Classifying Household Water Use Events into Indoor and Outdoor Use: Improving the Benefits of Basic Smart Meter Data Sets

Meyer, Bettina Elizabeth; Nguyen, Khoi; Beal, Cara; Jacobs, Heinz Erasmus; Buchberger, Steven G.

doi:10.1061/(asce)wr.1943-5452.0001471

Cited by 7 publications

(2 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Studies on public supply water withdrawal have been applied across different spatial and temporal scales, such as estimating water withdrawal annually for counties or even for individual households and communities (Aznar‐Sánchez et al., 2018; Dieter et al., 2018; Gato et al., 2007; Herrera et al., 2010; Maidment et al., 1985; McManamay et al., 2021; Meyer et al., 2021; Morales et al., 2013; Quesnel et al., 2020; Sankarasubramanian et al., 2017). Recent studies have increasingly adopted ML methods due to their ability to capture complex nonlinear relationships between dependent and independent variables (Wongso et al., 2020).…”

Section: Introductionmentioning

confidence: 99%

Next Generation Public Supply Water Withdrawal Estimation for the Conterminous United States Using Machine Learning and Operational Frameworks

Alzraiee,

Niswonger,

Luukkonen

et al. 2024

Water Resources Research

View full text Add to dashboard Cite

Estimation of human water withdrawals is more important now than ever due to uncertain water supplies, population growth, and climate change. Fourteen percent of the total water withdrawal in the United States is used for public supply, typically including deliveries to domestic, commercial, and occasionally including industrial, irrigation, and thermoelectric water withdrawal. Stewards of water resources in the USA require estimates of water withdrawals to manage and plan for future demands and sustainable water supplies. This study compiled the most comprehensive conterminous United States water withdrawal data set to date and developed a machine learning framework for estimating public supply withdrawals and associated uncertainty for the period 2000–2020. The modeling approach provides service area resolution estimates to allow for annual and monthly water withdrawal estimation while incorporating a complex array of driving factors that include hydroclimatic, demographic, socioeconomic, geographic, and land use factors. Model results reveal highly variable and lognormally distributed per‐capita water withdrawal, spanning from 30 to 650 gallons per capita per day (GPCD), across community, regional, and national scales, with pronounced seasonal variations. Analysis of estimated withdrawal trends indicates that the national annual average withdrawal experienced a decline at a rate of 0.58 GPCD/year during the period from 2000 to 2020. Model interpretation reveals a complex interplay between public supply withdrawal and key predictors, including population size, warm‐season precipitation, counts of large buildings and houses, and areas of urban and commercial land use. The developed models can forecast future public supply driven by various climate, demographic, and socioeconomic scenarios.

show abstract

Section: Introductionmentioning

confidence: 99%

Next Generation Public Supply Water Withdrawal Estimation for the Conterminous United States Using Machine Learning and Operational Frameworks

Alzraiee,

Niswonger,

Luukkonen

et al. 2024

Water Resources Research

View full text Add to dashboard Cite

show abstract

“…The authors in [15] proposed the use of an adaptable neuro-fuzzy network to classify water end-uses achieving high accuracy, using a limited dataset of flow measurements. In more recent studies, machine learning and data analytic algorithms were developed to address the problem of water end-use disaggregation, with promising results [16][17][18][19][20][21]. Several drawbacks that were noted in these studies include the need for a large amount of historical data to train the model and the absence of disaggregation techniques for combined water events.…”

Section: Introductionmentioning

confidence: 99%

Non-intrusive water usage classification considering limited training data

Pavlou,

Vrachimis,

Eliades

et al. 2022

Proceedings - 2nd International Join Conference on Water Distribution System Analysis (WDSA)& Computing and Control in the

View full text Add to dashboard Cite

Smart metering of domestic water consumption to continuously monitor the usage of different appliances has been shown to have an impact on people’s behavior towards water conservation. However, the installation of multiple sensors to monitor each appliance currently has a high initial cost and as a result, monitoring consumption from different appliances using sensors is not cost-effective. To address this challenge, studies have focused on analyzing measurements of the total domestic consumption using Machine Learning (ML) methods, to disaggregate water usage into each appliance. Identifying which appliances are in use through ML is challenging since their operation may be overlapping, while specific appliances may operate with intermittent flow, making individual consumption events hard to distinguish. Moreover, ML approaches require large amounts of labeled input data to train their models, which are typically not available for a single household, while usage characteristics may vary in different regions. In this work, we initially propose a data model that generates synthetic time series based on regional water usage characteristics and resolution to overcome the need for a large training dataset with real labeled data. The method requires a small number of real labeled data from the studied region. Following this, we propose a new algorithm for classifying single and overlapping household water usage events, using the total domestic consumption measurements. The classification procedure is described below: 1) During the offline feature learning stage, a dataset of labeled data corresponding to water-use profile signals is analyzed to some predefined features, such as event volume, event duration, event flow peak, and event signature, to extract its statistical properties, 2) The event classification stage monitors the provided measurement time-series for events between zero-flow intervals. The identified events are then classified using Dynamic Time Wrapping and an optimization procedure that finds the best label for the observed event based on the features learned in the first stage and similarity indices. Non-classified events are processed using a variation vector technique to identify the combined events which are then split into sub-single events and classified.Extended version of this article is selected for possible publication in a Special Issue in the Journal of Hydroinformatics.

show abstract

Pressure Reduction for Water Loss Management: A Statistical Evaluation of Water Consumption Considering Customer Storage Tanks

Sodek

Martim

Filho

et al. 2022

J. Water Resour. Plann. Manage.

View full text Add to dashboard Cite

Classifying Household Water Use Events into Indoor and Outdoor Use: Improving the Benefits of Basic Smart Meter Data Sets

Cited by 7 publications

References 27 publications

Next Generation Public Supply Water Withdrawal Estimation for the Conterminous United States Using Machine Learning and Operational Frameworks

Next Generation Public Supply Water Withdrawal Estimation for the Conterminous United States Using Machine Learning and Operational Frameworks

Non-intrusive water usage classification considering limited training data

Pressure Reduction for Water Loss Management: A Statistical Evaluation of Water Consumption Considering Customer Storage Tanks

Contact Info

Product

Resources

About