Graph theory algorithms for real time control of a sewer network

Research output: Chapter in Book/Conference proceedings/Edited volumeConference contributionScientificpeer-review

51 Downloads (Pure)

Abstract

Many Dutch sewer networks are combined sewer systems, they carry both storm
water and foul water. They consist of multiple sub-networks, linked by pumps into a tree structure with the Waste Water Treatment Plant as its root. Within sub-networks sewage transport is by gravity driven flow. Usually the original design assumed local control. Later changes, additions and extensions sometimes reduced the effectiveness of the original design. In these cases central control can improve the performance of the system without costly new construction. We apply two algorithms from graph theory, one is based on stable flows in time, the other on quickest evacuation flows. Results on local control are included to provide a lower bound on performance. A linear programming problem based of a perfect forecast of the whole event provides an upper bound on performance.
Original languageEnglish
Title of host publicationEPiC Series in Engineering
Subtitle of host publicationHIC 2018. 13th International Conference on Hydroinformatics
EditorsGoffredo La Loggia , Gabriele Freni, Valeria Puleo, Mauro De Marchis
PublisherEasyChair
Pages2127-2135
DOIs
Publication statusPublished - 2018
EventHIC 2018: 13th International Conference on Hydroinformatics - Palermo, Italy
Duration: 1 Jul 20186 Jul 2018
Conference number: 13
http://www.hydroinformatics.it/index.php?option=com_content&view=article&id=91:hic2018&catid=1:latest-news&Itemid=50

Publication series

NameEPiC Series in Engineering
PublisherEasyChair
Volume3
ISSN (Electronic)2516-2330

Conference

ConferenceHIC 2018: 13th International Conference on Hydroinformatics
Abbreviated titleHIC 2018
CountryItaly
CityPalermo
Period1/07/186/07/18
Internet address

Fingerprint Dive into the research topics of 'Graph theory algorithms for real time control of a sewer network'. Together they form a unique fingerprint.

Cite this