Disaggregation and classification of residential water events from high-resolution smart water meter data using unsupervised machine learning methods

The residential sector accounts for a significant amount of water consumption in the United States. Understanding this water consumption behavior provides opportunity for water savings, which are important for sustaining freshwater resources. This study analyzed 1-second resolution smart water meter data from a 4-person household over the course of one year. The smart meter data were disaggregated using derivative signals of the influent water flow rate at the water main entrance to the home to identify start and end times of water events. k-means clustering, an unsupervised machine learning method, then categorized these water events based on information collected from the appliance end-uses. The use of unsupervised learning substantially reduces the training data requirements and lowers the barrier of implementation for the model. Peak demand times for each day were determined and water use profiles were analyzed to identify seasonal, weekly, and daily trends. These results provide insight into opportunities to reduce water consumption within the household, including the reduction of water consumption during peak demand hours. The widespread implementation of this type of smart water metering and disaggregation system could provide opportunity to improve water conservation and efficiency on a larger scale and reduce stress on local infrastructure systems and water resources.

Name thesis:degree_name

M.S.

Level thesis:degree_level

Thesis

Discipline thesis:degree_discipline

Civil Engineering

Grantor

University of Illinois at Urbana-Champaign

Year dc:date

2020

Author dc:creator

• Bethke, Gabrielle M

Contributors dc:contributor

• Stillwell, Ashlynn S

• smart water metering
• machine learning
• water consumption
• residential water
• data disaggregation
• appliance end-uses

Statement dc:rights

• Copyright 2020 Gabrielle M. Bethke

Language dc:language

en

Handle dc:identifier

http://hdl.handle.net/2142/108047

OAI identifier oai:identifier

oai:www.ideals.illinois.edu:2142/108047