DONATE

IAHR Document Library


« Back to Library Homepage « Proceedings of the 2nd International Symposium on Water Syst...

Combined Clustering and Prediction of Daily Water and Energy Usage in Multi-Family Residential and Commercial Buildings

Author(s): M. Frankel; L. Xing; C. Chewning; L. Sela

Linked Author(s): Lu Xing

Keywords: Demand modelling and management; Smart urban water management; Water-energy nexus; Clustering

Abstract: As the threat of climate change intensifies and water and energy resources are becoming increasingly scarce, the need for novel demand reduction strategies to conserve water and energy resources has increased. Simultaneously, there has been a rapid increase in the use of meters continuously monitoring for water and energy consumption, collecting data with high temporal resolution that can provide insight into customer water and energy usage patterns. Analysis of the usage data can reveal customer consumption trends and the underlying causes of customer usage, which can be used to plan for future demands, as well as devising demand management strategies to reduce customer demands. Previous works in this topic have focused on extracting water end energy usage patterns [1,2], determining the explanatory variables which influence end user demands [3,4], and modelling future water and energy demands based on historical usage data [5,6]. While many works have focused on demand patterns of single-family residential customers, water and energy patterns of the commercial and multi-family residential sectors remain unexplored, and insights drawn from residential sectors may not apply to other building uses. In addition, while few studies have explored joint water and energy consumption patterns, there has not yet been an analysis of the underlying drivers of observed water and energy consumption patterns. This study seeks to illuminate the differences between water and energy consumption in multi-family residential and non-residential buildings, determine if general building characteristics can adequately segment buildings based on their usage profiles, and investigate the accuracy of a parametric and non-parametric model to predict future consumption.

DOI:

Year: 2021

Copyright © 2022 International Association for Hydro-Environment Engineering and Research. All rights reserved. | Terms and Conditions