Predictability analysis of domestic hot water consumption with neural networks: From single units to large residential buildings

Louis-Gabriel Maltais and Louis Gosselin

Energy, 2021, vol. 229, issue C

Abstract: Domestic hot water is a major energy load in current residential buildings. Improving the energy management and efficiency of water heating systems with smart predictive control requires forecasting the domestic hot water consumption. However, accurate prediction models have proven to be difficult to obtain, which is a major barrier to the applicability of predictive control strategies and their ability to reduce the energy consumption. This work optimizes the input parameters and architecture of neural networks to produce prediction models of the future domestic hot water demand. The methodology is tested on domestic water heating systems of various sizes by using data measured in a 40-unit multifamily residential building. The prediction model provides a good performance for the domestic hot water consumption of the entire building (R2 of 0.88). However, for smaller system sizes, the prediction performance is quite variable. This work proposes graphical tools to evaluate the expected predictability of DHW consumption profiles for systems of different sizes and reveals the close relation between the predictability of a given profile and its coefficient of variation.

Keywords: Buildings; Domestic hot water; Prediction; Machine learning; Neural networks (search for similar items in EconPapers)
Date: 2021
References: View references in EconPapers View complete reference list from CitEc
Citations: View citations in EconPapers (2)

Downloads: (external link)
http://www.sciencedirect.com/science/article/pii/S0360544221009075
Full text for ScienceDirect subscribers only

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:eee:energy:v:229:y:2021:i:c:s0360544221009075

DOI: 10.1016/j.energy.2021.120658

Access Statistics for this article

Energy is currently edited by Henrik Lund and Mark J. Kaiser

More articles in Energy from Elsevier
Bibliographic data for series maintained by Catherine Liu ().