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

被引:26
作者
Bonilla, Carlos A. [1 ,2 ]
Zanfei, Ariele [3 ]
Brentan, Bruno [2 ,4 ]
Montalvo, Idel [2 ,5 ]
Izquierdo, Joaquin [2 ]
机构
[1] Univ Pamplona, Fac Engn & Architecture, Dept Civil Environm & Chem Engn, Pamplona 543050, Colombia
[2] Univ Politecn Valencia, Fluing Inst Multidisciplinary Math, Valencia 46022, Spain
[3] Free Univ Bozen, Fac Sci & Technol, Piazza Univ 5, I-39100 Bolzano, Italy
[4] Univ Fed Minas Gerais, Sch Engn, Hydraul Engn & Water Resources Dept, BR-31270901 Belo Horizonte, MG, Brazil
[5] IngeniousWare GmbH, Business Dev, Jollystr 11, D-76137 Karlsruhe, Germany
关键词
graph convolutional neural networks; machine learning; state estimation; water distribution system; hydraulic modeling; digital twin; MODEL;
D O I
10.3390/w14040514
中图分类号
X [环境科学、安全科学];
学科分类号
08 ; 0830 ;
摘要
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.
引用
收藏
页数:16
相关论文
共 50 条
[1]  
[Anonymous], 2014, Advances in hydroinformatics, DOI DOI 10.1007/978-981-4451-42-0_44
[2]   A3T-GCN: Attention Temporal Graph Convolutional Network for Traffic Forecasting [J].
Bai, Jiandong ;
Zhu, Jiawei ;
Song, Yujiao ;
Zhao, Ling ;
Hou, Zhixiang ;
Du, Ronghua ;
Li, Haifeng .
ISPRS INTERNATIONAL JOURNAL OF GEO-INFORMATION, 2021, 10 (07)
[3]   Deep spatial-temporal sequence modeling for multi-step passenger demand prediction [J].
Bai, Lei ;
Yao, Lina ;
Wang, Xianzhi ;
Li, Can ;
Zhang, Xiang .
FUTURE GENERATION COMPUTER SYSTEMS-THE INTERNATIONAL JOURNAL OF ESCIENCE, 2021, 121 :25-34
[4]   ANN-based residential water end-use demand forecasting model [J].
Bennett, Christopher ;
Stewart, Rodney A. ;
Beal, Cara D. .
EXPERT SYSTEMS WITH APPLICATIONS, 2013, 40 (04) :1014-1023
[5]   Merging Fluid Transient Waves and Artificial Neural Networks for Burst Detection and Identification in Pipelines [J].
Bohorquez, Jessica ;
Simpson, Angus R. ;
Lambert, Martin F. ;
Alexander, Bradley .
JOURNAL OF WATER RESOURCES PLANNING AND MANAGEMENT, 2021, 147 (01)
[6]   Leak Detection and Topology Identification in Pipelines Using Fluid Transients and Artificial Neural Networks [J].
Bohorquez, Jessica ;
Alexander, Bradley ;
Simpson, Angus R. ;
Lambert, Martin F. .
JOURNAL OF WATER RESOURCES PLANNING AND MANAGEMENT, 2020, 146 (06)
[7]   Hybrid regression model for near real-time urban water demand forecasting [J].
Brentan, Bruno M. ;
Luvizotto, Edevar, Jr. ;
Herrera, Manuel ;
Izquierdo, Joaquin ;
Perez-Garcia, Rafael .
JOURNAL OF COMPUTATIONAL AND APPLIED MATHEMATICS, 2017, 309 :532-541
[8]   Geometric Deep Learning Going beyond Euclidean data [J].
Bronstein, Michael M. ;
Bruna, Joan ;
LeCun, Yann ;
Szlam, Arthur ;
Vandergheynst, Pierre .
IEEE SIGNAL PROCESSING MAGAZINE, 2017, 34 (04) :18-42
[9]   Digital Twins in Civil Infrastructure Systems [J].
Callcut, Matthew ;
Agliozzo, Jean-Paul Cerceau ;
Varga, Liz ;
McMillan, Lauren .
SUSTAINABILITY, 2021, 13 (20)
[10]   Near-Real Time Burst Location and Sizing in Water Distribution Systems Using Artificial Neural Networks [J].
Capelo, Miguel ;
Brentan, Bruno ;
Monteiro, Laura ;
Covas, Didia .
WATER, 2021, 13 (13)