A neural network solution to an architectual design problem: Design of a light shelf

Mohamed Boubekri, Z. Yin, R. Guy

Research output: Contribution to journalArticle

Abstract

The analysis and evaluation of light-shelves design are complex and involve numerous parameters. These features require the development of various methods and tools to help the designer analyse and evaluate the many possible configurations. Conventional statistical tools for generating performance algorithms are not easily adaptable to this problem due to the very large number of independent design variables involved in the equation. In this article Neural Networks (NNs) are introduced as a promising computation method for the analysis and evaluation of daylight illumination due to the light shelf. NNs are shown to have powerful capabilities for generalising a solution by learning a few examples, by associative and adaptive processing, and tolerating the fault inherited from incomplete data.

Original languageEnglish (US)
Pages (from-to)17-21
Number of pages5
JournalArchitectural Science Review
Volume40
Issue number1
DOIs
StatePublished - Mar 1 1997

Fingerprint

Neural networks
Lighting
Processing

Keywords

  • Daylight
  • Light shelves
  • Neural network

ASJC Scopus subject areas

  • Architecture

Cite this

A neural network solution to an architectual design problem : Design of a light shelf. / Boubekri, Mohamed; Yin, Z.; Guy, R.

In: Architectural Science Review, Vol. 40, No. 1, 01.03.1997, p. 17-21.

Research output: Contribution to journalArticle

@article{69539fc742e644d9a404486cb58fe5fe,
title = "A neural network solution to an architectual design problem: Design of a light shelf",
abstract = "The analysis and evaluation of light-shelves design are complex and involve numerous parameters. These features require the development of various methods and tools to help the designer analyse and evaluate the many possible configurations. Conventional statistical tools for generating performance algorithms are not easily adaptable to this problem due to the very large number of independent design variables involved in the equation. In this article Neural Networks (NNs) are introduced as a promising computation method for the analysis and evaluation of daylight illumination due to the light shelf. NNs are shown to have powerful capabilities for generalising a solution by learning a few examples, by associative and adaptive processing, and tolerating the fault inherited from incomplete data.",
keywords = "Daylight, Light shelves, Neural network",
author = "Mohamed Boubekri and Z. Yin and R. Guy",
year = "1997",
month = "3",
day = "1",
doi = "10.1080/00038628.1997.9697373",
language = "English (US)",
volume = "40",
pages = "17--21",
journal = "Architectural Science Review",
issn = "0003-8628",
publisher = "Earthscan",
number = "1",

}

TY - JOUR

T1 - A neural network solution to an architectual design problem

T2 - Design of a light shelf

AU - Boubekri, Mohamed

AU - Yin, Z.

AU - Guy, R.

PY - 1997/3/1

Y1 - 1997/3/1

N2 - The analysis and evaluation of light-shelves design are complex and involve numerous parameters. These features require the development of various methods and tools to help the designer analyse and evaluate the many possible configurations. Conventional statistical tools for generating performance algorithms are not easily adaptable to this problem due to the very large number of independent design variables involved in the equation. In this article Neural Networks (NNs) are introduced as a promising computation method for the analysis and evaluation of daylight illumination due to the light shelf. NNs are shown to have powerful capabilities for generalising a solution by learning a few examples, by associative and adaptive processing, and tolerating the fault inherited from incomplete data.

AB - The analysis and evaluation of light-shelves design are complex and involve numerous parameters. These features require the development of various methods and tools to help the designer analyse and evaluate the many possible configurations. Conventional statistical tools for generating performance algorithms are not easily adaptable to this problem due to the very large number of independent design variables involved in the equation. In this article Neural Networks (NNs) are introduced as a promising computation method for the analysis and evaluation of daylight illumination due to the light shelf. NNs are shown to have powerful capabilities for generalising a solution by learning a few examples, by associative and adaptive processing, and tolerating the fault inherited from incomplete data.

KW - Daylight

KW - Light shelves

KW - Neural network

UR - http://www.scopus.com/inward/record.url?scp=0342920661&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0342920661&partnerID=8YFLogxK

U2 - 10.1080/00038628.1997.9697373

DO - 10.1080/00038628.1997.9697373

M3 - Article

AN - SCOPUS:0342920661

VL - 40

SP - 17

EP - 21

JO - Architectural Science Review

JF - Architectural Science Review

SN - 0003-8628

IS - 1

ER -