A Generic Particle Model with Topologic Modeling Capabilities for the Computational Form-Finding of Elastic Structures

• Published in: Computer aided design Vol. 145; p. 103158
• Main Authors: Suzuki, Seiichi, Knippers, Jan
• Format: Journal Article
• Language: English
• Published: Amsterdam Elsevier Ltd 01-04-2022; Elsevier BV
• Subjects: Combinatorial analysis; Data structures; Design; Elastic deformation; Elastic structures; Exploratory form-finding; Graph theory; Mathematical models; Numerical models; Real-time physics; Simulation; Topologic modeling; Topology; Real-time physics; Exploratory form-finding; Elastic structures; Topologic modeling
• ISSN: 0010-4485; 1879-2685
• DOI: 10.1016/j.cad.2021.103158

Recent efforts on the research of bending- and tensile-active structures have shown the vast range of aesthetic and performative opportunities that elastic structures can offer for generating innovative architectural solutions. In this context, the task of conceptualizing and representing elastic structures is not trivial due to the presence of large deformations along the forming process. This condition demands to expand beyond current geometrical modeling schemes used in architecture and develop a more intuitive simulation-based design practice. Dynamic methods, such as Particle System and Dynamic Relaxation, have become an important topic on this subject with major efforts entirely focused on speeding up rates of numerical convergence. Fast simulations simplify the development of design iterations, but designers are still forced to break the integration process when topological changes need to be introduced for fulfilling a given design intention. As the exploration of elastic structures becomes more integrated within architectural design agendas, there is a growing necessity to introduce more flexible numerical models for design conceptualization. In this paper we introduce the conceptual framework for a generic Particle System model built on principles of combinatorial maps and graph theory that is designed to support topological transformations on the fly. The computational implementation of the proposed model is demonstrated with two types of data structures. The aim of this work is to enhance modeling capabilities when numerical simulations are used as design tools.
•Classification of exploratory form-finding approaches for elastic structures.•Activation of topologic modeling capabilities with real-time physics.•Comprehensive representation of particle system models using combinatorial maps.•Graph theory securing numerical stability when form-finding with dynamic topologies.