Distributed state estimation for leak detection in water supply networks

Abstract

In this paper, we focus on observer-based approaches of leak detection and isolation in pipeline networks, which has the potential to be extended to large-scale water supply network systems, while in previous studies only single pipeline scenario is considered. Due to several disadvantages of implementing centralized state estimation for large-scale systems, the distributed Kalman filter algorithm has been applied in this paper. The observers are designed on the basis of a set of nonlinear fluid models of pipelines, pumps and other hydraulic accessories, with uncertainties in leak position and magnitude. The approach assumes only flow and pressure measurements at the ends of each duct. Simulation results with a water transmission line are presented to assess the efficiency of the proposed distributed state estimation method while comparing with its centralized counterpart.

A Distributed Computing Solution Based on Distributed Kalman Filter for Leak Detection in WSN-Based Water Pipeline Monitoring

“…The drawback with the model-based methods is that they require a precise mathematical model of the pipeline system in order to accurately detect leaks. With the data-driven methods, large amounts of data are being collected and used to analyze, interpret, and extract useful information for operational and other purposes based on Artificial Intelligence (AI) techniques and other data-driven methods [ 61 ]. Their drawback is that they need a large amount of data and a long training time [ 61 ].…”

“…The Kalman filter minimizes the estimated error covariance in a linear stochastic system, has low memory requirements and low complexity [ 93 ], and it is capable of handling situations with a lot of noise or high uncertainty in the data. This therefore makes it a good candidate for improving the accuracy of noisy measured leak signals and detecting leaks in WWPM, as nodes are limited in their memory, processing power and energy [ 8 , 61 , 93 ].…”

“…The software-based methods use field sensors to monitor the operational and hydraulic conditions of the pipeline, such as the measurement of the flow, pressure and temperature [ 16 , 25 , 36 , 54 ] and smart computational algorithms to process the measurements in order to detect and localize the occurrence of leaks on the pipeline [ 29 ]. Some of the software-based methods available in the literature include methods based on model estimation (e.g., Kalman filter [ 19 , 29 , 61 ], state observer [ 62 ], system identification [ 63 ], the impedance method [ 64 ]), methods based on signal processing (Negative Wave Pressure [ 32 ], Mass Balance [ 65 ], Pressure Point Analysis [ 66 ], Acoustic Correlation Analysis [ 67 ], Spectral Analysis Response [ 68 ]) and data-driven methods (Support Vector Machine [ 69 ], K-nearest neighbor [ 70 ], Naïve Bayes [ 69 ]). They involve the use of either intrusive sensors or non- intrusive sensors to monitor the internal pipeline parameters.…”

An Optimization-Based State Estimation Framework for Large-Scale Natural Gas Networks

“…Most existing studies have been concerned with obtaining state estimates to perform gas leak detection. Among these studies, estimation methods used include adaptive particle filters, transfer functions, , and Kalman filters. − The drawbacks of these approaches, when contrasted with optimization-based approaches such as moving-horizon estimation, is well-documented and understood, , particularly in the context of large-scale systems . To the best of our knowledge, optimization-based estimation approaches for natural gas networks have not been studied previously.…”

A Multi-Leak Identification Scheme Using Multi-Classification for Water Distribution Infrastructure

“…To reduce the computational complexity, in [14] a model reduced by integrating multiple pipe branches into a single node was proposed to convert complicated WDI into a network. In terms of the practicability of internal automatic leak detection, various automatic leak detection algorithms have been developed, including state estimation algorithms [15,16], signal analysis algorithms [2,17], machine learning algorithms [12,18,19], etc. The paper [15] proposed a burst detection approach that utilizes an adaptive Kalman filter to perform hydraulic measurement of flow and pressure in a district metered area (DMA).…”