Cloud Based Machine Learning Approaches for Leakage Assessment and Management in Smart Water Networks. (2015)

• Record Type: Journal Article
• Authors: Mounce, S.R.; Pedraza, C.; Jackson, T.; Linford, P.; Boxall, J.B.
• Is Part Of: Procedia engineering. Volume 119(2015)
• Page Start: 43
• Page End: 52
• Publication Date: 2015
• Subjects: Leakage -- Smart Networks -- AMR -- Neural Networks -- Cloud computing.; Engineering -- Congresses; Engineering -- Periodicals; Engineering; Conference proceedings; Periodicals; 620.005
• DOI: 10.1016/j.proeng.2015.08.851
• Languages: English
• ISSNs: 1877-7058

Abstract

Abstract: One-third of utilities around the globe report a loss of more than 40 percent of clean water due to leaks. By reducing the amount of water leaked, smart water networks can help reduce the money wasted on producing or purchasing water, and the related energy required to pump water and treat water for distribution. A UK demo site is presented focusing on leak management, integrating fixed flow and pressure instrumentation, advanced (smart) metering infrastructure and novel instruments (capable of high resolution monitoring). Example data analysis results for this site using the AURA-Alert anomaly detection system for Condition Monitoring are presented.