Input feature optimization for ANN models predicting daylight in buildings

Lorenz, Clara-Larissa, Packianather, Michael ORCID: https://orcid.org/0000-0002-9436-8206, Bleil De Souza, Clarice ORCID: https://orcid.org/0000-0001-7823-1202, Spaeth, Achim Benjamin ORCID: https://orcid.org/0000-0003-2368-1542 and Lorenz, Tamara Irina 2019. Input feature optimization for ANN models predicting daylight in buildings. Presented at: 26th International Workshop on Intelligent Computing in Engineering, Leuven, Belgium, 30 June - 3 July 2019. Published in: Geyer, Philipp, Allacker, Karen, Schevenels, Mattias, De Troyer, Frank and Pauwels, Pieter eds. EG-ICE 2019. , vol.2394 (34) CEUR,

Preview

PDF - Accepted Post-Print Version Download (1MB) | Preview

Abstract

Artificial Neural Networks (ANNs) were used as prediction models to explore design solutions for the atrium design of a school building. To this end, a solution space of 165 design variants was generated via parametric modeling. This paper details the process of extracting and selecting the input features required for ANN training in order to predict the DA and sDA metric. The feature selection undertaken in this study mainly consisted of two steps: Firstly, a computationally less extensive machine learning model was used to rank the input features according to their relevance in predicting daylight levels. Secondly, ANNs were trained applying sequential forward selection. The proposed method is investigated in terms of achievable improvements to prediction accuracy, reduceable training time and the feasibility of the method.

Item Type:	Conference or Workshop Item - published (Paper)
Date Type:	Published Online
Status:	Published
Schools:	Schools > Engineering Schools > Architecture
Publisher:	CEUR
ISSN:	1613-0073
Date of First Compliant Deposit:	10 June 2019
Date of Acceptance:	15 May 2019
Last Modified:	14 May 2026 14:45
URI:	https://orca.cardiff.ac.uk/id/eprint/123329

Actions (repository staff only)

Edit Item