Applications of Deep Learning for Smart Water Networks

Wu, Zheng Yi; El-Maghraby, Mahmoud; Pathak, Sudipta

doi:10.1016/j.proeng.2015.08.870

Cited by 39 publications

(8 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Another point to consider is the massive quantity of data that can be recovered, calling for the utilisation of methods based on artificial intelligence to extract pertinent information from this mass of data. Works have been carried out in this direction (Arsene et al, 2022;Brous et al, 2016;Wu et al, 2015) and should be extended. Another current challenge relating to data is the development of emerging approaches that take into account data produced by the users (e.g., (Villesseche et al, 2017); https://adoptadrain.sfwater.org/) (Proposal 15).…”

Section: Directions Of Research Concerning Datamentioning

confidence: 99%

Water infrastructure asset management: state of the art and emerging research themes

Gat

Curt

Werey

et al. 2023

Structure and Infrastructure Engineering

View full text Add to dashboard Cite

The long-term maintenance of infrastructure requires asset management. This latter is defined as a series of actions aimed at the long-term maintenance of the capacity of an infrastructure to provide an efficient service for users, at costs and with impacts that are tolerable for society and the environment. The actions implemented concern notably the inventory of infrastructures, their monitoring, inspection and maintenance, the analysis of their state and performance, their reinforcement, renovation, and the definition of long-term technical-financial policies. This paper is structured around four themes based on the authors' experience faced with an analysis of the perspectives set out in recent articles. For each theme, a state of the art largely based on the literature in the field, as well as emerging and nevertheless pressing needs in terms of knowledge production and tools to support reflection and decision-making are presented: temporal and spatial dimensions of Water Infrastructure Asset Management and their consistency; multi-infrastructure management seen from the perspective of physical interdependences and decision-aids; governance, organisations and territories; digital representation of the socio-technical infrastructural system. The analysis is carried out following several disciplines from the engineering and, the human and social sciences.Infrastructure linked to water are studied.

show abstract

Section: Directions Of Research Concerning Datamentioning

confidence: 99%

Water infrastructure asset management: state of the art and emerging research themes

Gat

Curt

Werey

et al. 2023

Structure and Infrastructure Engineering

View full text Add to dashboard Cite

show abstract

“…Data-driven approaches using hydraulic models have shown promise for leakage detection. For example, Wu et al developed a pressure-dependent leakage detection (PDLD) methodology integrating leak simulation within hydraulic model calibration [23]. However, the methodology's performance suffers from computational complexity when applied to multiple partitioning networks, which have exponentially expanding solution search spaces, though it can successfully identify leakage.…”

Section: Introductionmentioning

confidence: 99%

A Unified Spatial-Pressure Sensitivity Partitioning and Leakage Detection Method within a Deep Learning Framework

Dong,

Shu,

2024

Water

View full text Add to dashboard Cite

This study introduces an innovative approach for leak detection in water distribution systems (WDSs), integrating three-order embedding, k-means clustering, and long short-term memory (LSTM) networks, with pressure-sensitive analysis techniques. This comprehensive methodology segments the network into distinct partitions, utilizes simulated leak events to train the deep learning networks, and establishes a sophisticated model for accurately identifying leak partitions. This approach generates a leak dataset by adjusting water demands, which could effectively pinpoint the leaks in a specific partition by leveraging both the pressure sensitivity and spatial coordinates of nodes, allowing for the elimination of the need for manual work and precise identification of leaks in targeted areas. Through the analysis of two case studies, the model demonstrates its ability to effectively pinpoint potential leak partitions, significantly enhancing operational efficiency and reliability in managing the complex problems of urban water resource management. This approach not only optimizes leak detection but also paves the way for advanced, data-driven strategies in WDSs, ensuring sustainable and secure water distribution in urban settings.

show abstract

“…Machine learning techniques are being used in the domain of water distribution networks for extracting relevant information from large amounts of data generated by smart meters and human observations [14,15]. According to [16], "A truly smart water network needs to be smart at each of the steps to achieve the best outcomes of water network management and operation." WDN models have evolved in the past few decades with increasing support for decision making given the heterogeneity of consumers (e.g., residential, industrial, commercial) and use cases (e.g., regional and metropolitan networks, agriculture and irrigation systems), and demand variation during the day and time of year.…”

Section: Introductionmentioning

confidence: 99%

“…The integration of machine learning provides a foundation for a new paradigm, where the high-level tasks can be delegated for improved data processing capabilities, by extracting relevant information from vast amounts of data [9,16]. With the developments in ML (machine learning), DL (deep learning) and RL (reinforcement learning), a data-driven approach has emerged as a flexible and highly adaptable alternative to more traditional modeling in water distribution systems [22].…”

Section: Introductionmentioning

confidence: 99%

An Advanced Learning-Based Multiple Model Control Supervisor for Pumping Stations in a Smart Water Distribution System

et al. 2020

View full text Add to dashboard Cite

Water distribution is fundamental to modern society, and there are many associated challenges in the context of large metropolitan areas. A multi-domain approach is required for designing modern solutions for the existing infrastructure, including control and monitoring systems, data science and Machine Learning. Considering the large scale water distribution networks in metropolitan areas, machine and deep learning algorithms can provide improved adaptability for control applications. This paper presents a monitoring and control machine learning-based architecture for a smart water distribution system. Automated test scenarios and learning methods are proposed and designed to predict the network configuration for a modern implementation of a multiple model control supervisor with increased adaptability to changing operating conditions. The high-level processing and components for smart water distribution systems are supported by the smart meters, providing real-time data, push-based and decoupled software architectures and reactive programming.

show abstract

Applications of Deep Learning for Smart Water Networks

Cited by 39 publications

References 5 publications

Water infrastructure asset management: state of the art and emerging research themes

Water infrastructure asset management: state of the art and emerging research themes

A Unified Spatial-Pressure Sensitivity Partitioning and Leakage Detection Method within a Deep Learning Framework

An Advanced Learning-Based Multiple Model Control Supervisor for Pumping Stations in a Smart Water Distribution System

Contact Info

Product

Resources

About