Approximate methods for analyzing water quality prediction uncertainty in water distribution systems

D. S. Kang, M. F K Pasha, Kevin E Lansey

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

5 Citations (Scopus)

Abstract

To assess the uncertainty of model predictions, Monte Carlo simulation (MCS) is commonly applied. However, when modeling water distribution system quality under unsteady conditions, MCS computation times can be excessive even for a reasonably sized system. The aim of this study is to evaluate alternative estimation schemes and examine their ability to predict model prediction uncertainty with less computational effort. Here, MCS results are compared with a point estimation method, the first order second moment (FOSM) method, and a quasi-MCS method, Latin hypercube sampling (LHS). Hydraulic and water quality simulations are performed using EPANET for a typical pressure zone sized system with 116 pipes and 90 nodes. The primary model outputs of interest are chlorine concentrations and water age. Nodal pressures are also evaluated. Preliminary analysis showed that these outputs are most sensitive to nodal demands of all system parameters thus only demand uncertainty results are presented. Results demonstrate that LHS provides very good estimates of the predicted means and variances for steady and unsteady conditions compared with MCS while FOSM did well for steady conditions but did poorly for some periods in the extended period simulation.

Original languageEnglish (US)
Title of host publicationRestoring Our Natural Habitat - Proceedings of the 2007 World Environmental and Water Resources Congress
StatePublished - 2007
Event2007 World Environmental and Water Resources Congress: Restoring Our Natural Habitat - Tampa, FL, United States
Duration: May 15 2007May 19 2007

Other

Other2007 World Environmental and Water Resources Congress: Restoring Our Natural Habitat
CountryUnited States
CityTampa, FL
Period5/15/075/19/07

Fingerprint

Water distribution systems
Water quality
water quality
prediction
simulation
Sampling
Method of moments
Chlorine
sampling
Pipe
estimation method
Hydraulics
Monte Carlo simulation
Uncertainty
water distribution system
method
chlorine
pipe
hydraulics
Water

ASJC Scopus subject areas

  • Environmental Engineering
  • Water Science and Technology

Cite this

Kang, D. S., Pasha, M. F. K., & Lansey, K. E. (2007). Approximate methods for analyzing water quality prediction uncertainty in water distribution systems. In Restoring Our Natural Habitat - Proceedings of the 2007 World Environmental and Water Resources Congress

Approximate methods for analyzing water quality prediction uncertainty in water distribution systems. / Kang, D. S.; Pasha, M. F K; Lansey, Kevin E.

Restoring Our Natural Habitat - Proceedings of the 2007 World Environmental and Water Resources Congress. 2007.

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

Kang, DS, Pasha, MFK & Lansey, KE 2007, Approximate methods for analyzing water quality prediction uncertainty in water distribution systems. in Restoring Our Natural Habitat - Proceedings of the 2007 World Environmental and Water Resources Congress. 2007 World Environmental and Water Resources Congress: Restoring Our Natural Habitat, Tampa, FL, United States, 5/15/07.
Kang DS, Pasha MFK, Lansey KE. Approximate methods for analyzing water quality prediction uncertainty in water distribution systems. In Restoring Our Natural Habitat - Proceedings of the 2007 World Environmental and Water Resources Congress. 2007
Kang, D. S. ; Pasha, M. F K ; Lansey, Kevin E. / Approximate methods for analyzing water quality prediction uncertainty in water distribution systems. Restoring Our Natural Habitat - Proceedings of the 2007 World Environmental and Water Resources Congress. 2007.
@inproceedings{67b34661f4544a66ac4e127ba52b63ae,
title = "Approximate methods for analyzing water quality prediction uncertainty in water distribution systems",
abstract = "To assess the uncertainty of model predictions, Monte Carlo simulation (MCS) is commonly applied. However, when modeling water distribution system quality under unsteady conditions, MCS computation times can be excessive even for a reasonably sized system. The aim of this study is to evaluate alternative estimation schemes and examine their ability to predict model prediction uncertainty with less computational effort. Here, MCS results are compared with a point estimation method, the first order second moment (FOSM) method, and a quasi-MCS method, Latin hypercube sampling (LHS). Hydraulic and water quality simulations are performed using EPANET for a typical pressure zone sized system with 116 pipes and 90 nodes. The primary model outputs of interest are chlorine concentrations and water age. Nodal pressures are also evaluated. Preliminary analysis showed that these outputs are most sensitive to nodal demands of all system parameters thus only demand uncertainty results are presented. Results demonstrate that LHS provides very good estimates of the predicted means and variances for steady and unsteady conditions compared with MCS while FOSM did well for steady conditions but did poorly for some periods in the extended period simulation.",
author = "Kang, {D. S.} and Pasha, {M. F K} and Lansey, {Kevin E}",
year = "2007",
language = "English (US)",
isbn = "9780784409275",
booktitle = "Restoring Our Natural Habitat - Proceedings of the 2007 World Environmental and Water Resources Congress",

}

TY - GEN

T1 - Approximate methods for analyzing water quality prediction uncertainty in water distribution systems

AU - Kang, D. S.

AU - Pasha, M. F K

AU - Lansey, Kevin E

PY - 2007

Y1 - 2007

N2 - To assess the uncertainty of model predictions, Monte Carlo simulation (MCS) is commonly applied. However, when modeling water distribution system quality under unsteady conditions, MCS computation times can be excessive even for a reasonably sized system. The aim of this study is to evaluate alternative estimation schemes and examine their ability to predict model prediction uncertainty with less computational effort. Here, MCS results are compared with a point estimation method, the first order second moment (FOSM) method, and a quasi-MCS method, Latin hypercube sampling (LHS). Hydraulic and water quality simulations are performed using EPANET for a typical pressure zone sized system with 116 pipes and 90 nodes. The primary model outputs of interest are chlorine concentrations and water age. Nodal pressures are also evaluated. Preliminary analysis showed that these outputs are most sensitive to nodal demands of all system parameters thus only demand uncertainty results are presented. Results demonstrate that LHS provides very good estimates of the predicted means and variances for steady and unsteady conditions compared with MCS while FOSM did well for steady conditions but did poorly for some periods in the extended period simulation.

AB - To assess the uncertainty of model predictions, Monte Carlo simulation (MCS) is commonly applied. However, when modeling water distribution system quality under unsteady conditions, MCS computation times can be excessive even for a reasonably sized system. The aim of this study is to evaluate alternative estimation schemes and examine their ability to predict model prediction uncertainty with less computational effort. Here, MCS results are compared with a point estimation method, the first order second moment (FOSM) method, and a quasi-MCS method, Latin hypercube sampling (LHS). Hydraulic and water quality simulations are performed using EPANET for a typical pressure zone sized system with 116 pipes and 90 nodes. The primary model outputs of interest are chlorine concentrations and water age. Nodal pressures are also evaluated. Preliminary analysis showed that these outputs are most sensitive to nodal demands of all system parameters thus only demand uncertainty results are presented. Results demonstrate that LHS provides very good estimates of the predicted means and variances for steady and unsteady conditions compared with MCS while FOSM did well for steady conditions but did poorly for some periods in the extended period simulation.

UR - http://www.scopus.com/inward/record.url?scp=80051616700&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=80051616700&partnerID=8YFLogxK

M3 - Conference contribution

SN - 9780784409275

BT - Restoring Our Natural Habitat - Proceedings of the 2007 World Environmental and Water Resources Congress

ER -