Conformal principal-asymptotic nets: Parametric design and modular construction of elastic gridshells

Abstract

Current modular design approaches often rely on numerical optimization to satisfy fabrication constraints, which can be computationally intensive and sensitive to initial geometric conditions. This work introduces a parametric design algorithm based on Conformal Principal-Asymptotic (CPA) networks on Minimal Surfaces (MS). This “pre-rationalized” approach enables the rapid generation of diverse shell morphologies while ensuring that construction logic, such as developability and node-orthogonality, is constantly preserved. The method provides a versatile design space for both surface geometry and beam configurations, while also facilitating geometric analyses of assembly-induced stresses. Our methodology is validated through a full-scale experimental timber model, providing a physical proof-of-concept that demonstrates a streamlined workflow for the design and modular assembly of complex elastic gridshells.