Article
Environmental Sciences
Bin Pan, Huan-Feng Duan, Alireza Keramat, Silvia Meniconi, Bruno Brunone
Summary: This study proposes a method for leak/burst detection in tree-shaped pipe networks, using forward and backward transient analysis to locate potential pipe defects and calculate their positions. Experimental and numerical tests show the effectiveness of this method, which is applicable to both transient and steady conditions, with high efficiency and wide applicability.
WATER RESOURCES RESEARCH
(2022)
Article
Engineering, Civil
Sehyeong Kim, Sanghoon Jun, Donghwi Jung
Summary: This study proposes an ensemble convolutional neural network model that employs multiple burst detection tools for more effective detection of pipe burst events in water distribution systems.
WATER RESOURCES MANAGEMENT
(2022)
Article
Green & Sustainable Science & Technology
Zukang Hu, Wenlong Chen, Helong Wang, Pei Tian, Dingtao Shen
Summary: Data-driven anomaly detection and early warning are crucial in water distribution systems to identify abnormal events such as pipe bursts and sensor failures. This study proposes a framework comprising four modules to accurately detect anomalies and provide early warnings.
JOURNAL OF CLEANER PRODUCTION
(2022)
Article
Environmental Sciences
Miguel Capelo, Bruno Brentan, Laura Monteiro, Didia Covas
Summary: The paper proposes a novel methodology using Multi-Layer Perceptron (MLP) for near-real time burst location and sizing in water distribution systems (WDS). The methodology involves constructing a pipe-burst database, formulating problems, defining ANN architecture, training and testing ANN, and applying it based on collected data. The trained ANN successfully locates and determines burst sizes with high accuracy in real-life networks, demonstrating its potential as a daily management tool for water distribution networks.
Article
Chemistry, Analytical
Danilo Aparecido Carnevale Castillo, Marco Carminati
Summary: Water leakage is a major issue in distribution infrastructures, with significant water loss in old networks. To tackle this problem, we developed an impedance sensor capable of detecting small water leaks. This sensor utilizes longitudinal electrodes placed on the external surface of pipes, which detect changes in impedance caused by the presence of water. We conducted numerical simulations and laboratory experiments to optimize the sensor's design and prove its effectiveness.
Article
Thermodynamics
Hayeol Kim, Jewhan Lee, Taekyeong Kim, Seong Jin Park, Hyungmo Kim, Im Doo Jung
Summary: In this study, a novel pipe-in-pipe (PIP) leakage detection system utilizing distributed temperature sensing (DTS) with machine learning (ML) is proposed. The system is able to detect liquid leakage between inner and outer pipe ranging from 0.2 to 7 ml/min using Fourier transformed spectrogram data from DTS, achieving an accuracy of 91.67% with a single embedded optical fiber. The system successfully distinguished leakage and non-leakage states using the optimized convolutional neural network under varying operating temperature. Our developed PIP leakage detection system can be deployed in safety-critical industrial systems for autonomous leakage detection.
CASE STUDIES IN THERMAL ENGINEERING
(2023)
Article
Mathematics
Ryul Kim, Young Hwan Choi
Summary: This research proposes a method that quantitatively evaluates the volume of leakage using deep learning technology and simultaneously detects the location of leakage through real-time monitoring. By using hydraulic data from a calibrated hydraulic model as training data and applying deep learning techniques, the study analyzes various scenarios regarding leakage volume and location to optimize leakage detection performance.
Article
Engineering, Multidisciplinary
Feng Li, Laibin Zhang, Shaohua Dong, Hang Zhang, Wenhe Wang, Yun You
Summary: This paper proposed a novel noise-pressure interaction model based on multi-information fusion processing, which can effectively filter interference signals and improve the accuracy and reliability of leakage detection.
Article
Environmental Sciences
Carlos Andres Macias Avila, Francisco-Javier Sanchez-Romero, P. Amparo Lopez-Jimenez, Modesto Perez-Sanchez
Summary: Water is a valuable resource, but high leakage rates in water distribution systems lead to maintenance costs and health risks. Different methods are used to estimate leakage levels and efficiency indicators are proposed. Analysis of cases where PATs reduce leaks and recover energy.
Article
Engineering, Environmental
Xinxin Yang, Xin Xu, Yisu Zhou, Yixin Yao, Chaofeng Shen, Jingqing Liu
Summary: Microplastics (MPs) are detected in both tap water and pipe scales samples collected from a typical drinking water distribution system (DWDS). The abundance and size of MPs vary along the DWDS with an increase in the stagnant-slow flow region and a decrease from upstream to downstream. The distribution of MPs is correlated with heavy metal concentration and dependent on the size and density of pipe scales.
JOURNAL OF HAZARDOUS MATERIALS
(2023)
Article
Environmental Sciences
Itai Arbiv, Husein Almuhtaram, Robert C. Andrews
Summary: This study compared the use of pipe loops to static pipe section reactors (PSRs) to assess chloramine decay. Unlined cast iron (UCI) and cement-lined ductile iron (CLDI) were compared to virgin polyvinyl chloride (PVC) pipe under different water velocities and hydraulic residence times. The results showed that pipe material had the greatest impact on chloramine decay, followed by flow velocity. The use of PSRs was found to be a viable and cost-effective alternative for assessing disinfectant decay.
SCIENCE OF THE TOTAL ENVIRONMENT
(2023)
Article
Environmental Sciences
Masaaki Kitajima, Mercedes C. Cruz, Rohan B. H. Williams, Stefan Wuertz, Andrew J. Whittle
Summary: This study analyzed microbial communities in biofilm and water samples collected from a DWDS where monochloramine is used as a residual disinfectant. The study revealed differences in microbial species and abundance in different pipe sections representing different water ages, providing novel insights into the microbial ecology of DWDS.
SCIENCE OF THE TOTAL ENVIRONMENT
(2021)
Article
Chemistry, Multidisciplinary
Sang Soo Lee, Ho-Hyun Lee, Yun-Jung Lee
Summary: A water supply enhancement project is being carried out in South Korea to reduce pipeline leakages and provide stable tap water. By establishing a District Metered Area and using IoT sensors, the minimum night flow can be monitored and predicted. The LSTM model performed better than the MLP model in predicting the minimum night flow.
APPLIED SCIENCES-BASEL
(2022)
Article
Water Resources
Priyanshu Jain, Ruchi Khare
Summary: An optimization procedure based on the Rao algorithm is proposed to determine the optimum locations for energy recovery and leakage reduction in water distribution networks. The method utilizes microturbines or pumps-as-turbines to reduce leakage and produce energy. Results show that the proposed approach converges to the best solution and can be applied to any pipe network regardless of size and location.
JOURNAL OF WATER AND CLIMATE CHANGE
(2023)
Article
Engineering, Multidisciplinary
Xudong Fan, Xiong (Bill) Yu
Summary: Despite significant progress in detecting and localizing leakage in underground water distribution networks, the challenge lies in the lack of labeled data under leaking conditions, making it difficult to utilize common machine learning models.
STRUCTURAL HEALTH MONITORING-AN INTERNATIONAL JOURNAL
(2022)