Abstract:In addition to the issues of water quality and economic losses, leakages in pipe networks are a potential danger to public safety. Early detection of water leaks in pipe networks is crucial for implementing countermeasures to reduce structural damage and water losses. The method of data collection based on sensors installed in pipeline networks has gained a lot of attention due to its potential for application in real-time monitoring systems for leak detection. However, many models developed on data collected from simulations, engineered tests or field experiments, and few have been validated using real network data. To make further investigations in real water distribution systems (WDSs), machine learning (ML)-based leak detection models were developed and validated using signal data collected by piezoelectric accelerometers installed in WDSs over several cities of China. Nonlinear features describing the non-stationary characteristics contained in leak signals are extracted from the signals to build feature sets as inputs to the common classifiers, i.e., support vector machine (SVM), decision tree (DT), and K-nearest neighbor (KNN). In addition, the spectrograms obtained by the short-time Fourier transform processing are used for the convolutional neural network SqueezeNet. The effects of different features, different dataset compositions, and different models for leak detection are compared and analyzed. SqueezeNet has the best performance with 95.15% accuracy in identifying leaks, and KNN is the best of the three classifiers with superior sensitivity and 88.17% accuracy in identification. This paper demonstrates the effectiveness and practicability of using machine learning models to identify leaks in real pipe networks. The application of machine learning models can perform effective leak detection with minimal human intervention, thus providing a meaningful contribution to the study of leak detection in real WDSs. The proposed model aims to identify leakages under operational conditions in real-life pipe networks with higher efficiency and reduced manual involvement. This paper demonstrates the applicability of accelerometers based on microelectromechanical systems for leak detection and establishes real network-based machine learning models; thereby, providing meaningful contribution to the literature regarding leak detection in a real WDS.

Real-time Detection of Urban Gas Pipeline Leakage Based on Machine Learning of IoT Time-Series Data

In-Situ Leak Detection in Urban Water Supply Pipelines Using Flexible Vibration Sensors

Pipeline leakage detection for district heating systems using multisource data in mid- and high-latitude regions

A Real-Time Method to Detect the Leakage Location in Urban Water Distribution Networks

Deep Learning Approach for Detection of Underground Natural Gas Micro-Leakage Using Infrared Thermal Images

Leakage detection of low-pressure gas distribution pipeline system based on linear fitting and extreme learning machine

Leak Detection in Water Distribution Systems by Classifying Vibration Signals

An unsupervised leak detection method with aggregating prediction and reconstruction along projection pathway for natural gas gathering pipelines

Research on Traceability Method of Gas Leak for Emergency Auxiliary Support

City Gas Pipeline Leakage Detection Based on Infrared Imaging Studies

Enhanced Deep Learning Method for Natural Gas Pipeline Flow Prediction Based on Integrated Learning

A new leak recognition method for natural gas pipelines in the urban underground space based on probabilistic neural network

Detection of underground natural gas pipeline micro-leakage based on UAV hyperspectral remote sensing and GIS

Hierarchical Leak Detection and Localization Method in Natural Gas Pipeline Monitoring Sensor Networks

Gas Leakage Recognition Based on Wide-Band Infrared Imaging with the Auxiliary Excitation Method and Machine Learning Model

Leakage Detection Approach of Natural Gas Pipelines Based on MCMC Algorithm

A Real-Time Early Warning Classification Method for Natural Gas Leakage Based on Random Forest

Leakage diagnosis of natural gas pipeline based on multi-source heterogeneous information fusion

Online Monitoring System for Natural Gas Leakage in Cities with Sensor Network

Pipeline Leakage Detection Using Acoustic Emission and Machine Learning Algorithms

Early Detection of Pipeline Natural Gas Leakage from Hyperspectral Imaging by Vegetation Indicators and Deep Neural Networks