Computational workflow for segmented shell structures: An ANN approach for fabrication efficiency

Effimia Giannopoulou*, Pablo Baquero, Angad Warang, Alberto T. Estévez

*Corresponding author for this work

Research output: Chapter in Book or Report/Conference proceedingConference Proceedingpeer-review

Abstract

An already established generative design methodology for bifurcating thin shell structures with similar physical and geometrical principles of biological processes, is using the logic of stripes as a construction system, linking relaxation, segmentation and fabrication processes as one unified system in equilibrium. However, a discussion is been raised upon the amount of stripes and their topology in relation with the branching system, structural performance, connectivity system, and material usage. The objective of this article is to explore the potentials and feasibility of a state-of-the-art machine learning approach which is based partly on the user's experience and knowledge. The method consists of a generation of a database of branching topologies which are utilized to train an Artificial Neural Network. This will allow the user to have a visual judgement of the numerical values based on multifunctional criteria and adjust the design based on the new information. The fact that the prediction mechanism preserves the intuitiveness of a generic design process enhances the designer's inventory by providing control over the scalability, is saving a lot of computational time and consequently helps in the design to fabrication process.

Original languageEnglish
Title of host publicationIASS Symposium 2019 - 60th Anniversary Symposium of the International Association for Shell and Spatial Structures; Structural Membranes 2019 - 9th International Conference on Textile Composites and Inflatable Structures, FORM and FORCE
EditorsCarlos Lazaro, Kai-Uwe Bletzinger, Eugenio Onate
PublisherInternational Center for Numerical Methods in Engineering
Pages2653-2660
Number of pages8
ISBN (Electronic)9788412110104
Publication statusPublished - 2019
Externally publishedYes
EventIASS Symposium 2019 - 60th Anniversary Symposium of the International Association for Shell and Spatial Structures; Structural Membranes 2019 - 9th International Conference on Textile Composites and Inflatable Structures, FORM and FORCE - Barcelona, Spain
Duration: 7 Oct 201910 Oct 2019

Publication series

NameIASS Symposium 2019 - 60th Anniversary Symposium of the International Association for Shell and Spatial Structures; Structural Membranes 2019 - 9th International Conference on Textile Composites and Inflatable Structures, FORM and FORCE

Conference

ConferenceIASS Symposium 2019 - 60th Anniversary Symposium of the International Association for Shell and Spatial Structures; Structural Membranes 2019 - 9th International Conference on Textile Composites and Inflatable Structures, FORM and FORCE
Country/TerritorySpain
CityBarcelona
Period7/10/1910/10/19

Keywords

  • Bio-inspired
  • Digital Fabrication
  • Machine Learning
  • Mesh Segmentation
  • Shell Structures

Fingerprint

Dive into the research topics of 'Computational workflow for segmented shell structures: An ANN approach for fabrication efficiency'. Together they form a unique fingerprint.

Cite this