Article
Automation & Control Systems
E. J. Perez-Perez, F. R. Lopez-Estrada, G. Valencia-Palomo, L. Torres, V Puig, J. D. Mina-Antonio
Summary: In this paper, a methodology using artificial neural networks (ANN) techniques and online measurements of pressure and flow rate is presented for detecting and locating water leaks in pipelines. The method estimates the friction factor of the pipe and uses this information to compute the leak position, showing good performance and applicability in experimental tests.
CONTROL ENGINEERING PRACTICE
(2021)
Review
Engineering, Chemical
Saikat Sinha Ray, Rohit Kumar Verma, Ashutosh Singh, Mahesh Ganesapillai, Young-Nam Kwon
Summary: In recent years, deep learning and machine learning have emerged as potential technologies widely applied in the fields of science, engineering, and technology, specifically in the optimization of seawater desalination and water treatment processes. Artificial intelligence has played a key role in addressing issues such as monitoring, management, and labor costs. This article thoroughly reviews the application of AI in the water treatment and seawater desalination sectors, compares conventional modeling approaches with artificial neural network modeling, and discusses challenges, shortcomings, and future prospects. The use of AI mechanisms in data processing, optimization, modeling, prediction, and decision-making during water treatment and seawater desalination processes are emphasized, along with innovative trends in these areas.
Article
Environmental Sciences
Michael E. Omeka, Ogbonnaya Igwe, Obialo S. Onwuka, Ogechukwu M. Nwodo, Samuel I. Ugar, Peter A. Undiandeye, Ifeanyi E. Anyanwu
Summary: Agricultural productivity can be affected by poor irrigation water quality, making vulnerability assessment and identification of influential water quality parameters important for water resource management. This study integrated GIS, AHP, and machine learning models to assess and predict irrigation water quality suitability. Results showed that 28.2% of the area was suitable for irrigation, 43.7% was moderately suitable, and 28.1% was unsuitable, with deterioration in the central-southeastern direction. MLP-NN outperformed MLR in predicting IWQ parameters, and HCO3, Cl, Mg, and SO4 were identified as having the greatest influence on irrigation water quality.
ENVIRONMENTAL SCIENCE AND POLLUTION RESEARCH
(2023)
Article
Engineering, Civil
Jessica Bohorquez, Angus R. Simpson, Martin F. Lambert, Bradley Alexander
Summary: This paper introduces a new technique that uses artificial neural networks to detect and identify bursts in pipelines by interpreting transient pressure waves. The technique is divided into model development and application stages. Experimental and numerical simulation results demonstrate that the technique can accurately predict the location of the burst, but predicting the burst size requires further steps to ensure accuracy.
JOURNAL OF WATER RESOURCES PLANNING AND MANAGEMENT
(2021)
Article
Engineering, Civil
Jessica Bohorquez, Martin F. Lambert, Bradley Alexander, Angus R. Simpson, Derek Abbott
Summary: Water losses through leakage pose a significant problem for asset management in water distribution systems. Previous approaches to locate and characterize leaks used the interpretation of fluid transient pressure waves, but these approaches were often model-driven and limited by existing knowledge of the system. Recently, the potential of using artificial neural networks (ANN) and fluid transient waves to detect, locate, and characterize anomalies in water pipelines has been proposed. However, applying this technique in more realistic conditions has been challenging. This paper demonstrates the enhanced detection of leaks in pressurized pipelines by deploying stochastic resonance, and presents a methodology for active inspection of pipelines using convolutional neural networks (CNNs). The results of their experiments on a real pipeline showed promising potential for developing CNN-based techniques to detect leaks and anomalies in water pipelines.
JOURNAL OF WATER RESOURCES PLANNING AND MANAGEMENT
(2022)
Article
Engineering, Multidisciplinary
Zahoor Ahmad, Tuan-Khai Nguyen, Jong-Myon Kim
Summary: This paper proposes a leak detection and size identification technique in fluid pipelines based on a new leak-sensitive feature called the vulnerability index (VI) and 1-D convolutional neural network (1D-CNN). The technique extracts acoustic emission hit features and applies a multiscale Mann-Whitney test to obtain the vulnerability index feature, which shows the pipeline's susceptibility to leak and changes according to the pipeline working conditions. Experimental results demonstrate that the proposed method outperforms state-of-the-art methods under variable fluid pressure conditions.
ENGINEERING APPLICATIONS OF COMPUTATIONAL FLUID MECHANICS
(2023)
Article
Green & Sustainable Science & Technology
Roberto Magini, Manuela Moretti, Maria Antonietta Boniforti, Roberto Guercio
Summary: In a water distribution network, an artificial neural network is used to estimate the pressure field based on limited nodal data, allowing for sustainable management of water resources without the need for monitoring all nodes.
Article
Computer Science, Artificial Intelligence
Alireza Keramat, Iman Ahmadianfar, Huan-Feng Duan, Qingzhi Hou
Summary: This study aims to combine metaheuristic and gradient-based optimization techniques to find the optimal solution of the novel objective function and locate the leak positions.
EXPERT SYSTEMS WITH APPLICATIONS
(2023)
Article
Computer Science, Artificial Intelligence
Nehal Elshaboury, Mohamed Marzouk
Summary: This research explores the application of optimized neural network models for assessing the condition of water pipelines in Shaker Al-Bahery, Egypt. Results indicate that the neural network trained with the Particle Swarm Optimization (PSO) algorithm outperforms other machine learning models. Additionally, swarm intelligence algorithms are utilized to determine optimal intervention strategies for pipeline maintenance.
Article
Chemistry, Multidisciplinary
Yao-Long Tsai, Hung-Chih Chang, Shih-Neng Lin, Ai-Huei Chiou, Tin-Lai Lee
Summary: To address the challenges brought by abnormal weather and industrial water consumption in Taiwan, the government has implemented measures such as transporting water and investing in backup water pipelines. However, the high leakage rate of water pipelines remains a concern. This study developed an intelligent sound-assisted water leak identification system using artificial intelligence and IoT technology, which has shown high accuracy and reliability in identifying and locating leaks.
APPLIED SCIENCES-BASEL
(2022)
Letter
Thermodynamics
Cherifa Kara Mostefa Khelil, Badia Amrouche, Kamel Kara, Aissa Chouder
Summary: This study examines the impact of different Artificial Neural Networks on fault diagnosis in PV installations. It shows that RBF ANNs affect the algorithm's reaction rate, while BPNNs and GRNN present the best results in terms of speed, high precision, and classification efficiency. PNN also stands out for achieving 100% accuracy in key statistical concepts.
ENERGY CONVERSION AND MANAGEMENT
(2021)
Article
Energy & Fuels
Omer Boyukdipi, Gokhan Tuccar, Hakan Serhad Soyhan
Summary: The experimental study investigated the effects of NH3 as a fuel additive on engine vibration parameters, revealing that increasing levels of NH3 additive led to increased engine vibration and had a negative impact on engine vibration when blended with sunflower biodiesel. High accuracy rates were achieved in predicting vibration data through artificial neural networks models.
Article
Chemistry, Analytical
Guang Yang, Hai Wang
Summary: This paper proposes a practical method to optimize the deployment of pressure sensors in water distribution networks in order to minimize the leakage ratio. The methodology generates leakage events through model simulation and calculates the essential sensors for leak identification. If there is a surplus budget, supplementary sensors can be determined to enhance the leak identification ability. Experimental results show that the method is highly suitable for real projects.
Article
Environmental Sciences
Vineet Tyagi, Prerna Pandey, Shashi Jain, Parthasarathy Ramachandran
Summary: This research proposes a two-stage model that uses data analysis to predict leak occurrences and their specific locations in water distribution networks. The model is cost-effective and easily deployable.
Article
Environmental Sciences
I. A. Tijani, S. Abdelmageed, A. Fares, K. H. Fan, Z. Y. Hu, T. Zayed
Summary: This study developed machine learning-based leak detection models for real water distribution networks by recording acoustic signals and extracting features from the signals. The models developed using features extracted from de-noised signals showed better classification accuracy compared to models developed using features from raw signals.
SCIENCE OF THE TOTAL ENVIRONMENT
(2022)
Article
Engineering, Civil
Jessica Bohorquez, Angus R. Simpson, Martin F. Lambert, Bradley Alexander
Summary: This paper introduces a new technique that uses artificial neural networks to detect and identify bursts in pipelines by interpreting transient pressure waves. The technique is divided into model development and application stages. Experimental and numerical simulation results demonstrate that the technique can accurately predict the location of the burst, but predicting the burst size requires further steps to ensure accuracy.
JOURNAL OF WATER RESOURCES PLANNING AND MANAGEMENT
(2021)
Article
Engineering, Civil
Xiao-xuan Du, Wei Zeng, Martin F. Lambert, Lei Chen, Eric Jing Hu
Summary: This paper presents an approach for pipe burst detection, localization, and cross-sectional area quantification based on changes in the discrete harmonic spectrogram and analysis of damping of fluid transients. The developed pressure signal processing algorithm allows for real-time data monitoring with lower data transmission and sampling rates compared to commonly used data acquisition systems. The algorithm has been verified both numerically and experimentally for its effectiveness in pipe burst detection.
JOURNAL OF WATER RESOURCES PLANNING AND MANAGEMENT
(2021)
Article
Engineering, Civil
Wei Zeng, Aaron C. Zecchin, Benjamin S. Cazzolato, Angus R. Simpson, Jinzhe Gong, Martin F. Lambert
Summary: A new higher-order paired-IRF has been proposed, along with a correlator to highlight anomaly-induced spikes and suppress noise. Experimental results show that the method is extremely sensitive in detecting anomalies and can accurately identify reflections even as small as 0.5% of the injected wave magnitude.
JOURNAL OF WATER RESOURCES PLANNING AND MANAGEMENT
(2021)
Article
Engineering, Civil
Jessica Bohorquez, Martin F. Lambert, Bradley Alexander, Angus R. Simpson, Derek Abbott
Summary: Water losses through leakage pose a significant problem for asset management in water distribution systems. Previous approaches to locate and characterize leaks used the interpretation of fluid transient pressure waves, but these approaches were often model-driven and limited by existing knowledge of the system. Recently, the potential of using artificial neural networks (ANN) and fluid transient waves to detect, locate, and characterize anomalies in water pipelines has been proposed. However, applying this technique in more realistic conditions has been challenging. This paper demonstrates the enhanced detection of leaks in pressurized pipelines by deploying stochastic resonance, and presents a methodology for active inspection of pipelines using convolutional neural networks (CNNs). The results of their experiments on a real pipeline showed promising potential for developing CNN-based techniques to detect leaks and anomalies in water pipelines.
JOURNAL OF WATER RESOURCES PLANNING AND MANAGEMENT
(2022)
Article
Engineering, Multidisciplinary
Chi Zhang, Bradley J. Alexander, Mark L. Stephens, Martin F. Lambert, Jinzhe Gong
Summary: The implementation of a smart water network is an effective approach to address challenges faced by water utilities. This paper develops a CNN-based model to classify acoustic wave files collected by the SWN and extract features using transfer learning. The developed models have been validated and shown to be an effective tool for water pipeline leak and crack detection, enabling proactive management of pipeline assets.
STRUCTURAL HEALTH MONITORING-AN INTERNATIONAL JOURNAL
(2023)
Article
Engineering, Civil
Chi Zhang, Mark L. Stephens, Martin F. Lambert, Bradley J. Alexander, Jinzhe Gong
Summary: This paper describes a technique for early detection of pipe cracks and its implementation in a smart water network using an acoustic monitoring system. The successful detection of multiple pipe cracks/leaks proves its effectiveness in proactive management of pipe breaks in water distribution systems.
JOURNAL OF WATER RESOURCES PLANNING AND MANAGEMENT
(2022)
Article
Engineering, Civil
A. C. Zecchin, N. Do, J. Gong, M. Leonard, M. F. Lambert, M. L. Stephens
Summary: This paper investigates the criteria for optimal sensor deployment in a water distribution system and develops a technique for determining the optimal sensor locations. By maximizing the network extent for detecting and locating hydraulic transient events, the concept of event locatability is proposed. The effectiveness of the proposed method is verified through case studies.
JOURNAL OF WATER RESOURCES PLANNING AND MANAGEMENT
(2022)
Article
Engineering, Civil
Byron Guerrero, Martin F. Lambert, Rey C. Chin
Summary: This work examines the flow dynamics contributing to the wall shear stress of accelerating and decelerating turbulent pipe flows. Results reveal the differences in time dependence and turbulence response between accelerating and decelerating flows. While an existing 1D unsteady friction model accurately predicts one component of the dynamic decomposition, it fails to capture the transient response of laminar and turbulent contributions. Therefore, a hybrid model based on the identified flow dynamics is proposed to improve the current approach.
JOURNAL OF HYDRAULIC ENGINEERING
(2022)
Article
Engineering, Civil
Wei Zeng, Jinzhe Gong, Aaron C. Zecchin, Martin F. Lambert, Benjamin S. Cazzolato, Angus R. Simpson
Summary: This paper proposes a technique for assessing the condition of pipeline walls in water distribution systems using persistent hydraulic transient waves. The technique focuses on detecting and reconstructing extended and irregular anomalies in the pipe walls. Numerical verifications show that the technique can successfully detect and accurately reconstruct nonuniformly deteriorated sections.
JOURNAL OF HYDRAULIC ENGINEERING
(2023)
Article
Engineering, Civil
Xiao-xuan Du, Martin F. Lambert, Lei Chen, Eric Jing Hu
Summary: The purpose of this study was to illustrate the relationship between methods that utilize the damping of fluid transients and approaches based on frequency response diagrams and to discuss their applications in both leak and burst detection. The mathematical relationship between the two methods was revealed and verified numerically and experimentally. Additionally, a comparison was made between the two methods in terms of input signal bandwidth, problem type, low sampling rate capability, robustness, and real-time data monitoring capability, discussing their applicability in these aspects.
JOURNAL OF HYDRAULIC ENGINEERING
(2023)
Article
Engineering, Civil
Wei Zeng, Aaron C. Zecchin, Martin F. Lambert
Summary: This paper introduces a novel Elastic Water Column (EWC) model for analyzing hydraulic transients in pipe networks, showing higher accuracy than standard models on 6- and 51-pipe networks.
JOURNAL OF HYDRAULIC ENGINEERING
(2022)
Article
Engineering, Civil
Nhu Cuong Do, Luke Dix, Martin Francis Lambert, Mark Leslie Stephens
Summary: This paper describes a continuous monitoring system for a real sewage network using ultrasonic water level sensors, implemented in Stonyfell, South Australia. Analysis of 62 data sets collected over 1 year identified two distinctive features of growing blockages, and an early detection method based on one of these features was formulated. Application of the method showed its effectiveness in detecting possible blockages and overflow events before they occur.
JOURNAL OF WATER RESOURCES PLANNING AND MANAGEMENT
(2023)
Article
Engineering, Civil
Xiao-xuan Du, Martin F. Lambert, Lei Chen, Eric Hu
Summary: The paper presents an approach for real-time pipe burst detection and location estimation based on changes in harmonics and damping analysis. Previous methods could not utilize higher order harmonics due to an increasing number of possible burst location solutions. However, the proposed algorithm overcomes this limitation and excludes incorrect solutions by analyzing damping. Additionally, real-time data analysis is made possible by setting the window gap between data windows. The approach has been verified numerically and experimentally with acceptable accuracy.
JOURNAL OF WATER RESOURCES PLANNING AND MANAGEMENT
(2023)