Reducing online contaminant monitoring uncertainty using a bayesian belief network

W. J. Dawsey, B. S. Minsker, V. L. VanBlaricum

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

Abstract

There is a great deal of uncertainty in real time characterization of water distribution system contamination events. Much of this uncertainty is due to the lack of targeted sensors which makes it necessary to use surrogate water quality parameters to indirectly measure the presence of a contaminant. A positive sensor detection can often be validated by pieces of evidence observed in a distribution system. This paper illustrates how Bayesian belief networks can be used to represent distribution system contamination scenarios. A framework was developed that integrated sensor data with other validating evidence of a contamination event. This framework was used to express causality between the events and observed evidence that comprise contamination scenarios. Copyright ASCE 2005.

Original languageEnglish (US)
Title of host publicationWorld Water Congress 2005
Subtitle of host publicationImpacts of Global Climate Change - Proceedings of the 2005 World Water and Environmental Resources Congress
Pages43
Number of pages1
DOIs
StatePublished - 2005
Externally publishedYes
Event2005 World Water and Environmental Resources Congress - Anchorage, AK, United States
Duration: May 15 2005May 19 2005

Publication series

NameWorld Water Congress 2005: Impacts of Global Climate Change - Proceedings of the 2005 World Water and Environmental Resources Congress

Other

Other2005 World Water and Environmental Resources Congress
Country/TerritoryUnited States
CityAnchorage, AK
Period5/15/055/19/05

ASJC Scopus subject areas

  • Water Science and Technology

Fingerprint

Dive into the research topics of 'Reducing online contaminant monitoring uncertainty using a bayesian belief network'. Together they form a unique fingerprint.

Cite this