Virtual sensors to improve on-line hydraulic model calibration

Daniel Goldsmith, Ami Preis, Michael Allen, Andrew J. Whittle

Research output: Chapter in Book/Report/Conference proceedingConference proceeding

3 Citations (Scopus)

Abstract

A new approach for monitoring a water distribution system involving Virtual Sensors is presented. In this approach, wireless sensor nodes are permanently deployed within the distribution system, providing continuous, on-line hydraulic data that can be assimilated into hydraulic models. In addition, temporary nodes are deployed for short periods (one week) around the distribution network. A Virtual Sensor is implemented using a data imputation technique called Gaussian Process Regression, which combines the historical data collected by the temporary node with correlated data from a subset of permanent sensor nodes. Use of spatially-correlated data accounts for new trends in the data that do not appear in the historical data collected by the temporary node. An increase in the number of sensors (a combination of real and virtual) is important for reducing the ill-conditioned state of the hydraulic model calibration procedure. The technique is demonstrated as a proof-of-concept using data collected from the WaterWiSe@SG testbed in Singapore, and is shown to predict pressure data trends with an accuracy of 0.76 PSI RMSE after a six-week test.

Original languageEnglish
Title of host publicationWater Distribution Systems Analysis 2010
Subtitle of host publicationProceedings of the 12th International Conference, WDSA 2010
EditorsKevin E. Lansey, Christopher Y. Choi, Avi Ostfeld, Ian L. Pepper
PublisherASCE
Pages1349-1361
Number of pages13
ISBN (Print)9780784412039
DOIs
Publication statusPublished - 2012
Event12th Annual International Conference on Water Distribution Systems Analysis 2010 - Tucson, AZ, United States
Duration: 12 Sep 201015 Sep 2010

Conference

Conference12th Annual International Conference on Water Distribution Systems Analysis 2010
Abbreviated titleWDSA 2010
CountryUnited States
CityTucson, AZ
Period12/09/1015/09/10

Fingerprint

sensor
calibration
hydraulics
distribution system
monitoring

Keywords

  • Data imputation
  • Gaussian Process Regression
  • hydraulic modeling
  • on-line operation

ASJC Scopus subject areas

  • Water Science and Technology

Cite this

Goldsmith, D., Preis, A., Allen, M., & Whittle, A. J. (2012). Virtual sensors to improve on-line hydraulic model calibration. In K. E. Lansey, C. Y. Choi, A. Ostfeld, & I. L. Pepper (Eds.), Water Distribution Systems Analysis 2010 : Proceedings of the 12th International Conference, WDSA 2010 (pp. 1349-1361). ASCE. https://doi.org/10.1061/41203(425)121

Virtual sensors to improve on-line hydraulic model calibration. / Goldsmith, Daniel; Preis, Ami; Allen, Michael; Whittle, Andrew J.

Water Distribution Systems Analysis 2010 : Proceedings of the 12th International Conference, WDSA 2010. ed. / Kevin E. Lansey; Christopher Y. Choi; Avi Ostfeld; Ian L. Pepper. ASCE, 2012. p. 1349-1361.

Research output: Chapter in Book/Report/Conference proceedingConference proceeding

Goldsmith, D, Preis, A, Allen, M & Whittle, AJ 2012, Virtual sensors to improve on-line hydraulic model calibration. in KE Lansey, CY Choi, A Ostfeld & IL Pepper (eds), Water Distribution Systems Analysis 2010 : Proceedings of the 12th International Conference, WDSA 2010. ASCE, pp. 1349-1361, 12th Annual International Conference on Water Distribution Systems Analysis 2010, Tucson, AZ, United States, 12/09/10. https://doi.org/10.1061/41203(425)121
Goldsmith D, Preis A, Allen M, Whittle AJ. Virtual sensors to improve on-line hydraulic model calibration. In Lansey KE, Choi CY, Ostfeld A, Pepper IL, editors, Water Distribution Systems Analysis 2010 : Proceedings of the 12th International Conference, WDSA 2010. ASCE. 2012. p. 1349-1361 https://doi.org/10.1061/41203(425)121
Goldsmith, Daniel ; Preis, Ami ; Allen, Michael ; Whittle, Andrew J. / Virtual sensors to improve on-line hydraulic model calibration. Water Distribution Systems Analysis 2010 : Proceedings of the 12th International Conference, WDSA 2010. editor / Kevin E. Lansey ; Christopher Y. Choi ; Avi Ostfeld ; Ian L. Pepper. ASCE, 2012. pp. 1349-1361
@inproceedings{29974769afa5455ca95e5df3543a6c4f,
title = "Virtual sensors to improve on-line hydraulic model calibration",
abstract = "A new approach for monitoring a water distribution system involving Virtual Sensors is presented. In this approach, wireless sensor nodes are permanently deployed within the distribution system, providing continuous, on-line hydraulic data that can be assimilated into hydraulic models. In addition, temporary nodes are deployed for short periods (one week) around the distribution network. A Virtual Sensor is implemented using a data imputation technique called Gaussian Process Regression, which combines the historical data collected by the temporary node with correlated data from a subset of permanent sensor nodes. Use of spatially-correlated data accounts for new trends in the data that do not appear in the historical data collected by the temporary node. An increase in the number of sensors (a combination of real and virtual) is important for reducing the ill-conditioned state of the hydraulic model calibration procedure. The technique is demonstrated as a proof-of-concept using data collected from the WaterWiSe@SG testbed in Singapore, and is shown to predict pressure data trends with an accuracy of 0.76 PSI RMSE after a six-week test.",
keywords = "Data imputation, Gaussian Process Regression, hydraulic modeling, on-line operation",
author = "Daniel Goldsmith and Ami Preis and Michael Allen and Whittle, {Andrew J.}",
year = "2012",
doi = "10.1061/41203(425)121",
language = "English",
isbn = "9780784412039",
pages = "1349--1361",
editor = "Lansey, {Kevin E.} and Choi, {Christopher Y.} and Avi Ostfeld and Pepper, {Ian L.}",
booktitle = "Water Distribution Systems Analysis 2010",
publisher = "ASCE",

}

TY - GEN

T1 - Virtual sensors to improve on-line hydraulic model calibration

AU - Goldsmith, Daniel

AU - Preis, Ami

AU - Allen, Michael

AU - Whittle, Andrew J.

PY - 2012

Y1 - 2012

N2 - A new approach for monitoring a water distribution system involving Virtual Sensors is presented. In this approach, wireless sensor nodes are permanently deployed within the distribution system, providing continuous, on-line hydraulic data that can be assimilated into hydraulic models. In addition, temporary nodes are deployed for short periods (one week) around the distribution network. A Virtual Sensor is implemented using a data imputation technique called Gaussian Process Regression, which combines the historical data collected by the temporary node with correlated data from a subset of permanent sensor nodes. Use of spatially-correlated data accounts for new trends in the data that do not appear in the historical data collected by the temporary node. An increase in the number of sensors (a combination of real and virtual) is important for reducing the ill-conditioned state of the hydraulic model calibration procedure. The technique is demonstrated as a proof-of-concept using data collected from the WaterWiSe@SG testbed in Singapore, and is shown to predict pressure data trends with an accuracy of 0.76 PSI RMSE after a six-week test.

AB - A new approach for monitoring a water distribution system involving Virtual Sensors is presented. In this approach, wireless sensor nodes are permanently deployed within the distribution system, providing continuous, on-line hydraulic data that can be assimilated into hydraulic models. In addition, temporary nodes are deployed for short periods (one week) around the distribution network. A Virtual Sensor is implemented using a data imputation technique called Gaussian Process Regression, which combines the historical data collected by the temporary node with correlated data from a subset of permanent sensor nodes. Use of spatially-correlated data accounts for new trends in the data that do not appear in the historical data collected by the temporary node. An increase in the number of sensors (a combination of real and virtual) is important for reducing the ill-conditioned state of the hydraulic model calibration procedure. The technique is demonstrated as a proof-of-concept using data collected from the WaterWiSe@SG testbed in Singapore, and is shown to predict pressure data trends with an accuracy of 0.76 PSI RMSE after a six-week test.

KW - Data imputation

KW - Gaussian Process Regression

KW - hydraulic modeling

KW - on-line operation

U2 - 10.1061/41203(425)121

DO - 10.1061/41203(425)121

M3 - Conference proceeding

SN - 9780784412039

SP - 1349

EP - 1361

BT - Water Distribution Systems Analysis 2010

A2 - Lansey, Kevin E.

A2 - Choi, Christopher Y.

A2 - Ostfeld, Avi

A2 - Pepper, Ian L.

PB - ASCE

ER -