2023
DOI: 10.3390/s23063226
|View full text |Cite
|
Sign up to set email alerts
|

Pipeline Leakage Detection Using Acoustic Emission and Machine Learning Algorithms

Abstract: Pipelines play a significant role in liquid and gas resource distribution. Pipeline leaks, however, result in severe consequences, such as wasted resources, risks to community health, distribution downtime, and economic loss. An efficient autonomous leakage detection system is clearly required. The recent leak diagnosis capability of acoustic emission (AE) technology has been well demonstrated. This article proposes a machine learning-based platform for leakage detection for various pinhole-sized leaks using t… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
3
1
1

Citation Types

0
11
0

Year Published

2023
2023
2024
2024

Publication Types

Select...
8
1

Relationship

0
9

Authors

Journals

citations
Cited by 34 publications
(11 citation statements)
references
References 29 publications
(29 reference statements)
0
11
0
Order By: Relevance
“…The algorithm presented exhibits high classification accuracy in detecting leaks of different sizes and fluid pressures. Ullah Niamat [62] and others put forward a platform based on machine learning, which is used to detect leaks with various pinhole sizes by using the channel information of acoustic emission sensors. Statistical measures are extracted from acoustic emission signals as the characteristics of the training machine learning model, and finally, the overall classification accuracy is 99%, which provides reliable and effective results suitable for implementing the proposed platform.…”
Section: Sound Sensormentioning
confidence: 99%
“…The algorithm presented exhibits high classification accuracy in detecting leaks of different sizes and fluid pressures. Ullah Niamat [62] and others put forward a platform based on machine learning, which is used to detect leaks with various pinhole sizes by using the channel information of acoustic emission sensors. Statistical measures are extracted from acoustic emission signals as the characteristics of the training machine learning model, and finally, the overall classification accuracy is 99%, which provides reliable and effective results suitable for implementing the proposed platform.…”
Section: Sound Sensormentioning
confidence: 99%
“…Yussif et al present a study that proposes a low-cost approach to locating leakages in urban water distribution networks using acoustic signal behavior and machine learning, achieving high validation accuracy with the developed models [28]. Ullah et al propose a machine-learning-based platform for detecting pipeline leaks of various pinhole sizes using acoustic emission sensor channel information and achieves an exceptional overall classification accuracy of 99% [29].…”
Section: Related Workmentioning
confidence: 99%
“…Natural gas has a low explosion limit, and the leaking gas is flammable and explosive when it reaches a certain concentration, which can easily cause major safety accidents that seriously threaten the lives and properties of the general public [14,15]. The current mainstream gas pipeline leak identification method is outside the pipeline detection, but outside the pipeline, detection excavation cost is high and can only identify the size of the larger leak points [16][17][18][19] and cannot meet the requirement of 1 mm minimum identifiable leak size for DN100 buried gas pipeline. Therefore, the design of a microleakage image recognition method applicable to the detection inside small, buried gas pipelines is of great significance to ensure people's safety and enhance economic efficiency.…”
Section: Introductionmentioning
confidence: 99%