Acoustic Identification of Water Supply Pipe Leakage Based on Bispectrum Analysis

Journal of Pipeline Systems Engineering and Practice, Journal Year: 2025, Volume and Issue: 16(3)

Published: April 29, 2025

Language: Английский

---

Parallel multi-layer sensor fusion for pipe leak detection using multi-sensors and machine learning

Engineering Applications of Artificial Intelligence, Journal Year: 2025, Volume and Issue: 153, P. 110923 - 110923

Published: April 26, 2025

Language: Английский

---

Real-Time Pipeline Leak Detection: A Hybrid Deep Learning Approach Using Acoustic Emission Signals

Applied Sciences, Journal Year: 2024, Volume and Issue: 15(1), P. 185 - 185

Published: Dec. 28, 2024

This study introduces an advanced deep-learning framework for the real-time detection of pipeline leaks in smart city infrastructure. The methodology transforms acoustic emission (AE) signals from time domain into scalogram images using continuous wavelet transform (CWT) to enhance leak-related features. A Gaussian filter minimizes background noise and clarifies these features further. core combines convolutional neural networks (CNNs) with long short-term memory (LSTM), ensuring a comprehensive examination both spatial temporal AE signals. genetic algorithm (GA) optimizes network by isolating most important leak detection. final classification stage uses fully connected categorize health conditions as either ‘leak’ or ‘non-leak’. Experimental validation on real-world data demonstrated framework’s efficacy, achieving accuracy rates 99.69%. approach significantly advances capabilities monitoring maintenance, offering durable scalable solution proactive infrastructure management.

Language: Английский

---

Adaptive Real-time Leak Detection in Water Distribution Systems Using Online Learning

Published: May 16, 2024

In this paper, we address the critical challenge of real-time leak detection in water distribution systems using online learning algorithms. The data collected by accelerometers was exploited to identify distinctive characteristics leaks. Our study focuses exclusively on Online Gradient Boosting Machines (Online GBM) method following preprocessing. analysis reveals that GBM model, optimised through random search for its hyperparameters, excels detection, achieving an accuracy 92.30%. These results, obtained a test set, demonstrate effectiveness managing large sets and reliability as rapid tool. article highlights significant potential techniques, particularly GBM, enhancing resource management effectively reducing losses due results research offer promising path towards improving monitoring maintenance infrastructure.

Language: Английский

---

Water Leakage Classification With Acceleration, Pressure, and Acoustic Data: Leveraging the Wavelet Scattering Transform, Unimodal Classifiers, and Late Fusion

IEEE Access, Journal Year: 2024, Volume and Issue: 12, P. 84923 - 84951

Published: Jan. 1, 2024

Language: Английский

---

Novel EMD with Optimal Mode Selector, MFCC, and 2DCNN for Leak Detection and Localization in Water Pipeline

Applied Sciences, Journal Year: 2023, Volume and Issue: 13(23), P. 12892 - 12892

Published: Dec. 1, 2023

Significant water loss caused by pipeline leaks emphasizes the importance of effective leak detection and localization techniques to minimize wastage. All state-of-the-art approaches use deep learning (DL) for cross-correlation localization. The existing methods’ complexity is very high, as they detect localize using two different architectures. This paper aims present an independent architecture with a single sensor detecting localizing enhanced performance. proposed approach combines novel EMD optimal mode selector, MFCC, two-dimensional convolutional neural network (2DCNN). suggested technique uses acousto-optic data from real-time setup in UTAR, Malaysia. collected are noisy, redundant, one-dimensional time series. So, must be denoised prepared before being fed 2DCNN selector denoises series identifies desired IMF. IMF passed MFCC then leak. assessment criteria employed this study prediction accuracy, precision, recall, F-score, R-squared. helps validate method’s detection-only credibility. also implements variants show EMD’s algorithm. reliability cross-verified cross-correlation. findings demonstrate that surpasses alternative methods methods. method gives 99.99% accuracy across all metrics. 99.54%, precision 98.92%, recall 98.86%, F-score 98.89%, R-square 99.09%.

Language: Английский

---

Separation Anomaly Distribution with a Triplet Network for Leakage Detection

Journal of Digital Contents Society, Journal Year: 2024, Volume and Issue: 25(5), P. 1273 - 1279

Published: May 31, 2024

Language: Английский

---

Research on the Loss Rule of the Leakage Problem in Residential Construction Based on Water Spray and Storage Tests

Buildings, Journal Year: 2024, Volume and Issue: 14(11), P. 3386 - 3386

Published: Oct. 25, 2024

Leakage issues have received increasing attention as the most common and significant source of complaints in residential construction quality problems. In this study, based on classification leakage problems, 1947 water spray tests 2333 storage were conducted 18 projects. An empirical analysis 432 cases was to determine loss law for a single point well laws different grades Through analysis, it can be concluded that more than 90% problems are third-level. To better understand quantitative problem, total model developed. Finally, is summarized, measures reduce proposed. This research provide theoretical basis tools inherent defect insurance help companies control risks drive promotion insurance.

Language: Английский

---