Diagnostic tools of energy performance for supermarkets using Artificial Neural Network algorithms

File Description SizeFormat 
Manuscript_for_archiving.pdfAccepted version992.81 kBAdobe PDFView/Open
Title: Diagnostic tools of energy performance for supermarkets using Artificial Neural Network algorithms
Authors: Mavromatidis, G
Acha, S
Shah, N
Item Type: Journal Article
Abstract: Supermarket performance monitoring is of vital importance to ensure systems perform adequately and guarantee operating costs and energy use are kept at a minimum. Furthermore, advanced monitoring techniques can allow early detection of equipment faults that could disrupt store operation. This paper details the development of a tool for performance monitoring and fault detection for supermarkets focusing on evaluating the Store's Total Electricity Consumption as well as individual systems, such as Refrigeration, HVAC, Lighting and Boiler. Artificial Neural Network (ANN) models are developed for each system to provide the energy baseline, which is modelled as a dependency between the energy consumption and suitable explanatory variables. The tool has two diagnostic levels. The first level broadly evaluates the systems performance, in terms of energy consumption, while the second level applies more rigorous criteria for fault detection of supermarket subsystems. A case study, using data from a store in Southeast England, is presented and results show remarkable accuracy for calculating hourly energy use, thus marking the ANN method as a viable tool for diagnosis purposes. Finally, the generic nature of the methodology approach allows the development and application to other stores, effectively offering a valuable analytical tool for better running of supermarkets.
Issue Date: 21-Mar-2013
Date of Acceptance: 9-Mar-2013
URI: http://hdl.handle.net/10044/1/38980
DOI: http://dx.doi.org/10.1016/j.enbuild.2013.03.020
ISSN: 1872-6178
Publisher: Elsevier
Start Page: 304
End Page: 314
Journal / Book Title: Energy and Buildings
Volume: 62
Copyright Statement: © 2013 Elsevier. Licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International http://creativecommons.org/licenses/by-nc-nd/4.0/
Sponsor/Funder: Sainsbury's Supermarkets Ltd
Funder's Grant Number: n/a
Keywords: Science & Technology
Construction & Building Technology
Energy & Fuels
Engineering, Civil
Supermarket energy use
Energy forecasting
Fault detection
Predicative maintenance
Artificial neural networks
Building & Construction
09 Engineering
12 Built Environment And Design
Publication Status: Published
Appears in Collections:Faculty of Engineering
Centre for Environmental Policy
Chemical Engineering
Faculty of Natural Sciences

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.

Creative Commons