Cardiff University | Prifysgol Caerdydd ORCA
Online Research @ Cardiff 
WelshClear Cookie - decide language by browser settings

BIM-based machine learning engine for building energy prediction

Bie, Sisi and Li, Haijiang 2019. BIM-based machine learning engine for building energy prediction. Presented at: 36th CIB W78 2019 Conference, Newcastle, England, UK, 18-20 September 2019.
Item availability restricted.

[thumbnail of w78-2019-paper-029.pdf] PDF - Published Version
Restricted to Repository staff only

Download (910kB)


Buildings consume a large amount of energy; it is therefore important to find a better way to manage building energy performance more effectively. The recent developments in data science and machine learning show a good potential to leverage the different sources of data to predict energy consumption. This paper reviews the current research in these areas and proposes a BIM (Building Information Modelling) based approach to provide data and information to train the relevant machine learning (ML) engine. The method has been demonstrated with a daylight illuminance prediction analysis. The presented preliminary result can potentially lead to a more generic tool, where rich data and information embedded in BIM models can be utilized for better decision making.

Item Type: Conference or Workshop Item (Paper)
Status: In Press
Schools: Engineering
Date of First Compliant Deposit: 6 April 2020
Date of Acceptance: 18 September 2019
Last Modified: 07 Apr 2020 12:00

Actions (repository staff only)

Edit Item Edit Item


Downloads per month over past year

View more statistics