State Estimation in Water Distribution Networks through a New Successive Linear Approximation

Wang, Shen; Taha, Ahmad F.; Sela, Lina; Gatsis, Nikolaos; Giacomoni, Marcio

doi:10.1109/cdc40024.2019.9029744

Cited by 9 publications

(13 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this way, the DSE solution remains deterministic. The studies in [7]- [10], [12] make up the bulk of recent DSE literature.…”

Section: A Literature Reviewmentioning

confidence: 99%

“…The authors in [7], [14] propose a constrained method to overcome the downside of a weighted based approach via adding parts of objective functions to the constraints. Recently, we propose a novel geometric programming (GP)-based method to solve the SE problem is proposed in our recent paper [12].…”

Section: A Literature Reviewmentioning

confidence: 99%

“…The research gaps are summarized as follows. First, the deterministic SE studies [7]- [12] do not consider critical uncertainty (demand and pipe coefficient uncertainty) in the network that could lead to significantly different estimates for unmeasured state variables. Second, the methods based on CLA or uncertainty quantification [14]- [21] suffer from the following limitations.…”

Section: B Research Gaps Paper Contributions and Organizationmentioning

confidence: 99%

“…. , x 0 (k + T )} through solving a scalable, deterministic state estimation routine (see [12] and references therein) given the means of all uncertainty for all time-steps. For example, we consider that a demand prediction is given from k to k + T .…”

Section: Formulating and Solving The Pse Problemmentioning

confidence: 99%

“…There are two scenarios for SE in water networks [12], [37]: (i) Sufficient scenario is described by having the same number of equations and unknowns. When measurements include the heads at tanks and reservoirs, and demand pseudomeasurements are available for state estimation, we consider it as a sufficient scenario.…”

Section: B Sufficient and Over-determined Scenariosmentioning

confidence: 99%

See 4 more Smart Citations

Probabilistic State Estimation in Water Networks

Wang

Taha

Gatsis

et al. 2020

Preprint

Self Cite

View full text Add to dashboard Cite

State estimation in water distribution networks (WDN), the problem of estimating all unknown network heads and flows given select measurements, is challenging due to the nonconvexity of hydraulic models and significant uncertainty from demands, network parameters, and measurements. To this end, probabilistic modeling for state estimation (PSE) in WDNs is proposed. Regardless of uncertainty's statistical distributions, the PSE shows that the covariance matrix of unknown system states (unmeasured heads and flows) can be linearly expressed by the covariance matrix of uncertainty from measurement noise, network parameters, and demand at arbitrary operating pointsafter linearizing the nonlinear hydraulic models. Rather than computing point-based estimates for unknown states, the proposed PSE approach (i) yields variances of individual unknown states, (ii) amounts to solving a highly scalable linear systems of equations, (iii) is also useful for uncertainty quantification, extended period simulations, or confidence limit analysis, and (iv) includes the modeling of various types of valves and measurement scenarios in water networks. Thorough case studies demonstrate the effectiveness and scalability of the proposed approach.

show abstract

“…In this way, the DSE solution remains deterministic. The studies in [7]- [10], [12] make up the bulk of recent DSE literature.…”

Section: A Literature Reviewmentioning

confidence: 99%

Section: A Literature Reviewmentioning

confidence: 99%

Section: B Research Gaps Paper Contributions and Organizationmentioning

confidence: 99%

Section: Formulating and Solving The Pse Problemmentioning

confidence: 99%

Section: B Sufficient and Over-determined Scenariosmentioning

confidence: 99%

See 3 more Smart Citations

Probabilistic State Estimation in Water Networks

Wang

Taha

Gatsis

et al. 2020

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

An Efficient Approach for Nodal Water Demand Estimation in Large-scale Water Distribution Systems

Chu

Zhang

Zhou

et al. 2022

Water Resour Manage

View full text Add to dashboard Cite

Real-time modeling of the water distribution system (WDS) is a critical step for the control and operation of such systems. The nodal water demand as the most important time-varying parameter must be estimated in real-time. The computational burden of nodal water demand estimation is intensive, leading to inefficiency for the modeling of the large-scale network. The Jacobian matrix computation and Hessian matrix inversion are the processes that dominate the main computation time. To address this problem, an approach to shorten the computational time for the real-time demand estimation in the large-scale network is proposed. The approach can efficiently compute the Jacobian matrix based on solving a system of linear equations, and a Hessian matrix inversion method based on matrix partition and Iterative Woodbury-Matrix-Identity Formula is proposed. The developed approach is applied to a large-scale network, of which the number of nodal water demand is 12523, and the number of measurements ranging from 10 to 2000.Results show that the time consumptions of both Jacobian computation and Hessian matrix inversion are significantly shortened compared with the existing approach.

show abstract

How Effective is Model Predictive Control in Real‐Time Water Quality Regulation? State‐Space Modeling and Scalable Control

Wang

Taha

Abokifa

2021

Water Resources Research

Self Cite

View full text Add to dashboard Cite

Before drinking water leaves water treatment plants, chemical disinfection is typically applied to ensure the microbiological safety of the treated water. Water utilities worldwide rely on chlorine-based disinfectants due to their strong antimicrobial activity and low cost. Excess chlorine is usually applied at the treatment plant to prevent microbial recontamination of the treated drinking water as it moves through the pipes of water distribution networks (WDN).Residual chlorine concentrations are routinely monitored to verify that a sufficient residual is maintained throughout WDN. Maintenance of a detectable residual is also typically mandated by state and federal regulations in many countries. For instance, water utilities in the US are required to preserve detectable chlorine residual throughout their WDNs under the Surface Water Treatment Rule (SWTR) (Haas, 1999), and many states have established even more stringent numerical thresholds on the minimum residual concentration (Roth & Cornwell, 2018).Nevertheless, determining the appropriate chlorine dosage to ensure a sufficient residual, particularly at the far ends of WDNs where the water age is the highest, is rather challenging. Applying large doses of chlorine-based disinfectants at the treatment plant has been associated with multiple issues, including the excessive formation of disinfection byproducts as well as aesthetic issues with water taste and odor (Fisher et al., 2011;Hua et al., 2015). Alternatively, the disinfectant can be injected in smaller doses at multiple locations in the network, a practice commonly known as booster disinfection, to maintain a uniform disinfectant concentration throughout the WDN (Tryby et al., 1999). Most recently, to solve the problem of low disinfectant concentrations at critical dead-end nodes with no need of increasing disinfectant dose at sources or installing additional booster stations, the modulation of nodal outflows in WDN is proposed (Avvedimento et al., 2020). For more context of real-time control of water quality in WDN (see Creaco et al., 2019). Literature ReviewOver the past two decades, many studies have investigated the water quality control problem (WQC) of optimizing the locations and/or dosing schedules of booster disinfection systems.A wide range of optimization-based methods was used to solve the WQC problem, including linear programming (LP), quadratic programming (QP), heuristic algorithms such as genetic algorithm (GA), and

show abstract

State Estimation in Water Distribution Networks through a New Successive Linear Approximation

Cited by 9 publications

References 17 publications

Probabilistic State Estimation in Water Networks

Probabilistic State Estimation in Water Networks

An Efficient Approach for Nodal Water Demand Estimation in Large-scale Water Distribution Systems

How Effective is Model Predictive Control in Real‐Time Water Quality Regulation? State‐Space Modeling and Scalable Control

Contact Info

Product

Resources

About