Leak localization in water distribution networks using a mixed model-based/data-driven approach

• The proposed method for leak localization in WDNs combines models with classifiers.
• A pressure model of the considered WDN is used in a first stage to compute residuals.
• Then, a classifier is applied to the obtained residuals to perform leak localization.
• The training data is generated by simulations of different faults and uncertainties.
• The results are tested in different WDNs and compared with current approaches.

This paper proposes a new method for leak localization in water distribution networks (WDNs). In a first stage, residuals are obtained by comparing pressure measurements with the estimations provided by a WDN model. In a second stage, a classifier is applied to the residuals with the aim of determining the leak location. The classifier is trained with data generated by simulation of the WDN under different leak scenarios and uncertainty conditions. The proposed method is tested both by using synthetic and experimental data with real WDNs of different sizes. The comparison with the current existing approaches shows a performance improvement.