Identification of leakage in water supply pipelines based on time-frequency convolutional neural network

doi:10.3785/j.issn.1008-973X.2025.01.019

Journal of ZheJiang University (Engineering Science)

2025, Vol. 59

Issue (1): 196-204 DOI: 10.3785/j.issn.1008-973X.2025.01.019

Identification of leakage in water supply pipelines based on time-frequency convolutional neural network

Lingxuan LAI(

),Jingqing LIU*(

),Yisu ZHOU,Xiujuan LI

College of Civil Engineering and Architecture, Zhejiang University, Hangzhou 310058, China

Download:

HTML

PDF(2092KB) HTML
Export: BibTeX | EndNote (RIS)

Abstract

A leakage pressure identification model in the water supply pipeline was proposed to resolve problems that the existing studies of leakage detection failed to determine the pressure of leakage. The proposed model was based on the short-time Fourier transform with window parameter optimization and convolutional neural network. Pipeline sound signals were collected from a full-scale pipeline leakage pilot experimental platform. Leakage conditions of the same leak area but different shapes were subjected to short-time Fourier transform processing to obtain two-dimensional time-frequency spectra. The spectra contained characteristics related to leakage at three different pressure levels and were used as input for the convolutional neural network. The leakage pressure identification model was constructed based on the optimization of window parameters and network hyperparameters. Experimental results showed that the proposed model achieved an overall accuracy of 95.2%, and the recognition of high-pressure, middle-pressure and low-pressure conditions were 93.5%, 92.9% and 92.4% respectively. The proposed model has a higher accuracy in identifying leakage and pressure of leakage than traditional machine learning models, and the model can be used for identifying leakage pressure levels in real water supply networks.

Key words： water supply pipeline leakage pressure identification sound signal monitoring short-time Fourier transform convolutional neural network window parameters optimization

Received: 16 January 2024 Published: 18 January 2025

CLC:

TU 991

Corresponding Authors: Jingqing LIU E-mail: 22112072@zju.edu.cn;liujingqing@zju.edu.cn

	Service
	E-mail this article
	Add to my bookshelf
	Add to citation manager
	E-mail Alert
	RSS
	Articles by authors
	Lingxuan LAI
	Jingqing LIU
	Yisu ZHOU
	Xiujuan LI

Cite this article:

Lingxuan LAI,Jingqing LIU,Yisu ZHOU,Xiujuan LI. Identification of leakage in water supply pipelines based on time-frequency convolutional neural network. Journal of ZheJiang University (Engineering Science), 2025, 59(1): 196-204.

URL:

https://www.zjujournals.com/eng/10.3785/j.issn.1008-973X.2025.01.019 OR https://www.zjujournals.com/eng/Y2025/V59/I1/196

基于时频卷积神经网络的供水管道漏损识别

现有供水管道漏损检测研究不能判断漏损压力，为此依托全尺寸管网漏损中试实验平台采集管道声信号，提出基于窗参数优化的短时傅里叶变换和卷积神经网络的供水管道漏损压力识别模型. 对于相同面积、不同形状的漏口，采用短时傅里叶变换处理声信号，得到包含三压力水平下漏损特征的二维时频谱图，作为卷积神经网络的输入. 在窗参数和网络超参数优化的基础上，构建漏损压力识别模型. 实验结果表明：所提模型总体识别准确率为95.2%，高、中、低压漏损工况识别准确率为93.5%、92.9%、92.4%；相比传统机器学习模型，所提模型识别漏损和压力准确率更高，可用于实际供水管网的漏损压力识别.

关键词： 供水管道, 漏损压力识别, 声信号监测, 短时傅里叶变换, 卷积神经网络, 窗参数优化

Fig.1 Schematic diagram of water supply pipeline leakage experimental platform

Fig.2 Signal waveform diagram for a leak free condition

Tab.1 Partition of leakage condition identification dataset

Fig.3 Time-frequency characteristics of sound signals from different conditions of water supply pipeline

Fig.4 Time-frequency spectrum of different leakage conditions in water supply pipelines

Tab.2 Design of time-frequency spectra parameters and model accuracy

Fig.5 Time-frequency spectra with different window parameters

Fig.6 Water supply pipeline leakage identification model based on shot-time Founer transform and convolutional neural network

Tab.3 Parameters of water supply pipeline leakage identification model based on short-time Fourier transform and convolutional neural network

Tab.4 Optimization design of hyperparameter combination

Fig.7 Accuracy of different hyperparameter combinations on datasets

Fig.8 Performance of water supply pipeline leakage identification model based on shot-time Fourier transform and convolutional neural network

Tab.5 Results of modular ablation experiment

Tab.6 Performance comparison of different classification models

Fig.9 Recall and precision of classification model


[1]	AHOPELTO S, VAHALA R Cost-benefit analysis of leakage reduction methods in water supply networks[J]. Water, 2020, 12 (1): 195 doi: 10.3390/w12010195

[2]	SARKAMARYAN S, HAGHIGHI A, ADIB A Leakage detection and calibration of pipe networks by the inverse transient analysis modified by Gaussian functions for leakage simulation[J]. Journal of Water Supply: Research and Technology-Aqua, 2018, 67 (4): 404- 413 doi: 10.2166/aqua.2018.176

[3]	COLOMBO A F, KARNEY B W Energy and costs of leaky pipes: towards comprehensive picture[J]. Journal of Water Resources Planning and Management, 2002, 128 (6): 441- 450 doi: 10.1061/(ASCE)0733-9496(2002)128:6(441)

[4]	ZHU L J, WANG D M, YUE J K, et al Leakage detection method of natural gas pipeline combining improved variational mode decomposition and Lempel-Ziv complexity analysis[J]. Transactions of the Institute of Measurement and Control, 2022, 44 (15): 2865- 2876 doi: 10.1177/01423312221088080

[5]	XIE J, ZHANG Y B, HE Z Y, et al Automated leakage detection method of pipeline networks under complicated backgrounds by combining infrared thermography and Faster R-CNN technique[J]. Process Safety and Environmental Protection, 2023, 174: 39- 52 doi: 10.1016/j.psep.2023.04.006

[6]	AMRAN T S T, ISMAIL M P, AHMAD M R, et al. Detection of underground water distribution piping system and leakages using ground penetrating radar (GPR) [C]// Advancing Nuclear Science and Engineering for Sustainable Nuclear Energy Knowledge . [S. l.]: AIP Publishing, 2017, 1799: 030004.

[7]	GONG W J, SURESH M A, SMITH L, et al Mobile sensor networks for optimal leak and back flow detection and localization in municipal water networks[J]. Environmental Modelling and Software, 2016, 80: 306- 321 doi: 10.1016/j.envsoft.2016.02.001

[8]	QI Z X, ZHENG F F, GUO D L, et al Better understanding of the capacity of pressure sensor systems to detect pipe burst within water distribution networks[J]. Journal of Water Resources Planning and Management, 2018, 144 (7): 04018035 doi: 10.1061/(ASCE)WR.1943-5452.0000957

[9]	GAO Y, LIU Y Y Theoretical and experimental investigation into structural and fluid motions at low frequencies in water distribution pipes[J]. Mechanical Systems and Signal Processing, 2017, 90: 126- 140 doi: 10.1016/j.ymssp.2016.12.018

[10]	ADEDEJI K B, HAMAM Y, ABE B T, et al Towards achieving a reliable leakage detection and localization algorithm for application in water piping networks: an overview[J]. IEEE Access, 2017, 5 (10): 20272- 20285

[11]	LU H F, LSELEY T, BEHBAHANI S, et al Leakage detection techniques for oil and gas pipelines: state-of-the-art[J]. Tunnelling and Underground Space Technology, 2020, 98: 103249 doi: 10.1016/j.tust.2019.103249

[12]	ZHOU M F, XU Y Z, CUI B H, et al Pipeline leakage detection and localization using a reliable pipeline-mechanism model incorporating a Bayesian model updating approach[J]. Water, 2022, 14 (8): 1255 doi: 10.3390/w14081255

[13]	EI-ZAHAB S, ASAAD A, ABDELKADER E M, et al Collective thinking approach improving leak detection systems[J]. Smart Water, 2016, 2: 3 doi: 10.1186/s40713-017-0007-9

[14]	KANG J, PARK Y J, LEE J, et al Novel leakage detection by ensemble CNN-SVM and graph-based localization in water distribution systems[J]. IEEE Transactions on Industrial Electronics, 2018, 65 (5): 4279- 4289 doi: 10.1109/TIE.2017.2764861

[15]	XU Z H, MA W, LIN P, et al Deep learning of rock images for intelligent lithology identification[J]. Computers and Geosciences, 2021, 154: 104799 doi: 10.1016/j.cageo.2021.104799

[16]	朱丹宸, 张永祥, 潘洋洋, 等基于多传感器信号和卷积神经网络的滚动轴承故障诊断[J]. 振动与冲击, 2020, 39 (4): 172- 178 ZHU Danchen, ZHANG Yongxiang, PAN Yangang, et al Fault diagnosis for rolling element bearings based on multi-sensor signals and CNN[J]. Journal of Vibration and Shock, 2020, 39 (4): 172- 178

[17]	GUO G C, YU X P, LIU S M, et al Leakage detection in water distribution systems based on time-frequency convolutional neural network[J]. Journal of Water Resources Planning and Management, 2021, 147 (2): 04020101 doi: 10.1061/(ASCE)WR.1943-5452.0001317

[18]	IOFFE S, SZEGEDY C. Batch normalization: accelerating deep network training by reducing internal covariate shift [C]// International Conference on Machine Learning . San Diego: MIT Press, 2015, 37: 448–456.

[19]	KAMALAKKANNAN R, ROOPKUMAR R, ZAYED A Short time coupled fractional fourier transform and the uncertainty principle[J]. Fractional Calculus and Applied Analysis, 2021, 24: 667- 688 doi: 10.1515/fca-2021-0029

[20]	ZHOU X, TANG Z H, XU W R, et al. Deep learning identifies accurate burst locations in water distribution networks [J]. Water Research , 2019, 166: 115058.

[21]	赵林硕, 叶郭煊, 申永刚, 等基于时频卷积神经网络的供水管道漏损检测[J]. 中国给水排水, 2023, 39 (17): 53- 58 ZHAO Linshuo, YE Guoxuan, SHEN Yonggang, et al Leakage detection of water supply pipeline based on time-frequency convolutional neural network[J]. China Water and Wastewater, 2023, 39 (17): 53- 58

[22]	BUTTERFIELD J D, KRYNKIN A, COLLINS R P, et al Experimental investigation into vibro-acoustic emission signal processing techniques to quantify leak flow rate in plastic water distribution pipes[J]. Applied Acoustics, 2017, 199: 146- 155

[23]	JALIL M, BUTT F A, MALIK A. Short-time energy, magnitude, zero crossing rate and autocorrelation measurement for discriminating voiced and unvoiced segments of speech signals [C]// 2013 IEEE International Conference on Technological Advances in Electrical, Electronics and Computer Engineering . Konya: IEEE, 2013: 208–212.

[24]	CHUANG W Y, TSAI Y L, WANG L H. Leak detection in water distribution pipes based on CNN with Mel-frequency cepstral coefficients [C]// The 3rd International Conference on Innovation in Artificial Intelligence . Suzhou: [s.n.], 2019: 83–86.

[1]	Haijun WANG,Tao WANG,Cijun YU. CFRP ultrasonic detection defect identification method based on recursive quantitative analysis[J]. Journal of ZheJiang University (Engineering Science), 2024, 58(8): 1604-1617.

[2]	Jinye LI,Yongqiang LI. Spatial-temporal multi-graph convolution for traffic flow prediction by integrating knowledge graphs[J]. Journal of ZheJiang University (Engineering Science), 2024, 58(7): 1366-1376.

[3]	Zhiwei XING,Shujie ZHU,Biao LI. Airline baggage feature perception based on improved graph convolutional neural network[J]. Journal of ZheJiang University (Engineering Science), 2024, 58(5): 941-950.

[4]	Mingjun SONG,Wen YAN,Yizhao DENG,Junran ZHANG,Haiyan TU. Light-weight algorithm for real-time robotic grasp detection[J]. Journal of ZheJiang University (Engineering Science), 2024, 58(3): 599-610.

[5]	Xiaofeng FU,Weiqi CHEN,Yao SUN,Yuze PAN. Bimodal software classification model based on bidirectional encoder representation from transformer[J]. Journal of ZheJiang University (Engineering Science), 2024, 58(11): 2239-2246.

[6]	Zhiqiang GENG,Wei CHEN,Bo MA,Yongming HAN. Bearing intelligent fault diagnosis method based on continuous wavelet convolutional neural network[J]. Journal of ZheJiang University (Engineering Science), 2024, 58(10): 2069-2075.

[7]	Zhao-yang SONG,Xiao-qiang ZHAO,Yong-yong HUI,Hong-mei JIANG. Image super-resolution reconstruction algorithm based on multi-level continuous encoding and decoding[J]. Journal of ZheJiang University (Engineering Science), 2023, 57(9): 1885-1893.

[8]	Dian-hai WANG,Rui XIE,Zheng-yi CAI. Prediction of urban interrupted traffic flow based on optimal convergence time interval[J]. Journal of ZheJiang University (Engineering Science), 2023, 57(8): 1607-1617.

[9]	Wei QUAN,Yong-qing CAI,Chao WANG,Jia SONG,Hong-kai SUN,Lin-xuan LI. VR sickness estimation model based on 3D-ResNet two-stream network[J]. Journal of ZheJiang University (Engineering Science), 2023, 57(7): 1345-1353.

[10]	Xin-dong LV,Jiao LI,Zhen-nan DENG,Hao FENG,Xin-tong CUI,Hong-xia DENG. Structured image super-resolution network based on improved Transformer[J]. Journal of ZheJiang University (Engineering Science), 2023, 57(5): 865-874.

[11]	Jian-zhao ZHANG,Ji-chang GUO,Yu-dong WANG. Underwater image enhancement algorithm via fusing reverse medium transmission map[J]. Journal of ZheJiang University (Engineering Science), 2023, 57(5): 921-929.

[12]	Chuan-hua ZHOU,Li-chun CAO,Jia-yi ZHOU,Feng ZHAN. Identification of critical nodes in temporal networks based on graph convolution union computing[J]. Journal of ZheJiang University (Engineering Science), 2023, 57(5): 930-938.

[13]	Jun HE,Ya-sheng ZHANG,Can-bin YIN. Operating modes identification of spaceborne SAR based on deep learning[J]. Journal of ZheJiang University (Engineering Science), 2022, 56(8): 1676-1684.

[14]	Ren-peng MO,Xiao-sheng SI,Tian-mei LI,Xu ZHU. Bearing life prediction based on multi-scale features and attention mechanism[J]. Journal of ZheJiang University (Engineering Science), 2022, 56(7): 1447-1456.

[15]	Zhu-peng WEN,Jie CHEN,Lian-hua LIU,Ling-ling JIAO. Fault diagnosis of wind power gearbox based on wavelet transform and improved CNN[J]. Journal of ZheJiang University (Engineering Science), 2022, 56(6): 1212-1219.

Viewed

Full text

Abstract

Cited

Shared

Discussed