Acoustic Emission Signal Detection for Internal Valve Leakage in Liquid‐Filled Pipelines Using Kernel Principal Component Analysis.
Saved in:
| Title: | Acoustic Emission Signal Detection for Internal Valve Leakage in Liquid‐Filled Pipelines Using Kernel Principal Component Analysis. |
|---|---|
| Authors: | Zhang, Runlin1 (AUTHOR), He, Xueming1 (AUTHOR), Wu, Chao1 (AUTHOR), Wang, Lei2 (AUTHOR), Wu, Di1 (AUTHOR) zhrl202508@163.com, Biswas, Arnab (AUTHOR) arnbiswas@wiley.com |
| Source: | Shock & Vibration. 4/27/2026, Vol. 2026, p1-12. 12p. |
| Subjects: | Acoustic emission, Principal components analysis, Pipeline transportation, Dimensional reduction algorithms, Signal detection, Feature extraction, Time-frequency analysis |
| Abstract: | To detect internal valve leakage in liquid‐filled pipelines, a method using kernel principal component analysis (KPCA) is proposed to analyze acoustic emission (AE) signals for leakage detection. Time‐frequency analysis and root mean square (RMS) values of AE signals across various frequency bands reveal that the dominant frequency band for valve leakage signals consistently falls within 160–180 kHz across the tested pressure and leakage rate ranges. Using these findings, key feature parameters were selected and subjected to dimensionality reduction via principal component analysis (PCA). A kernel function was integrated to enhance the efficacy of feature extraction for leakage detection. The results show that KPCA effectively distinguishes background noise from leakage signals when appropriate feature parameters are selected. The integration of the kernel function increased the explained variance ratio of the first principal component from 77.84% to 88.75% compared to conventional PCA. Evaluation of multiple metrics confirms that the KPCA method with a sigmoid kernel achieves optimal performance in processing AE signal feature parameters. [ABSTRACT FROM AUTHOR] |
| Copyright of Shock & Vibration is the property of Wiley-Blackwell and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.) | |
| Database: | Engineering Source |
|
Full text is not displayed to guests.
Login for full access.
|
|
| Abstract: | To detect internal valve leakage in liquid‐filled pipelines, a method using kernel principal component analysis (KPCA) is proposed to analyze acoustic emission (AE) signals for leakage detection. Time‐frequency analysis and root mean square (RMS) values of AE signals across various frequency bands reveal that the dominant frequency band for valve leakage signals consistently falls within 160–180 kHz across the tested pressure and leakage rate ranges. Using these findings, key feature parameters were selected and subjected to dimensionality reduction via principal component analysis (PCA). A kernel function was integrated to enhance the efficacy of feature extraction for leakage detection. The results show that KPCA effectively distinguishes background noise from leakage signals when appropriate feature parameters are selected. The integration of the kernel function increased the explained variance ratio of the first principal component from 77.84% to 88.75% compared to conventional PCA. Evaluation of multiple metrics confirms that the KPCA method with a sigmoid kernel achieves optimal performance in processing AE signal feature parameters. [ABSTRACT FROM AUTHOR] |
|---|---|
| ISSN: | 10709622 |
| DOI: | 10.1155/vib/3367991 |
