the optimization are defined in relation to the structure's curvature, providing a direct link between the structure's geometry and the optimization process. Through the application of a control surface, the dimensionality of the design space is drastically reduced, rendering this method applicable to large-scale projects. A case study of an existing structure of complex geometry is presented, and this method achieves a significant reduction in the construction complexity in a robust and computationally efficient way.