EKF-based observers for multi-leak diagnosis in branched pipeline systems

Abstract

The present work deals with the multi-leak diagnosis problem in a branched pipeline configuration as in water distribution systems. Here, it is assumed that the flow rate and pressure head measurements are available upstream and at all delivering points of the network. The proposed Leak Detection and Isolation (LDI) scheme basically involves two essential steps: leak region identification based on flow-rate residuals with a related k-Nearest Neighbors (k-NN) classifier, and then leak parameter identification (magnitude and position) via the use of the so-called Extended Kalman Filters (EKFs) for each leak based on a simple generic model and fed with pressure head estimations provided by an initial EKF. For the sake of illustration, successful experimental results of a two sequential leak scenario are provided using databases generated by a test bed plant with two branchings built at the Tuxtla Gutiérrez Institute of Technology.