%0 Journal Article %A Younis, Mohammed Chachan %A Keedwell, Edward %A Savic, Dragan %A Raine, Anthony %D 2017 %T CCWI2017: F78 'Evaluating Classification Algorithms for Improved Wastewater System Calibration' %U https://orda.shef.ac.uk/articles/journal_contribution/CCWI2017_F78_Evaluating_Classification_Algorithms_for_Improved_Wastewater_System_Calibration_/5363806 %R 10.15131/shef.data.5363806.v1 %2 https://ndownloader.figshare.com/files/9218269 %K Urban drainage %K InfoWorks ICM %K Image classification %K CCWI2017 %K Civil Engineering not elsewhere classified %X Hydraulic models provide an approximate model of rainfall collection (storm), wastewater collection (foul), and combined (both rainfall and wastewater collections) network performance, capturing the large scale element of the system, but such systems require calibration with real world data to achieve reliable and accurate results. However, many factors surrounding real-world network and catchment characteristics are unknown and can influence the hydraulic performance of the network. Consequently, calibrating wastewater models to accurately reflect real-world conditions is a time consuming and complex process. This paper employs a two stage urban runoff forecasting approach with a combination of image classification techniques and modelling using InfoWorks ICM. The image classification section consists of the automated processing of the satellite image as an aid to modelling wastewater in the urban environment by classifying the land- cover as pervious and impervious segments of water from rainfall events. By classifying such areas with an urban catchment, and by using the Wallingford PR Equation across all tests of Area 1 (roads), 2 (roofs) and 3 (permeable area), we explore the potential for a partially automated network system calibration process of a wastewater network.
%I The University of Sheffield