《油气储运》

【目的】阀门是保障天然气管道系统稳定运行的重要元件，为了实现天然气阀门早期泄漏的实时监测及缺陷识别，亟需建立一种可检测并识别不同微小泄漏缺陷的有效方法。【方法】利用自主搭建的阀门微小泄漏声发射检测实验装置，设置0.2MPa、0.4MPa及0.6MPa共3种压力，采用声发射检测方法采集了5种阀门状态（健康阀门、阀杆腐蚀外漏缺陷、阀门封闭不严内漏缺陷、阀芯划伤内漏缺陷及法兰垫片破损外漏缺陷）的声发射信号，并得到其时频域特征矩阵。通过主成分分析方法（Principal Component Analysis,PCA）将复杂、冗余的时频域特征矩阵转化为二维、三维特征因子矩阵，计算得到各个特征因子矩阵的最佳簇数，并通过K-means、K-medoids聚类算法对二维、三维特征因子矩阵进行聚类分析。【结果】对于较低压力（0.2MPa）下天然气阀门微小泄漏缺陷的识别，三维PCA聚类效果达到97.8％，优于二维PCA；随着压力增至0.6MPa，二维PCA聚类效果不断优化，并逐渐超过三维PCA。在聚类方法评价中，K-medoids聚类算法的稳定程度优于K-means聚类算法，且K-medoids聚类算法的运算效率更有优势。【结论】基于声发射技术+特征提取+PCA+聚类算法的声发射信号分类处理方法，提高了阀门微小泄漏检测的准确率与稳定性，实现了天然气管道阀门缺陷的分类识别，可为天然气管道系统的安全运行提供检测手段及保障措施。（图7，表2，参[26]）

[Objective] The pivotal role valves play in ensuring the stable operation of natural gas pipeline systems necessitates real-time monitoring and defect identification to eliminate early leakage in natural gas valves. To this end, it is imperative to develop an effective method for detecting and identifying micro-leakage defects. [Methods] An acoustic emission detection experimental setup was initially established targeting valve micro-leakage, with three pressure settings of 0.2 MPa, 0.4 MPa, and 0.6 MPa. This self-developed setup was utilized to capture the acoustic emission signals across five valve conditions (healthy state, external leakage due to valve stem corrosion, internal leakage due to untightness in valve closure, internal leakage due to valve core scratch, and external leakage due to flange gasket damage) through the acoustic emission detection method. The collected signals were then used to create time-frequency domain feature matrices. These intricate and redundant time-frequency domain feature matrices were transformed into two-dimensional and three-dimensional feature factor matrices utilizing Principal Component Analysis (PCA). The optimal number of clusters for each feature factor matrix was determined through subsequent calculations. The two-dimensional and three-dimensional feature factor matrices were further analyzed using K-means and K-medoids clustering methods. [Results] For the identification of micro-leakage defects of natural gas valves under low pressure conditions (0.2 MPa), the clustering effect of 3D PCA reached 97.8%, surpassing that of 2D PCA. As the pressure was gradually raised to 0.6 MPa, the clustering effect of 2D PCA continued to improve, eventually surpassing that of 3D PCA. In the evaluation of clustering methods, it was observed that the K-medoids method exhibited greater stability and higher operational efficiency compared to the K-means method. [Conclusion] The classified processing method for acoustic emission signals, based on acoustic emission technology, feature extraction, PCA, and clustering algorithms, can enhance the accuracy and stability of valve micro-leakage detection. Moreover, it enables the classified identification of defects in natural gas pipeline valves. This approach not only offers monitoring means but also establishes safeguard measures crucial for ensuring the secure operation of natural gas pipeline systems. (7 Figures, 2 Tables, 26 References)

引言
1 基本方法及原理
2 声发射信号采集
3 缺陷识别处理
4 结论

图1 天然气管道阀门微小泄漏时频特征体系图Fig. 1 Time-frequency domain feature system for micro-leakage of natural gas pipeline valves

图2 天然气管道阀门微小泄漏声发射检测实验装置示意图Fig. 2 Experimental setup of acoustic emission detection for micro-leakage in natural gas pipeline valves

图3 压力为0.2 MPa下健康阀门与阀芯划伤缺陷时频域对比图Fig. 3 Time-frequency domain comparison between healthy valve and valve with core scratch at 0.2 MPa

表1 压力0.2 MPa下健康阀门声发射特征矩阵表Table 1 Acoustic emission feature matrix of healthy valve at 0.2 MPa

图4 不同压力下天然气管道阀门5种状态的PCA降维结果对比图Fig. 4 Pressure-varying comparison of PCA dimensionality reduction results of natural gas pipeline valves across five conditions

图5 不同压力下天然气管道阀门簇内误差平方和结果对比图Fig. 5 Pressure-varying comparison of cluster sum of squared errors for natural gas pipeline valve

图6 压力为0.2 MPa下天然气管道阀门状态K-means与K-medoids聚类效果对比图Fig. 6 Comparison of K-means and K-medoids clustering effects across natural gas pipeline valve conditions at 0.2 MPa

图7 压力0.6 MPa下天然气管道阀门状态K-means与K-medoids聚类效果对比图Fig. 7 Comparison of K-means and K-medoids clustering effects across natural gas pipeline valve conditions at 0.6 MPa

表2 不同压力下天然气管道阀门两种聚类结果对比表Table 2 Comparison of two clustering results for natural gas pipeline valves under different pressures

[1] 熊毅，高萍，赵潇，赵立康.天然气管道泄漏应急抢修过程的动态AHP风险评价[J].油气储运，2021，40（12）：1423-1429. 10.6047/j.issn.1000-8241.2021.12.014. XIONG Y，GAO P，ZHAO X，ZHAO L K. Dynamic AHP-based risk assessment for emergency repair of natural gas pipeline leakage[J]. Oil & Gas Storage and Transportation，2021，40(12):1423-1429.
[2] 李振林，毕治强，刘刚，夏广辉，张海峰，李洪星.基于声发射的天然气管道球阀内漏检测[J].油气储运，2013，32（6）：647-650. 10.6047/j.issn.1000-8241.2013.06.019. LI Z L，BI Z Q，LIU G，XIA G H，ZHANG H F，LI H X. Detection of internal leakage of ball valves in gas pipeline based on acoustic emission theory[J]. Oil & Gas Storage and Transportation，2013，32(6):647-650.
[3] LIU G J，CHEN Y Q，ZHANG X L，JIANG W X，XIE Y C，TIAN X J，et al. A deep residual shrinkage network based on multi-scale attention module for subsea Christmas tree valve leakage detection[J]. Measurement，2022，198:110970. DOI:10.1016/j.measurement.2022.110970.
[4] LYU B K，XU D，LI L F，NISHIMURA A，LIU H M，ZHANG H C，et al. A 20 K cryogen-free leak detection system for cryogenic valves by using a GM cryocooler[J]. Cryogenics，2021，117:103332. DOI:10.1016/j.cryogenics.2021.103332.
[5] XIA G H，GOU W J. The design of globe valve innerleakage detection system based on acoustic emission in natural gas pipeline[J]. IOP Conference Series:Earth and Environmental Science，2020，450(1):012083. DOI:10.1088/1755-1315/450/1/012083.
[6] SIM H Y，RAMLI R，SAIFIZUL A. Reciprocating compressor valve damage estimation under varying speeds through the acoustic emission technique[J]. International Journal of Structural Integrity，2019，10(5):621-633. DOI:10.1108/IJSI-03-2019-0024.
[7] SHI M J，LIANG Y B，QIN L S，ZHENG Z，HUANG Z Q. Prediction method of ball valve internal leakage rate based on acoustic emission technology[J]. Flow Measurement and Instrumentation，2021，81:102036. DOI:10.1016/j.flowmeasinst.2021.102036.
[8] YE G Y，XU K J，WU W K. Mixed multiple-variable modelling of acoustic emission signals for valve internal leakage detection[J]. IET Science Measurement & Technology，2021，15(6):487-498. DOI:10.1049/smt2.12049.
[9] YE G Y，XU K J，WU W K. Multi-variable classification model for valve internal leakage based on acoustic emission time-frequency domain characteristics and random forest[J]. Review of Scientific Instruments，2021，92(2):025108. DOI:10.1063/5.0024611.
[10] YE G Y，XU K J，WU W K. Multivariable modeling of valve inner leakage acoustic emission signal based on Gaussian process[J]. Mechanical Systems and Signal Processing，2020，140:106675. DOI:10.1016/j.ymssp.2020.106675.
[11] ZHANG Y Y，YANG G L，ZHANG D H，WANG T. Investigation on recognition method of acoustic emission signal of the compressor valve based on the deep learning method[J]. Energy Reports，2021，7(S7):62-71. DOI:10.1016/j.egyr.2021.10.053.
[12] KAEWWAEWNOI W，PRATEEPASEN A，KAEWTRAKULPONG P. Investigation of the relationship between internal fluid leakage through a valve and the acoustic emission generated from the leakage[J]. Measurement，2010，43(2):274-282. DOI:10.1016/j.measurement.2009.10.005.
[13] SIM H Y，RAMLI R，SAIFIZUL A，SOONG M F. Detection and estimation of valve leakage losses in reciprocating compressor using acoustic emission technique[J]. Measurement，2020，152:107315. DOI:10.1016/j.measurement.2019.107315.
[14] ZHU S B，LI Z L，ZHANG S M，YING Y Y，ZHANG H F. Deep belief network-based internal valve leakage rate prediction approach[J]. Measurement，2019，133:182-192. DOI:10.1016/j.measurement.2018.10.020.
[15] LIU Y K，ZHOU W，AYODEJI A，ZHOU X Q，PENG M J，CHAO N. A multi-layer approach to DN 50 electric valve fault diagnosis using shallow-deep intelligent models[J]. Nuclear Engineering and Technology，2021，53(1):148-163. DOI:10.1016/j.net.2020.07.001.
[16] ZHU L，ZOU B，GAO S H，JIANG M，LI Z P. Research on gate valve internal leakage detection based on acoustic emission signal processing[J]. Advanced Materials Research，2014，1037:169-173. DOI:10.4028/www.scientific.net/AMR.1037.169.
[17] SIM H Y，RAMLI R，SAIFIZUL A A，ABDULLAH M A K. Empirical investigation of acoustic emission signals for valve failure identification by using statistical method[J]. Measurement，2014，58:165-174. DOI:10.1016/j.measurement.2014.08.028.
[18] FAN H，TARIQ S，ZAYED T. Acoustic leak detection approaches for water pipelines[J]. Automation in Construction，2022，138:104226. DOI:10.1016/j.autcon.2022.104226.
[19] 吴同，邓忠华，沈亮，谢祝祝，陈煜凯，刘翠伟，等.长距离输油管道泄漏监测技术研究进展[J].油气储运，2023，42（3）：259-275. 10.6047/j.issn.1000-8241.2023.03.003. WU T，DENG Z H，SHEN L，XIE Z Z，CHEN Y K，LIU C W，et al. Research progress of long-distance oil pipeline leakage monitoring technology[J]. Oil & Gas Storage and Transportation，2023，42(3):259-275.
[20] 张海峰，杨淼，张一峰，王雪莉，王建伟，李东阳.基于小波包熵的天然气管道阀门内漏分析方法[J].油气储运，2017，36（11）：1308-1314. 10.6047/j.issn.1000-8241.2017.11.013. ZHANG H F，YANG M，ZHANG Y F，WANG X L，WANG J W，LI D Y. A new analysis method for the internal leakage of gas pipeline valves based on wavelet packet entropy[J]. Oil &Gas Storage and Transportation，2017，36(11):1308-1314.
[21] 徐长航，刘鹏谦，李振兴，杜莎莎，张源.基于声发射信号的气液两相流管道泄漏识别方法[J].油气储运，2021，40（10）：1131-1137. 10.6047/j.issn.1000-8241.2021.10.007. XU C H，LIU P Q，LI Z X，DU S S，ZHANG Y. Leakage identification method of gas-liquid two-phase flow pipeline based on acoustic emission signal[J]. Oil & Gas Storage and Transportation，2021，40(10):1131-1137.
[22] LI Z L，ZHANG H F，TAN D J，CHEN X，LEI H X. A novel acoustic emission detection module for leakage recognition in a gas pipeline valve[J]. Process Safety and Environmental Protection，2017，105:32-40. DOI:10.1016/j.psep.2016.10.005.
[23] ZHU S B，LI Z L，ZHANG S M，LIANG L L，ZHANG H F. Natural gas pipeline valve leakage rate estimation via factor and cluster analysis of acoustic emissions[J]. Measurement，2018，125:48-55. DOI:10.1016/j.measurement.2018.04.076.
[24] 刘真，潘文菊，刘佳，温凯，宫敬.基于K-means聚类＋灰色理论＋BP神经网络的区域天然气长期需求预测[J].油气储运，2024，43（1）：103-110. 10.6047/j.issn.1000-8241.2024.01.012. LIU Z，PAN W J，LIU J，WEN K，GONG J. A combined approach for forecasting regional long-term natural gas demand integrating K-means clustering，grey theory，and BP neural network[J]. Oil & Gas Storage and Transportation，2024，43(1):103-110.
[25] 吴猛猛，高艳，明岳，李海斌，钟小雁，陈明辉.基于声比拟的压力管路阀门内漏噪声数值预报研究[J].声学技术，2023，42（2）：234-241. 10.16300/j.cnki.1000-3630.2023.02.017. WU M M，GAO Y，MING Y，LI H B，ZHONG X Y，CHEN M H. Numerical prediction of internal leakage noise in pressure pipeline valve based on acoustic analogy[J]. Technical Acoustics，2023，42(2):234-241.
[26] 赵锦鹏，周文静，白云龙，张永海，魏进家.基于MCMC算法的天然气管道泄漏检测方法[J].油气储运，2024，43（1）：49-56. 10.6047/j.issn.1000-8241.2024.01.006. ZHAO J P，ZHOU W J，BAI Y L，ZHANG Y H，WEI J J. Leakage detection approach of natural gas pipelines based on MCMC algorithm[J]. Oil & Gas Storage and Transportation，2024，43(1):49-56.

备注

引言

1 基本方法及原理

2 声发射信号采集

3 缺陷识别处理

4 结论