User centered neuro-fuzzy energy management through semantic-based optimization

Howell, Shaun K., Wicaksono, Hendro, Yuce, Baris ORCID: https://orcid.org/0000-0002-9937-1535, McGlinn, Kris and Rezgui, Yacine ORCID: https://orcid.org/0000-0002-5711-8400 2019. User centered neuro-fuzzy energy management through semantic-based optimization. IEEE Transactions on Cybernetics 49 (9) , pp. 3278-3292. 10.1109/TCYB.2018.2839700

Preview

PDF - Published Version Available under License Creative Commons Attribution. Download (3MB) | Preview

Abstract

This paper presents a cloud-based building energy management system, underpinned by semantic middleware, that integrates an enhanced sensor network with advanced analytics, accessible through an intuitive Web-based user interface. The proposed solution is described in terms of its three key layers: 1) user interface; 2) intelligence; and 3) interoperability. The system’s intelligence is derived from simulation-based optimized rules, historical sensor data mining, and a fuzzy reasoner. The solution enables interoperability through a semantic knowledge base, which also contributes intelligence through reasoning and inference abilities, and which are enhanced through intelligent rules. Finally, building energy performance monitoring is delivered alongside optimized rule suggestions and a negotiation process in a 3-D Web-based interface using WebGL. The solution has been validated in a real pilot building to illustrate the strength of the approach, where it has shown over 25% energy savings. The relevance of this paper in the field is discussed, and it is argued that the proposed solution is mature enough for testing across further buildings.

Item Type:	Article
Date Type:	Publication
Status:	Published
Schools:	Engineering
Additional Information:	This is an open access article under the terms of the CC-BY Attribution 4.0 International license.
Publisher:	IEEE
ISSN:	2168-2267
Date of First Compliant Deposit:	19 July 2018
Date of Acceptance:	18 May 2018
Last Modified:	06 May 2023 22:54
URI:	https://orca.cardiff.ac.uk/id/eprint/113267

Citation Data

Cited 14 times in Scopus. View in Scopus. Powered By Scopus® Data

Actions (repository staff only)

Edit Item