Burst Area Identification of Water Supply Network by Improved DenseNet Algorithm with Attention Mechanism

Cheng, Jing; Peng, Sen; Cheng, Rui; Wu, Xingqi; Xu, Fang

doi:10.1007/s11269-022-03316-9

Cited by 5 publications

(3 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Zhou et al [19] inputs the image signals, which have been denoised using improved spline-local mean decomposition, into a CNN network to predict the location of leak points. Cheng et al [20] proposed a DenseNet network based on attention mechanism, which detects leaks by monitoring system pressure. Song and Li [21] separately used the time-domain signals and the time-frequency images obtained from wavelet transformation as inputs for a CNN to accomplish leak detection work.…”

Section: Introductionmentioning

confidence: 99%

Liquid-filled pipeline leak detection and localization based on multi-scale residual networks

Zhao,

Zhou,

Liu

et al. 2024

Meas. Sci. Technol.

View full text Add to dashboard Cite

Effective ways to improve the accuracy of liquid-filled pipeline leak detection are one of the key issues that need to be addressed urgently in a conservation-oriented society. Recently, pipeline leak detection methods based on deep learning have developed rapidly. To improve the learning ability of convolutional neural network for pipeline leak signal features and leak detection accuracy, a Multi-Scale Residual Networks (MSRN) model is proposed in this paper for liquid-filled pipeline leak detection and localization. The model uses convolutional kernels of different scales to extract multiscale features of pipeline leakage signals based on deep residual networks (DRNs) and uses fully connected layers to fuse the features, thus improving the accuracy of pipeline leakage detection and localization. Among them, the large convolution kernel can acquire the low-frequency information of the signal due to its sizeable perceptual field, the medium convolution kernel can capture the local and global features of the signal, and the small convolution kernel is more sensitive to the high-frequency information of the signal. Meanwhile, a pipeline leakage test platform is built to evaluate the proposed model. The test results show that the accuracy of leak detection and localization of MSRN model is 98.3%, which is better than that of single-scale DRN model. In addition, the proposed MSRN model is verified to have good generalization and noise immunity through testing and analyzing the leakage signals under different pressures and background noises.

show abstract

Section: Introductionmentioning

confidence: 99%

Liquid-filled pipeline leak detection and localization based on multi-scale residual networks

Zhao,

Zhou,

Liu

et al. 2024

Meas. Sci. Technol.

View full text Add to dashboard Cite

show abstract

“…For example, Zhou et al [12] proposed a burst localization framework based on FL-DenseNet, and after optimizing the placement of pressure sensors, the model can effectively identify the location of a pipe burst from potential areas (e.g., District Metering Areas). Cheng et al [13] built a burst localization framework, first dividing the WDN into monitoring areas based on hydraulic characteristics caused by burst events, then identifying and localizing bursts with a fully linear DenseNet model enhanced by an attention mechanism and Bayesian hyperparameter optimization. A subsequent study replaced FL-DenseNet with FL-ResNet, obtaining comparable average accuracy but with less training time [14].…”

Section: Introductionmentioning

confidence: 99%

Burst Diagnosis Multi-Stage Model for Water Distribution Networks Based on Deep Learning Algorithms

Peng,

Wang,

Fang

et al. 2024

Water

Self Cite

View full text Add to dashboard Cite

Pipe bursts in water distribution networks (WDNs) pose significant threats to the safety of distribution networks, driving attention to deep learning-based burst detection and localization. However, the applicability of different pressure features still needs to be compared and verified. A large number of nodes challenges deep learning with the excessive number of classification categories and low recognition accuracy. To address these problems, this paper extracts different burst pressure features, including pressure value, pressure difference, and pressure fluctuation ratio, and inputs one of these features into a Burst Diagnosis Multi-Stage Model (BDMM) based on three CS-LSTMs (a combination of the Cuckoo Search algorithm and a long short-term memory network). The first model addresses a binary classification problem, outputting labels indicating whether a pipe burst has occurred. The second one solves a multi-classification problem, outputting the label of the burst partition, and the third model also solves a multi-classification problem, outputting the ID of the bursting junction. The model is tested on a real network and outperforms ELM. For basic burst identification tasks using CS-LSTM, differences among the three features are minimal, while pressure difference and pressure fluctuation ratio exhibit superior performance to pressure value when resolving more complex problems like burst junction localization.

show abstract

“…To accommodate such challenges, it becomes necessary to increase the number of pipes branching from a location closer to the central water distribution unit [17]. As a result, municipalities must be even more vigilant about maintaining these additional pipes as their numbers increase.…”

mentioning

confidence: 99%

Water distribution pipe lifespans: Predicting when to repair the pipes in municipal water distribution networks using machine learning techniques

Farajzadeh,

Sadeghzadeh,

Jokar

2024

PLOS Water

View full text Add to dashboard Cite

Water is one of the essential matters that keeps living species alive; yet, the lifespan of pipes has two direct impacts on wasting water in very great amounts: pipe leakages and pipe bursts. Consequently, the proper detection of aged pipes in the water distribution networks has always been an issue in overcoming the problem. This makes water pipe monitoring an important duty of municipalities. Traditionally, leakages and bursts were only detected visually or through reports in local areas, leading municipalities to change the old pipes. Although this helps to fix the issue, a more desired way is to perhaps let officials know about the possibilities of such problems in advance by predicting which pipes are aged, so they can prevent the wastage. Therefore, to automate the detection process, in this study, we take the initial steps to predict the pipes needing repair in a particular area using machine learning methods. We first obtain a private dataset provided by the municipality of Saveh, Iran which outlines pipes that were damaged previously. We then train three machine learning algorithms to predict whether a set of pipes in an area is prone to damage. To achieve this, One-Class (OC) Classification methods such as OC-SVM, Isolation Forest, and Elliptic Envelope are used and they achieved the highest accuracy of 0.909. This study is of value since it requires zero additional devices (i.e., sensors).

show abstract

Burst Area Identification of Water Supply Network by Improved DenseNet Algorithm with Attention Mechanism

Cited by 5 publications

References 24 publications

Liquid-filled pipeline leak detection and localization based on multi-scale residual networks

Liquid-filled pipeline leak detection and localization based on multi-scale residual networks

Burst Diagnosis Multi-Stage Model for Water Distribution Networks Based on Deep Learning Algorithms

Water distribution pipe lifespans: Predicting when to repair the pipes in municipal water distribution networks using machine learning techniques

Contact Info

Product

Resources

About