Model based leakage isolation in water distribution system: a neural classifier approach

Abstract

The paper presents an approach to detect and isolate leakages in water distribution networks. The first step of the methodology is the definition of the most sensitive nodes of the water distribution network that could be chosen as the places of pressure measurement points. Next, the numerical calculations of the hydraulic model for the purpose of determining the pressure variations for the standard, and during a leakage, operation of the water network were performed. At last neural classifiers were estimated so as to isolate the place of a leakage.