Leak Detection and Isolation in Pressurized Water Pipe Networks using Interval LPV Models

In this paper, a leak detection and isolation methodology in pressurized water pipe networks is proposed. The methodology is based on computing residuals which are obtained comparing measured pressures (heads) in selected points of the network and their estimated values by means of an interval Linear Parameter Varying (LPV) model. The structure of the LPV models is obtained from the nonlinear mathematical model of the network. The proposed detection method uses interval LPV models to obtain uncertainty intervals for the estimated heads that allow to indicate when a leak appears in the water network. The isolation task employs an algorithm based on the residual fault sensitivity analysis. Finally, a typical water pipe network is employed to validate the proposed methodology. This network is simulated using EPANET software. Parameters of LPV models and their uncertainty bounded by intervals are estimated from data coming from this simulator. Several leak scenarios allow to assess the effectiveness of the proposed approach.