Article
Engineering, Multidisciplinary
Jingpin Jiao, Jiawei Zhang, Yubao Ren, Guanghai Li, Bin Wu, Cunfu He
Summary: In this paper, acoustic emission signals are analyzed using a sparse representation method to extract the main components associated with pipeline leaks. Dictionary learning is performed to estimate the main leakage components, and cross-correlation analysis is used to determine the leak location. Experimental results demonstrate that the proposed method effectively improves the signal-to-noise ratio and enhances the accuracy and reliability of pipeline leak location.
Article
Engineering, Multidisciplinary
Hongjin Liu, Hongyuan Fang, Xiang Yu, Fuming Wang, Xuan Yang, Shaohui Li
Summary: Leak location studies of water supply pipes are crucial for reducing water waste and eliminating potential hazards. In this paper, an adaptive time-delay estimation method (CEENDAN-CC) is proposed, which can extract feeble leakage signals and improve the accuracy and noise immunity of time-delay estimation.
Article
Engineering, Electrical & Electronic
Lin Mei, Shuaiyong Li, Chao Zhang, Mingxiu Han
Summary: An adaptive signal enhancement method based on genetic algorithm optimized VMD and SVD is proposed in this study to address the low SNR issue in leak location in water-supply pipelines. Experimental results demonstrate that the proposed method is effective in reducing leak location errors.
IEEE SENSORS JOURNAL
(2021)
Article
Mathematics
Vincent F. Yu, Grace Aloina, Hadi Susanto, Mohammad Khoirul Effendi, Shih-Wei Lin
Summary: Municipal waste management is a challenging issue, especially in developing countries. This research proposes a new model to determine depot locations and vehicle routes for waste collection in each region, considering government policy requirements, and aims to fulfill the collection needs at a minimum cost. The effectiveness of the proposed method is demonstrated through numerical examples using actual data.
Article
Acoustics
Fangli Ning, Zhanghong Cheng, Di Meng, Juan Wei
Summary: The paper presents a new framework combining the acoustic features extraction method and Random Forest algorithm for gas pipeline leak detection and classification. By extracting acoustic features and using the Random Forest algorithm for detection and classification, the framework achieves effective condition monitoring of industrial equipment, improving accuracy and applicability.
Article
Computer Science, Artificial Intelligence
Arun Kumar Sangaiah, Raheleh Khanduzi
Summary: Designing a reliable hub network for transporting goods has become crucial, with a focus on hub location, protective measures, and risk assessment. This study introduces a model to optimize transportation cost through primary and backup hubs, utilizing metaheuristic algorithms such as tabu search and simulated annealing. Experimental results show superior performance of these algorithms in solving the complex problem, with a proposed hybrid approach showing promising results in sensitivity analysis and real-world applications.
APPLIED SOFT COMPUTING
(2022)
Article
Engineering, Civil
Hongjin Liu, Hongyuan Fang, Xiang Yu, Fuming Wang, Yangyang Xia
Summary: This study proposes an empirical mode decomposition and cross-correlation (EMDCC) method for water pipeline leak localization, which adaptively extracts effective leak signals from low signal-to-noise ratio (SNR) detection signals, and significantly improves the accuracy of localization.
Article
Engineering, Multidisciplinary
Qiansheng Fang, Haojie Wang, Chenlei Xie, Jie Chen
Summary: This paper proposes a leak location method for water supply pipelines based on a multivariate variational mode decomposition algorithm, which can locate the leakage point more accurately with an average relative positioning error of less than 2.2%.
CMES-COMPUTER MODELING IN ENGINEERING & SCIENCES
(2022)
Article
Environmental Sciences
Yongxin Shen, Weiping Cheng
Summary: This study develops and compares three machine learning models for assisting leak detection operations in water distribution systems. By conducting numerous on-site tests and analyzing signal features, it was found that certain features can effectively distinguish leakage signals from non-leakage signals.
Article
Computer Science, Artificial Intelligence
Xuguang Hu, Huaguang Zhang, Dazhong Ma, Rui Wang, Tianbiao Wang, Xiangpeng Xie
Summary: Traditional leak location methods often yield inaccurate results due to the uncertainty of pressure change points. This article proposes an adaptive dynamic programming approach to address this issue. By introducing a pipeline model, a value iteration scheme, and neural networks, the proposed method provides real-time leak location for long-distance pipelines.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2023)
Article
Automation & Control Systems
Xuguang Hu, Huaguang Zhang, Dazhong Ma, Rui Wang, Pengfei Tu
Summary: In this article, a data-driven method for small leak location is proposed, which accurately locates the leak point through the construction of a pipeline model and heuristic dynamic programming method, and demonstrates satisfactory performance in field tests.
IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS
(2022)
Article
Computer Science, Information Systems
Tuoru Li, Senxiang Lu, Enjie Xu
Summary: The paper proposes a method to locate internal detectors in pipelines using vibration signals extracted by the VMD algorithm and classified by machine learning classifiers. This method shows high accuracy and adaptability in experimental research.
Article
Chemistry, Analytical
Niamat Ullah, Zahoor Ahmed, Jong-Myon Kim
Summary: Pipelines are important for distributing liquid and gas resources, but leaks can lead to resource waste, health risks, distribution downtime, and economic loss. This article proposes a machine learning-based platform that uses acoustic emission (AE) technology to detect pinhole-sized leaks. Statistical features extracted from the AE signal are used to train machine learning models. The proposed platform achieves an exceptional overall classification accuracy of 99% for detecting leaks and pinhole-sized leaks.
Article
Thermodynamics
Peifeng Lin, Donghui Lei, Jiang Liao
Summary: In this study, experimental and numerical methods are employed to locate pipeline leakage, with particular consideration to the influence of curvature radius on the accuracy of leak location. Results show that smaller curvature radius leads to lower accuracy, while larger curvature radius or increased inlet pressure improves accuracy. It is found that the elbow with a curvature radius three to four times of pipe diameter provides the most accurate measurement of leakage location.
ADVANCES IN MECHANICAL ENGINEERING
(2021)
Article
Engineering, Electrical & Electronic
Xianming Lang
Summary: The traditional method of negative pressure wave for locating pipeline leaks requires accurate acquisition of the inflection point of pressure signal, otherwise the leaks cannot be located; a fusion signal-based leak localization method is proposed, which locates leaks based on pressure and flow signal fusion, reducing the localization error effectively.
IEEE SENSORS JOURNAL
(2021)