A Digital Twin of a Water Distribution System by Using Graph Convolutional Networks for Pump Speed-Based State Estimation

Abstract

Water distribution system monitoring is currently carried out using advanced real-time control technologies to achieve a higher operational efficiency. Data analysis techniques can be implemented for condition estimation, which are crucial tools for managing, developing, and operating water networks using the monitored flow rate and pressure data at some network pipes and nodes. This work proposes a state estimation methodology that enables one to infer the hydraulic state of the operating speed of pumping systems from these pressure and flow measurements. The presented approach suggests using graph convolutional neural network theory linked to hydraulic models for generating a digital twin of the water system. It is validated on two benchmark hydraulic networks: the Patios-Villa del Rosario, Colombia, and the C-Town networks. The results show that the proposed model effectively predicts the state estimation in the two hydraulic networks used. The results of the evaluation metrics indicate low values of mean squared error and mean absolute error and high values of the coefficient of determination, reflecting high predictive ability and that the prediction results adequately represent the real data.

Digital Twin-Driven Machine Condition Monitoring: A Literature Review

“…Physical-based Principle-based modelling [59]; finite element analysis [60]; computational fluid dynamics [68]; equivalent modelling [48] Data-driven Machine learning [53]; neural network [61]; deep learning [69] Hybrid Reduced-order modelling [30]; surrogate modelling [23] 8…”

“…Recently, hybrid approaches combining physical-based and data-driven models have been investigated in academics to reduce the computing and mapping time of a high-order virtual model. Bonilla et al used graph convolutional neural network theory and a hydraulic modelling method to generate a digital twin of the water system [61]. Magargle et al built a…”

Digital Twin-Based Integrated Monitoring System: Korean Application Cases

“…They presented an example of a digital twin implementation of a water distribution network in Valencia (Spain) and its metropolitan area (1.6 million citizens). Bonilla et al [ 23 ] introduced a digital twin-based water distribution system and applied and verified the system in two regions of Colombia. Ramos et al [ 24 ] demonstrated to Water Distribution Networks that water savings of up to 28% can be achieved through fast detection of leaks based on a digital twin system.…”

Dynamic effects of a regulating valve in the assessment of water leakages in single pipelines

“…Because all supply systems eventually suffer from leakages, efficient management is required to effectively identify the origin of leaks ( Ávila et al, 2021;Adedeji et al, 2017). Hence, water utility companies must establish pipe-leakage mitigation programmes aimed at improving the service efficiency and decreasing costs borne by water treatment plants, which is of the utmost important for establishing reliable digital twins in water distribution systems (Bonilla et al, 2022;Galdiero et al, 2015). According to the International Water Association (IWA) conventions (Lambert and Hirner, 2000;Lambert et al, 1999), water flow in the supply systems is divided into two types of flows: (i) the flow rate measured, which represents the authorised/billed consumption at households, industries, businesses and institutions and (ii) the uncontrolled flow rate, which is subdivided into apparent losses (customer flow meter inaccuracies, unmetered consumption, unauthorised consumption, water used for firefighting and water used for cleaning streets and public areas) and physical losses (leakages).…”