Full Article - Open Access.

Idioma principal | Segundo idioma

Clustering metamodel for predictive performance for dynamic shading facade

Clustering metamodel for predictive performance for dynamic shading facade

Houang Daher, Cassio ; Coeli Ruschel, Regina ;

Full Article:

Dynamic shading facades present a challenge for the computational simulation of illuminance performance in the project's design phase. Such architectural elements have conflicting functions: shading without blocking natural light. The evaluation of these dynamic elements depends on multiple parameters and combinations, which results in higher complexity in compositions and difficulty in understanding. The objective is to identify the optimized positioning of dynamic elements of the facade; still in the early stages of the project, to have good illuminance performance for all busy hours of the year. The research conducted here is experimental, using computer simulation. Our model considers as the dependent variable the average annual illuminance. The independent variables analyzed are shoebox orientation, positioning of the fins, date, and time. The contribution of this research is to test the set of results of the independent variables by training an algorithm capable of replacing the simulation.

Full Article:

Dynamic shading facades present a challenge for the computational simulation of illuminance performance in the project's design phase. Such architectural elements have conflicting functions: shading without blocking natural light. The evaluation of these dynamic elements depends on multiple parameters and combinations, which results in higher complexity in compositions and difficulty in understanding. The objective is to identify the optimized positioning of dynamic elements of the facade; still in the early stages of the project, to have good illuminance performance for all busy hours of the year. The research conducted here is experimental, using computer simulation. Our model considers as the dependent variable the average annual illuminance. The independent variables analyzed are shoebox orientation, positioning of the fins, date, and time. The contribution of this research is to test the set of results of the independent variables by training an algorithm capable of replacing the simulation.

Palavras-chave: Machine Learning, Metamodel, Dynamic Facade, Performance Evaluation, Clustering,

Palavras-chave: Machine Learning, Metamodel, Dynamic Facade, Performance Evaluation, Clustering,

DOI: 10.5151/sigradi2023-196

Referências bibliográficas
  • [1] -
Como citar:

Houang Daher, Cassio; Coeli Ruschel, Regina; "Clustering metamodel for predictive performance for dynamic shading facade", p. 446-458 . In: . São Paulo: Blucher, 2024.
ISSN 2318-6968, DOI 10.5151/sigradi2023-196

últimos 30 dias | último ano | desde a publicação


downloads


visualizações


indexações