Ant colony algorithm for building energy optimisation problems and comparison with benchmark algorithms

Bamdad, Keivan and Cholette, Michael E and Guan, Lisa and Bell, John (2017) Ant colony algorithm for building energy optimisation problems and comparison with benchmark algorithms. Energy and Buildings, 154. pp. 404-414. ISSN 0378-7788


Abstract

In the design of low-energy buildings, mathematical optimisation has proven to be a powerful tool for minimising energy consumption. Simulation-based optimisation methods are widely employed due to the nonlinear thermal behaviour of buildings. However, finding high-quality solutions with reasonable computational cost remains a significant challenge in the building industry.

In this paper, Ant Colony Optimisation for continuous domain (ACOR) is developed and applied to optimise a commercial building in Australia. The results for a typical commercial building showed that optimisation can achieve an additional energy savings of more than 11.4%, even after some common energy saving measures were implemented (e.g. double pane windows). The performance of ACOR was compared to three benchmark optimisation algorithms: Nelder-Mead (NM) algorithm, Particle Swarm Optimisation with Inertia Weight (PSOIW) and the hybrid Particle Swarm Optimisation and Hooke-Jeeves (PSO-HJ). This comparison showed that ACOR was able to consistently find better solutions in less time than the benchmark algorithms. The findings demonstrate that ACOR can further facilitate the design of low-energy buildings.


Statistics for USQ ePrint 39349
Statistics for this ePrint Item
Item Type: Article (Commonwealth Reporting Category C)
Refereed: Yes
Item Status: Live Archive
Additional Information: Files associated with this item cannot be displayed due to copyright restrictions.
Faculty/School / Institute/Centre: Current - Research and Innovation Division (12 Jul 2012 -)
Faculty/School / Institute/Centre: Current - Research and Innovation Division (12 Jul 2012 -)
Date Deposited: 04 Sep 2020 01:34
Last Modified: 11 Sep 2020 02:29
Uncontrolled Keywords: Optimisation algorithm benchmarking, Building optimisation, Ant colony optimisation, Particle swarm optimisation, Australian commercial building
Fields of Research (2008): 09 Engineering > 0999 Other Engineering > 099999 Engineering not elsewhere classified
Identification Number or DOI: dpi:10.1016/j.enbuild.2017.08.071
URI: http://eprints.usq.edu.au/id/eprint/39349

Actions (login required)

View Item Archive Repository Staff Only