Smart meters data for modeling and forecasting water demand at the user-level

Smart meters installed at the user-level provide a new data source for managing water infrastructure. This research explores the use of machine learning methods, including Random Forests (RFs), Artificial Neural Networks (ANNs), and Support Vector Regression (SVR) to forecast hourly water demand at...

Full description

Saved in:
Bibliographic Details
Published in:Environmental modelling & software : with environment data news 2020-03, Vol.125, p.104633, Article 104633
Main Authors: Pesantez, Jorge E., Berglund, Emily Zechman, Kaza, Nikhil
Format: Article
Language:English
Subjects:
AMI
Artificial neural networks
Clustering
Datasets
Demand
Economic forecasting
Forecasting model
Hourly water demand
Learning algorithms
Learning theory
Machine learning
Mathematical models
Modelling
Neural networks
Seasonal variations
Smart water meters
Support vector machines
Time of use
Urban water systems
User-level data
Water demand
Water demand management
Water management
Water supply systems
Weather
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Smart meters installed at the user-level provide a new data source for managing water infrastructure. This research explores the use of machine learning methods, including Random Forests (RFs), Artificial Neural Networks (ANNs), and Support Vector Regression (SVR) to forecast hourly water demand at 90 accounts using smart-metered data. Demands are predicted using lagged demand, seasonality, weather, and household characteristics. Time-series clustering is applied to delineate data based on the time of day and day of the week, which improves model performance. Two modeling approaches are compared. Individual models are developed separately for each meter, and a Group model is trained using a data set of multiple meters. Individual models predict demands at meters in the original data set with lower error than Group models, while the Group model predicts demands at new meters with lower error than Individual models. Results demonstrate that RF and ANN perform better than SVR across all scenarios. •A forecasting model is developed using smart water meter data at the user-level.•Predictors include hourly past demand, weather data, and household characteristics.•The approach uses clustering and machine learning methods.•Clustering and exogenous predictors improve model performance.
ISSN:1364-8152
1873-6726
DOI:10.1016/j.envsoft.2020.104633