Optimising a Fuzzy Logic Real-Time Control System for Sewer Flooding Reduction using a Genetic Algorithm

Will Shepherd, Steve Mounce, Sonja Ostojin, Mohamad Abdel-Aal, Alma Schellart, Peter Skipworth, Simon Tait

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

47 Downloads (Pure)


CENTAUR aims to provide an innovative, cost effective, local autonomous data driven in-sewer flow control system whose operation will reduce urban flood risk. The system comprises of a specially designed flow control device and a wireless local water level monitoring and control system. A data driven algorithm has been developed that is able to analyse the water level data and issue instructions to the flow control device to reduce flood risk at the downstream flooding location. This Fuzzy Logic control algorithm has been linked to a SWMM model to allow virtual testing to take place and provide the basis for a Genetic Algorithm to optimise the Fuzzy Logic membership functions. Methods for generating the initial starting membership functions for input to the Genetic Algorithm have also been investigated. Results confirm that the best Genetic Algorithm optimised Fuzzy Logic controllers reduce flood volume by up to 25% depending on the timestep at which the algorithm is run and the membership function initialisation method.
Original languageEnglish
Title of host publicationCCWI– Computing and Control for the Water Industry
Place of PublicationSheffield
PublisherUniversity of Sheffield
Publication statusPublished - 2017
EventComputing and Control for the Water Industry 2017 - Sheffield, United Kingdom
Duration: 5 Sept 20177 Sept 2017


ConferenceComputing and Control for the Water Industry 2017
Abbreviated titleCCWI 2017
Country/TerritoryUnited Kingdom


Dive into the research topics of 'Optimising a Fuzzy Logic Real-Time Control System for Sewer Flooding Reduction using a Genetic Algorithm'. Together they form a unique fingerprint.

Cite this