Article(id=1156986625868784228, tenantId=1146029695717560320, journalId=1146123166801305609, issueId=1156963927277003616, articleNumber=null, orderNo=null, doi=10.12404/j.issn.1671-1815.2402034, pmid=null, cstr=null, oa=null, hot=null, price=null, onlineType=0, articleFormat=0, articleType=null, articleTypeStr=null, receivedDate=1710950400000, receivedDateStr=2024-03-21, revisedDate=1734278400000, revisedDateStr=2024-12-16, acceptedDate=null, acceptedDateStr=null, onlineDate=1753776707378, onlineDateStr=2025-07-29, pubDate=1742227200000, pubDateStr=2025-03-18, doiRegisterDate=null, doiRegisterDateStr=null, onlineIssueDate=1753776707378, onlineIssueDateStr=2025-07-29, onlineJustAcceptDate=null, onlineJustAcceptDateStr=null, onlineFirstDate=null, onlineFirstDateStr=null, sourceXml=null, magXml=null, createTime=1753776707378, creator=13701087609, updateTime=1753776707378, updator=13701087609, issue=Issue{id=1156963927277003616, tenantId=1146029695717560320, journalId=1146123166801305609, year='2025', volume='25', issue='8', pageStart='3079', pageEnd='3528', issueExtLink='null', onlineDate='null', pubDate='null', beforeIssueId=null, nextIssueId=null, price=null, status=1, issueComplete=1, articleOrder=1, issueType=-1, specialIssue=0, createTime=1753771295613, creator=13701087609, updateTime=1753777038876, updator=13701087609, preIssue=null, nextIssue=null, ext={EN=IssueExt(id=1156988016305726153, tenantId=1146029695717560320, journalId=1146123166801305609, issueId=1156963927277003616, language=EN, specialIssueTitle=, coverIllustrator=, specialIssueEditor=, specialIssueAbout=), CN=IssueExt(id=1156988016305726154, tenantId=1146029695717560320, journalId=1146123166801305609, issueId=1156963927277003616, language=CN, specialIssueTitle=, coverIllustrator=, specialIssueEditor=, specialIssueAbout=)}, issueFiles=null}, startPage=3296, endPage=3303, ext={EN=ArticleExt(id=1156986627806552687, articleId=1156986625868784228, tenantId=1146029695717560320, journalId=1146123166801305609, language=EN, title=Identification and Location of Leakage Aperture of Underground Water Supply Pipeline in Coal Mine, columnId=1156963928489157478, journalTitle=Science Technology and Engineering, columnName=Automation and Computational Technology, runingTitle=null, highlight=null, articleAbstract=

In order to quickly identify the location of the leakage point and the leak aperture in the coal mine, a model was proposed for identifying and locating the leak aperture by using the pressure and flow signals generated when the water supply pipeline leaked. Modal energy entropy and genetic algorithm combined with envelope entropy were used to optimize the parameters of variational mode decomposition (VMD), and then VMD was used to denoise the pressure signal. Convolutional neural network (CNN) was used to extract the deep feature sequence of pressure and flow signal, and the long short-term memory network (LSTM) was used to extract the time sequence of deep feature sequence to identify and locate the leak aperture. The experimental results show that compared with Kalman filter, mean value filter and low-pass filter, the variational modal decomposition with optimized parameters has higher root-mean-square error (RMSE), mean absolute error (MAE), signal-to-noise ratio (SNR) and normalized cross correlation (NCC), which indicates that it can effectively reduce noise components and retain effective signals. Compared with LSTM, the MAE, mean absolute percentage error (MAPE) and RMSE of CNN-LSTM in leak location decrease by 65.97%, 61.22% and 59.11%. In the identification of leak aperture, MAE decreases by 12.04%, MAPE decreases by 22.45%, and RMSE decreases by 3.29%, which proves that CNN-LSTM can make full use of the spatial and temporal characteristics of pipeline pressure and flow signals to identify the leak location and aperture, and its detection effect is more accurate and stable than LSTM.

, correspAuthors=Zhen CHEN, authorNote=null, correspAuthorsNote=null, copyrightStatement=null, copyrightOwner=null, extLink=null, articleAbsUrl=null, sourceXml=null, magXml=null, pdfUrl=null, pdf=null, pdfFileSize=null, pdfExtLink=null, richHtmlUrl=null, mobilePdfUrl=null, reviewReport=null, pdfFirstPage=null, abstractGraph=null, abstractGraphContent=null, abstractVideo=null, citation=null, cebUrl=null, magXmlContent=null, mapNumber=null, authorCompany=null, fund=null, authors=null, authorsList=Jing-yi DU, Zhen CHEN, Jia-wei ZHANG, Chen LI, Rui GAO, Peng WANG), CN=ArticleExt(id=1156986809591882028, articleId=1156986625868784228, tenantId=1146029695717560320, journalId=1146123166801305609, language=CN, title=煤矿井下供水管道泄漏孔径识别与定位, columnId=1154013913436902243, journalTitle=科学技术与工程, columnName=自动化技术、计算机技术, runingTitle=null, highlight=null, articleAbstract=

为快速识别煤矿井下泄漏点的位置及泄漏孔径,利用供水管道泄漏时产生的压力及流量信号,提出一种泄漏孔径识别与定位模型。首先利用模态能量熵和遗传算法结合包络熵对变分模态分解(variational mode decomposition,VMD) 进行参数优化,再使用VMD 对压力信号进行降噪处理;采用卷积神经网络(convolutional neural networks, CNN) 提取压力及流量信号的深层特征序列,长短时记忆网络(long short- term memory, LSTM) 提取深层特征序列的时序特征,进行泄漏孔径识别与定位。实验结果表明:经过参数优化的变分模态分解,相较卡尔曼滤波、均值滤波、低通滤波在均方根误差(root mean square error,RMSE) 、平均绝对误差(mean absolute error, MAE) 、信噪比(signal to noise ratio, SNR) 、归一化互相关系数(normalized cross correlation, NCC) 上均有提高,表明其能够有效降低噪声成分,保留有效信号;CNN-LSTM 相较LSTM,在泄漏点定位中,MAE降低了65. 97% ,平均绝对百分比误差( mean absolute percentage error, MAPE) 降低了61. 22% ,RMSE 降低了59. 11% 。在泄漏孔径识别中,MAE 降低了12. 04% ,MAPE 降低了22. 45% ,RMSE 降低了3. 29% ,证明CNN-LSTM 可以充分利用管道压力及流量信号的空间及时间特征进行泄漏位置及孔径的识别,其检测效果相较LSTM 更加准确和稳定。

, correspAuthors=陈镇, authorNote=null, correspAuthorsNote=
*陈镇(1999—),男,汉族,安徽六安人,硕士研究生。研究方向:模式识别与嵌入式系统。E-mail:1745293122@qq.com。
, copyrightStatement=null, copyrightOwner=null, extLink=null, articleAbsUrl=null, sourceXml=wQiegXYryOeMYXgxM1PIBA==, magXml=x63RQJOWkOl20wbwtA8D4Q==, pdfUrl=null, pdf=7Gn7miDlJg78PTocFbb4gA==, pdfFileSize=null, pdfExtLink=null, richHtmlUrl=null, mobilePdfUrl=null, reviewReport=null, pdfFirstPage=null, abstractGraph=null, abstractGraphContent=null, abstractVideo=null, citation=null, cebUrl=null, magXmlContent=gdquUnyJUipFv0ycfk0vEw==, mapNumber=null, authorCompany=null, fund=null, authors=

杜京义(1965—),男,汉族,山东淄博人,博士,教授。研究方向:检测技术及其自动化。E-mail:517571853@qq.com。

, authorsList=杜京义, 陈镇, 张嘉伟, 李晨, 高瑞, 王鹏)}, authors=[Author(id=1156986812360122750, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, orderNo=0, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=517571853@qq.com, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1156986812423037313, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, authorId=1156986812360122750, language=EN, stringName=Jing-yi DU, firstName=Jing-yi, middleName=null, lastName=DU, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1 College of Electrical and Control Engineering Xi'an University of Science and Technology Xi'an 710054 China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1156986812481757571, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, authorId=1156986812360122750, language=CN, stringName=杜京义, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1 西安科技大学 电气与控制工程学院 西安 710054, bio={"content":"

杜京义(1965—),男,汉族,山东淄博人,博士,教授。研究方向:检测技术及其自动化。E-mail:517571853@qq.com。

"}, bioImg=null, bioContent=

杜京义(1965—),男,汉族,山东淄博人,博士,教授。研究方向:检测技术及其自动化。E-mail:517571853@qq.com。

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1156986812104270193, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, xref=1, ext=[AuthorCompanyExt(id=1156986812112658802, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, companyId=1156986812104270193, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 College of Electrical and Control Engineering Xi'an University of Science and Technology Xi'an 710054 China), AuthorCompanyExt(id=1156986812116853107, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, companyId=1156986812104270193, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 西安科技大学 电气与控制工程学院 西安 710054)])]), Author(id=1156986812536283526, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, orderNo=1, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=1745293122@qq.com, emailSecond=null, emailThird=null, correspondingAuthor=1, authorType=1, ext={EN=AuthorExt(id=1156986812653724040, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, authorId=1156986812536283526, language=EN, stringName=Zhen CHEN, firstName=Zhen, middleName=null, lastName=CHEN, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, *, address=1 College of Electrical and Control Engineering Xi'an University of Science and Technology Xi'an 710054 China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1156986812704055691, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, authorId=1156986812536283526, language=CN, stringName=陈镇, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, *, address=1 西安科技大学 电气与控制工程学院 西安 710054, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1156986812104270193, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, xref=1, ext=[AuthorCompanyExt(id=1156986812112658802, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, companyId=1156986812104270193, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 College of Electrical and Control Engineering Xi'an University of Science and Technology Xi'an 710054 China), AuthorCompanyExt(id=1156986812116853107, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, companyId=1156986812104270193, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 西安科技大学 电气与控制工程学院 西安 710054)])]), Author(id=1156986812754387342, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, orderNo=2, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1156986813840712097, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, authorId=1156986812754387342, language=EN, stringName=Jia-wei ZHANG, firstName=Jia-wei, middleName=null, lastName=ZHANG, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1 College of Electrical and Control Engineering Xi'an University of Science and Technology Xi'an 710054 China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1156986813895238052, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, authorId=1156986812754387342, language=CN, stringName=张嘉伟, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1 西安科技大学 电气与控制工程学院 西安 710054, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1156986812104270193, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, xref=1, ext=[AuthorCompanyExt(id=1156986812112658802, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, companyId=1156986812104270193, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 College of Electrical and Control Engineering Xi'an University of Science and Technology Xi'an 710054 China), AuthorCompanyExt(id=1156986812116853107, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, companyId=1156986812104270193, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 西安科技大学 电气与控制工程学院 西安 710054)])]), Author(id=1156986813966541224, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, orderNo=3, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1156986814042038698, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, authorId=1156986813966541224, language=EN, stringName=Chen LI, firstName=Chen, middleName=null, lastName=LI, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2 College of Communication and Information Engineering Xi'an University of Science and Technology Xi'an 710054 China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1156986814100758956, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, authorId=1156986813966541224, language=CN, stringName=李晨, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2 西安科技大学 通信与信息工程学院 西安 710054, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1156986812175573365, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, xref=2, ext=[AuthorCompanyExt(id=1156986812183961974, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, companyId=1156986812175573365, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 College of Communication and Information Engineering Xi'an University of Science and Technology Xi'an 710054 China), AuthorCompanyExt(id=1156986812188156279, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, companyId=1156986812175573365, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 西安科技大学 通信与信息工程学院 西安 710054)])]), Author(id=1156986814172062124, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, orderNo=4, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1156986814234976687, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, authorId=1156986814172062124, language=EN, stringName=Rui GAO, firstName=Rui, middleName=null, lastName=GAO, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1 College of Electrical and Control Engineering Xi'an University of Science and Technology Xi'an 710054 China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1156986814323057072, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, authorId=1156986814172062124, language=CN, stringName=高瑞, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1 西安科技大学 电气与控制工程学院 西安 710054, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1156986812104270193, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, xref=1, ext=[AuthorCompanyExt(id=1156986812112658802, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, companyId=1156986812104270193, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 College of Electrical and Control Engineering Xi'an University of Science and Technology Xi'an 710054 China), AuthorCompanyExt(id=1156986812116853107, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, companyId=1156986812104270193, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 西安科技大学 电气与控制工程学院 西安 710054)])]), Author(id=1156986814406943154, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, orderNo=5, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1156986814469857716, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, authorId=1156986814406943154, language=EN, stringName=Peng WANG, firstName=Peng, middleName=null, lastName=WANG, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=3, address=3 College of Mechanical Engineering Xi'an University of Science and Technology Xi'an 710054 China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1156986814515995062, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, authorId=1156986814406943154, language=CN, stringName=王鹏, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=3, address=3 西安科技大学 机械工程学院 西安 710054, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1156986812276236665, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, xref=3, ext=[AuthorCompanyExt(id=1156986812280430970, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, companyId=1156986812276236665, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3 College of Mechanical Engineering Xi'an University of Science and Technology Xi'an 710054 China), AuthorCompanyExt(id=1156986812284625275, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, companyId=1156986812276236665, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3 西安科技大学 机械工程学院 西安 710054)])])], keywords=[Keyword(id=1156986815879143883, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, language=EN, orderNo=1, keyword=variational mode decomposition (VMD)), Keyword(id=1156986815954641357, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, language=EN, orderNo=2, keyword=convolutional neural networks (CNN)), Keyword(id=1156986816009167311, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, language=EN, orderNo=3, keyword=long short-term memory (LSTM)), Keyword(id=1156986816059498960, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, language=EN, orderNo=4, keyword=modal energy entropy), Keyword(id=1156986816143385041, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, language=EN, orderNo=5, keyword=genetic algorithm (GA)), Keyword(id=1156986816202105298, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, language=EN, orderNo=6, keyword=envelope entropy), Keyword(id=1156986816260825556, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, language=CN, orderNo=1, keyword=变分模态分解(VMD)), Keyword(id=1156986816311157206, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, language=CN, orderNo=2, keyword=卷积神经网络(CNN)), Keyword(id=1156986816357294552, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, language=CN, orderNo=3, keyword=长短时记忆网络(LSTM)), Keyword(id=1156986816403431898, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, language=CN, orderNo=4, keyword=模态能量熵), Keyword(id=1156986816462152156, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, language=CN, orderNo=5, keyword=遗传算法(GA)), Keyword(id=1156986816512483806, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, language=CN, orderNo=6, keyword=包络熵)], refs=[Reference(id=1156986821969273405, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, doi=null, pmid=null, pmcid=null, year=2014, volume=62, issue=3, pageStart=531, pageEnd=544, url=null, language=null, rfNumber=[1], rfOrder=0, authorNames=Dragomiretskiy K, Zosso D, journalName=IEEE Transactions on Signal Process, refType=null, unstructuredReference=Dragomiretskiy K, Zosso D. Variational mode decomposition[J]. IEEE Transactions on Signal Process, 2014. 62(3): 531-544., articleTitle=Variational mode decomposition, refAbstract=null), Reference(id=1156986822023799358, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, doi=null, pmid=null, pmcid=null, year=2019, volume=36, issue=3, pageStart=69, pageEnd=73, url=null, language=null, rfNumber=[2], rfOrder=1, authorNames=王秀芳, 朱道鸿, 葛延良, journalName=压力容器, refType=null, unstructuredReference=王秀芳, 朱道鸿, 葛延良. 基于VMD-EV的天然气管道小泄漏信号去噪研究[J]. 压力容器, 2019. 36(3): 69-73., articleTitle=基于VMD-EV的天然气管道小泄漏信号去噪研究, refAbstract=null), Reference(id=1156986822103491135, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, doi=null, pmid=null, pmcid=null, year=2019, volume=36, issue=3, pageStart=69, pageEnd=73, url=null, language=null, rfNumber=[2], rfOrder=2, authorNames=Wang Xiufang, Zhu Daohong, Ge Yanliang, journalName=Pressure Vessel Technology, refType=null, unstructuredReference=Wang Xiufang, Zhu Daohong, Ge Yanliang. Research on denoising small leakage signal of natural gas pipeline based on VMD-EV[J]. Pressure Vessel Technology, 2019. 36(3): 69-73., articleTitle=Research on denoising small leakage signal of natural gas pipeline based on VMD-EV, refAbstract=null), Reference(id=1156986822174794304, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, doi=null, pmid=null, pmcid=null, year=2015, volume=49, issue=5, pageStart=73, pageEnd=81, url=null, language=null, rfNumber=[3], rfOrder=3, authorNames=唐贵基, 王晓龙, journalName=西安交通大学学报, refType=null, unstructuredReference=唐贵基, 王晓龙. 参数优化变分模态分解方法在滚动轴承早期故障诊断中的应用[J]. 西安交通大学学报, 2015. 49(5): 73-81., articleTitle=参数优化变分模态分解方法在滚动轴承早期故障诊断中的应用, refAbstract=null), Reference(id=1156986822233514561, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, doi=null, pmid=null, pmcid=null, year=2015, volume=49, issue=5, pageStart=73, pageEnd=81, url=null, language=null, rfNumber=[3], rfOrder=4, authorNames=Tang Guiji, Wang Xiaolong, journalName=Journal of Xi'an Jiaotong University, refType=null, unstructuredReference=Tang Guiji, Wang Xiaolong. Early fault diagnosis of rolling bearing with IVMD fusion singular value difference spectrum[J]. Journal of Xi'an Jiaotong University, 2015. 49(5): 73-81., articleTitle=Early fault diagnosis of rolling bearing with IVMD fusion singular value difference spectrum, refAbstract=null), Reference(id=1156986822279651906, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, doi=null, pmid=null, pmcid=null, year=2021, volume=41, issue=4, pageStart=86, pageEnd=93, url=null, language=null, rfNumber=[4], rfOrder=5, authorNames=张萍, 张文海, 赵新贺, journalName=噪声与振动控制, refType=null, unstructuredReference=张萍, 张文海, 赵新贺, 等. WOA-VMD算法在轴承故障诊断中的应用[J]. 噪声与振动控制, 2021. 41(4): 86-93., articleTitle=WOA-VMD算法在轴承故障诊断中的应用, refAbstract=null), Reference(id=1156986822367732291, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, doi=null, pmid=null, pmcid=null, year=2021, volume=41, issue=4, pageStart=86, pageEnd=93, url=null, language=null, rfNumber=[4], rfOrder=6, authorNames=Zhang Ping, Zhang Wenhai, Zhao Xinhe, journalName=Noise and Vibration Control, refType=null, unstructuredReference=Zhang Ping, Zhang Wenhai, Zhao Xinhe, et al. Application of WOA-VMD algorithm in bearing fault diagnosis[J]. Noise and Vibration Control, 2021. 41(4): 86-93., articleTitle=Application of WOA-VMD algorithm in bearing fault diagnosis, refAbstract=null), Reference(id=1156986822430646852, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, doi=null, pmid=null, pmcid=null, year=2020, volume=120, issue=2, pageStart=312, pageEnd=328, url=null, language=null, rfNumber=[5], rfOrder=7, authorNames=Qin W, Lü HC, Liu CL, journalName=Industrial Management & Data Systems, refType=null, unstructuredReference=Qin W, HC, Liu CL, et al. Remaining useful life prediction for lithium-ion batteries using particle filter and artificial neural network[J]. Industrial Management & Data Systems, 2020. 120(2): 312-328., articleTitle=Remaining useful life prediction for lithium-ion batteries using particle filter and artificial neural network, refAbstract=null), Reference(id=1156986822480978501, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, doi=null, pmid=null, pmcid=null, year=2017, volume=76, issue=18, pageStart=18749, pageEnd=18770, url=null, language=null, rfNumber=[6], rfOrder=8, authorNames=Kuang YT, Singh R, Singh S, journalName=Multimedia Tools and Applications, refType=null, unstructuredReference=Kuang YT, Singh R, Singh S, et al. A novel macroeconomic forecasting model based on revised multimedia assisted BP neural network model and ant Colony algorithmk[J]. Multimedia Tools and Applications, 2017. 76(18): 18749-18770., articleTitle=A novel macroeconomic forecasting model based on revised multimedia assisted BP neural network model and ant Colony algorithmk, refAbstract=null), Reference(id=1156986822569058886, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, doi=null, pmid=null, pmcid=null, year=2023, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[7], rfOrder=9, authorNames=刘源, journalName=null, refType=null, unstructuredReference=刘源. 基于1DCNN的天然气管道泄漏孔径实时识别[D]. 大庆: 东北石油大学, 2023., articleTitle=基于1DCNN的天然气管道泄漏孔径实时识别, refAbstract=null), Reference(id=1156986822623584840, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, doi=null, pmid=null, pmcid=null, year=2023, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[7], rfOrder=10, authorNames=Liu Yuan, journalName=null, refType=null, unstructuredReference=Liu Yuan. Real-time identification of natural gas pipeline leak aperture based on 1DCNN[D]. Daqing: Northeast Petroleum University, 2023., articleTitle=Real-time identification of natural gas pipeline leak aperture based on 1DCNN, refAbstract=null), Reference(id=1156986822673916489, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, doi=null, pmid=null, pmcid=null, year=2011, volume=11, issue=7, pageStart=1601, pageEnd=1604, url=null, language=null, rfNumber=[8], rfOrder=11, authorNames=罗伟, 张奎, 杨艳, journalName=科学技术与工程, refType=null, unstructuredReference=罗伟, 张奎, 杨艳. 输油管道泄漏信号去噪的研究[J]. 科学技术与工程, 2011. 11(7): 1601-1604., articleTitle=输油管道泄漏信号去噪的研究, refAbstract=null), Reference(id=1156986822732636746, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, doi=null, pmid=null, pmcid=null, year=2011, volume=11, issue=7, pageStart=1601, pageEnd=1604, url=null, language=null, rfNumber=[8], rfOrder=12, authorNames=Luo Wei, Zhang Kui, Yang Yan, journalName=Science Technology and Engineering, refType=null, unstructuredReference=Luo Wei, Zhang Kui, Yang Yan. Research on denoising of oil pipeline leakage signal[J]. Science Technology and Engineering, 2011. 11(7): 1601-1604., articleTitle=Research on denoising of oil pipeline leakage signal, refAbstract=null), Reference(id=1156986822782968395, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, doi=null, pmid=null, pmcid=null, year=2021, volume=21, issue=1, pageStart=144, pageEnd=151, url=null, language=null, rfNumber=[9], rfOrder=13, authorNames=纪健, 傅晓宁, 王毅, journalName=科学技术与工程, refType=null, unstructuredReference=纪健, 傅晓宁, 王毅, 等. 气液两相流管道泄漏流场特性数值模拟分析[J]. 科学技术与工程, 2021. 21(1): 144-151., articleTitle=气液两相流管道泄漏流场特性数值模拟分析, refAbstract=null), Reference(id=1156986822833300044, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, doi=null, pmid=null, pmcid=null, year=2021, volume=21, issue=1, pageStart=144, pageEnd=151, url=null, language=null, rfNumber=[9], rfOrder=14, authorNames=Ji Jian, Fu Xiaoning, Wang Yi et al, journalName=Science Technology and Engineering, refType=null, unstructuredReference=Ji Jian, Fu Xiaoning, Wang Yi et al. Numerical simulation analysis of leakage flow field characteristics in gas-liquid two-phase flow pipeline[J]. Science Technology and Engineering, 2021. 21(1): 144-151., articleTitle=Numerical simulation analysis of leakage flow field characteristics in gas-liquid two-phase flow pipeline, refAbstract=null), Reference(id=1156986822879437389, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, doi=null, pmid=null, pmcid=null, year=2020, volume=20, issue=33, pageStart=13660, pageEnd=13666, url=null, language=null, rfNumber=[10], rfOrder=15, authorNames=王俊, 封辉, 高琦, journalName=科学技术与工程, refType=null, unstructuredReference=王俊, 封辉, 高琦, 等. 埋地管道泄漏数值模拟分析[J]. 科学技术与工程, 2020. 20(33): 13660-13666., articleTitle=埋地管道泄漏数值模拟分析, refAbstract=null), Reference(id=1156986822925574734, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, doi=null, pmid=null, pmcid=null, year=2020, volume=20, issue=33, pageStart=13660, pageEnd=13666, url=null, language=null, rfNumber=[10], rfOrder=16, authorNames=Wang Jun, Feng Hui, Gao Qi, journalName=Science Technology and Engineering, refType=null, unstructuredReference=Wang Jun, Feng Hui, Gao Qi, et al. Numerical simulation analysis of buried pipeline leakage[J]. Science Technology and Engineering, 2020. 20(33): 13660-13666., articleTitle=Numerical simulation analysis of buried pipeline leakage, refAbstract=null), Reference(id=1156986822980100687, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, doi=null, pmid=null, pmcid=null, year=2014, volume=14, issue=26, pageStart=224, pageEnd=227, url=null, language=null, rfNumber=[11], rfOrder=17, authorNames=李宏, 申瑞琦, journalName=科学技术与工程, refType=null, unstructuredReference=李宏, 申瑞琦. 基于改进的经验模态分解方法的热网管道泄漏检测方法[J]. 科学技术与工程, 2014. 14(26): 224-227., articleTitle=基于改进的经验模态分解方法的热网管道泄漏检测方法, refAbstract=null), Reference(id=1156986823043015248, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, doi=null, pmid=null, pmcid=null, year=2014, volume=14, issue=26, pageStart=224, pageEnd=227, url=null, language=null, rfNumber=[11], rfOrder=18, authorNames=Li Hong, Shen Ruiqi, journalName=Science Technology and Engineering, refType=null, unstructuredReference=Li Hong, Shen Ruiqi. Pipeline leakage detection method of heat supply network based on improved empirical mode decomposition method[J]. Science Technology and Engineering, 2014. 14(26): 224-227., articleTitle=Pipeline leakage detection method of heat supply network based on improved empirical mode decomposition method, refAbstract=null), Reference(id=1156986823093346897, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, doi=null, pmid=null, pmcid=null, year=2022, volume=22, issue=12, pageStart=4797, pageEnd=4803, url=null, language=null, rfNumber=[12], rfOrder=19, authorNames=汪建新, 王恩浩, journalName=科学技术与工程, refType=null, unstructuredReference=汪建新, 王恩浩. 管道泄漏源定位中最优小波基变换[J]. 科学技术与工程, 2022. 22(12): 4797-4803., articleTitle=管道泄漏源定位中最优小波基变换, refAbstract=null), Reference(id=1156986823139484242, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, doi=null, pmid=null, pmcid=null, year=2022, volume=22, issue=12, pageStart=4797, pageEnd=4803, url=null, language=null, rfNumber=[12], rfOrder=20, authorNames=Wang Jianxin, Wang Enhao, journalName=Science Technology and Engineering, refType=null, unstructuredReference=Wang Jianxin, Wang Enhao. Optimal wavelet basis transformation in pipeline leakage source location[J]. Science Technology and Engineering, 2022. 22(12): 4797-4803., articleTitle=Optimal wavelet basis transformation in pipeline leakage source location, refAbstract=null), Reference(id=1156986823181427283, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, doi=null, pmid=null, pmcid=null, year=2022, volume=45, issue=16, pageStart=34, pageEnd=41, url=null, language=null, rfNumber=[13], rfOrder=21, authorNames=马广兴, 曲波, 常琛, journalName=电子测量技术, refType=null, unstructuredReference=马广兴, 曲波, 常琛, 等. 基于CNN的供热管道泄漏识别方法研究[J]. 电子测量技术, 2022. 45(16): 34-41., articleTitle=基于CNN的供热管道泄漏识别方法研究, refAbstract=null), Reference(id=1156986823231758932, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, doi=null, pmid=null, pmcid=null, year=2022, volume=45, issue=16, pageStart=34, pageEnd=41, url=null, language=null, rfNumber=[13], rfOrder=22, authorNames=Ma Guangxing, Qu Bo, Chang Chen, journalName=Electronic Measurement Technology, refType=null, unstructuredReference=Ma Guangxing, Qu Bo, Chang Chen, et al. Research on leakage identification method of Heating pipeline based on CNN[J]. Electronic Measurement Technology, 2022. 45(16): 34-41., articleTitle=Research on leakage identification method of Heating pipeline based on CNN, refAbstract=null)], funds=[Fund(id=1156986821885387322, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, awardId=2023-JC-YB-362, language=CN, fundingSource=陕西省自然科学基础研究计划(2023-JC-YB-362), fundOrder=null, country=null)], companyList=[AuthorCompany(id=1156986812104270193, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, xref=1, ext=[AuthorCompanyExt(id=1156986812112658802, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, companyId=1156986812104270193, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 College of Electrical and Control Engineering Xi'an University of Science and Technology Xi'an 710054 China), AuthorCompanyExt(id=1156986812116853107, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, companyId=1156986812104270193, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 西安科技大学 电气与控制工程学院 西安 710054)]), AuthorCompany(id=1156986812175573365, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, xref=2, ext=[AuthorCompanyExt(id=1156986812183961974, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, companyId=1156986812175573365, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 College of Communication and Information Engineering Xi'an University of Science and Technology Xi'an 710054 China), AuthorCompanyExt(id=1156986812188156279, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, companyId=1156986812175573365, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 西安科技大学 通信与信息工程学院 西安 710054)]), AuthorCompany(id=1156986812276236665, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, xref=3, ext=[AuthorCompanyExt(id=1156986812280430970, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, companyId=1156986812276236665, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3 College of Mechanical Engineering Xi'an University of Science and Technology Xi'an 710054 China), AuthorCompanyExt(id=1156986812284625275, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, companyId=1156986812276236665, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3 西安科技大学 机械工程学院 西安 710054)])], figs=[ArticleFig(id=1156986819909870098, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, language=EN, label=Fig. 1, caption=Leakage detection and location model, figureFileSmall=KyrsoUn/5A+h7iBX6jQYGw==, figureFileBig=e2YYRavWDc4OQKdxX1l3pw==, tableContent=null), ArticleFig(id=1156986819976978963, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, language=CN, label=图1, caption=泄漏孔径识别与定位模型

IMF$n$ 为第$n$ 个变分模态分解模态分量; Dropout 为随机失活层

, figureFileSmall=KyrsoUn/5A+h7iBX6jQYGw==, figureFileBig=e2YYRavWDc4OQKdxX1l3pw==, tableContent=null), ArticleFig(id=1156986820031504916, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, language=EN, label=Fig. 2, caption=Parameter optimization workflow, figureFileSmall=bjSEZd2qjJVkwIvCf1gg5Q==, figureFileBig=RPxkmsP81PfOcBKIArsd4g==, tableContent=null), ArticleFig(id=1156986820098613781, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, language=CN, label=图2, caption=参数寻优工作流程, figureFileSmall=bjSEZd2qjJVkwIvCf1gg5Q==, figureFileBig=RPxkmsP81PfOcBKIArsd4g==, tableContent=null), ArticleFig(id=1156986820169916950, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, language=EN, label=Fig. 3, caption=Three gate structures of LSTM, figureFileSmall=DiVSLwtuEuzuKspC/sUufQ==, figureFileBig=luVSsdilXaZj30NPpIzB1Q==, tableContent=null), ArticleFig(id=1156986820228637207, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, language=CN, label=图3, caption=LSTM 3 种门结构

${\mathbf{C}}_{t - 1}$ 为上一时刻的状态;${\mathbf{C}}_{t}$ 为当前时刻的状态;${\mathbf{h}}_{t - 1}$ 为前一时刻的隐藏状态信息;${\mathbf{h}}_{t}$ 为当前时刻隐藏状态;$\sigma$ 为 sigmoid 激活函数; tanh 为双曲正切激活函数

, figureFileSmall=DiVSLwtuEuzuKspC/sUufQ==, figureFileBig=luVSsdilXaZj30NPpIzB1Q==, tableContent=null), ArticleFig(id=1156986820295746072, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, language=EN, label=Fig. 4, caption=Leakage transient simulation calculation model, figureFileSmall=+n4rqNArfvx+yCG/9lJsKw==, figureFileBig=hEadmzfW/URcodVI0bkujw==, tableContent=null), ArticleFig(id=1156986820341883417, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, language=CN, label=图4, caption=泄漏瞬态仿真计算模型, figureFileSmall=+n4rqNArfvx+yCG/9lJsKw==, figureFileBig=hEadmzfW/URcodVI0bkujw==, tableContent=null), ArticleFig(id=1156986820408992282, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, language=EN, label=Fig. 5, caption=Original signal and noisy signal, figureFileSmall=TEdZqSG0SuQ4cpQpK8rEKA==, figureFileBig=vpBOP8MlbdQ8Y/KWATpgNg==, tableContent=null), ArticleFig(id=1156986820471906843, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, language=CN, label=图5, caption=原始信号与带噪信号, figureFileSmall=TEdZqSG0SuQ4cpQpK8rEKA==, figureFileBig=vpBOP8MlbdQ8Y/KWATpgNg==, tableContent=null), ArticleFig(id=1156986820534821404, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, language=EN, label=Fig. 6, caption=Energy change ratio iteration diagram, figureFileSmall=Vs5mEVfqQdFfigHAv+WxLQ==, figureFileBig=hB7GC4UKRbwLfKBWNaPRUw==, tableContent=null), ArticleFig(id=1156986820589347357, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, language=CN, label=图6, caption=能量变化比迭代图, figureFileSmall=Vs5mEVfqQdFfigHAv+WxLQ==, figureFileBig=hB7GC4UKRbwLfKBWNaPRUw==, tableContent=null), ArticleFig(id=1156986820648067614, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, language=EN, label=Fig. 7, caption=Each IMF component, figureFileSmall=05F7oGNH3ZC+GkaHNKGouw==, figureFileBig=x54Xa7KeNyn5tmrmZqnTng==, tableContent=null), ArticleFig(id=1156986820706787872, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, language=CN, label=图7, caption=各 IMF 分量, figureFileSmall=05F7oGNH3ZC+GkaHNKGouw==, figureFileBig=x54Xa7KeNyn5tmrmZqnTng==, tableContent=null), ArticleFig(id=1156986820765508129, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, language=EN, label=Fig. 8, caption=IMF spectrum diagram, figureFileSmall=yXSN5Rp7GvjnVENcSwBqnw==, figureFileBig=XSlWapUaMPPZfwbyqTdrqQ==, tableContent=null), ArticleFig(id=1156986820811645474, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, language=CN, label=图8, caption=各 IMF 频谱图, figureFileSmall=yXSN5Rp7GvjnVENcSwBqnw==, figureFileBig=XSlWapUaMPPZfwbyqTdrqQ==, tableContent=null), ArticleFig(id=1156986820861977123, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, language=EN, label=Fig. 9, caption=Reconstruction signal, figureFileSmall=jh7IOZ0MYFuTnEbPxL/MVg==, figureFileBig=Az1bSU/Zb2cXGWFxdiCb+g==, tableContent=null), ArticleFig(id=1156986820954251812, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, language=CN, label=图9, caption=重构信号, figureFileSmall=jh7IOZ0MYFuTnEbPxL/MVg==, figureFileBig=Az1bSU/Zb2cXGWFxdiCb+g==, tableContent=null), ArticleFig(id=1156986821004583461, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, language=EN, label=Fig. 10, caption=Algorithm evaluation radar chart, figureFileSmall=fuyEEs0tTH+EBRKbYzqOJg==, figureFileBig=LMNWZor6NlPG7vq07aCrHQ==, tableContent=null), ArticleFig(id=1156986821063303718, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, language=CN, label=图10, caption=算法评估雷达图, figureFileSmall=fuyEEs0tTH+EBRKbYzqOJg==, figureFileBig=LMNWZor6NlPG7vq07aCrHQ==, tableContent=null), ArticleFig(id=1156986821122023975, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, language=EN, label=Fig. 11, caption=Model loss iteration diagram, figureFileSmall=K6lIWeWkHc08BEzqAu15zw==, figureFileBig=CKck5rs0txVq45U6MUa12g==, tableContent=null), ArticleFig(id=1156986821172355624, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, language=CN, label=图11, caption=模型损失迭代图, figureFileSmall=K6lIWeWkHc08BEzqAu15zw==, figureFileBig=CKck5rs0txVq45U6MUa12g==, tableContent=null), ArticleFig(id=1156986821281407529, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, language=EN, label=Fig. 12, caption=Test set leak detection and location results, figureFileSmall=l+UfKTvSliJcANUBdJmCRQ==, figureFileBig=uL3URnOh8Diw24vUnjWUMw==, tableContent=null), ArticleFig(id=1156986821428208172, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, language=CN, label=图12, caption=测试集泄漏孔径识别与定位结果, figureFileSmall=l+UfKTvSliJcANUBdJmCRQ==, figureFileBig=uL3URnOh8Diw24vUnjWUMw==, tableContent=null), ArticleFig(id=1156986821482734126, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, language=EN, label=Table 1, caption=Filter algorithm evaluation, figureFileSmall=null, figureFileBig=null, tableContent=
算法 RMSE MAE SNR NCC
卡尔曼 0.066 0.057 35.36 0.985
均值 0.080 0.051 34.54 0.982
低通 0.042 0.057 37.27 0.990
VMD 0.020 0.044 40.62 0.995
), ArticleFig(id=1156986821566620209, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, language=CN, label=表1, caption=滤波算法评估, figureFileSmall=null, figureFileBig=null, tableContent=
算法 RMSE MAE SNR NCC
卡尔曼 0.066 0.057 35.36 0.985
均值 0.080 0.051 34.54 0.982
低通 0.042 0.057 37.27 0.990
VMD 0.020 0.044 40.62 0.995
), ArticleFig(id=1156986821629534771, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, language=EN, label=Table 2, caption=Assessment results of CNN-LSTM and LSTM, figureFileSmall=null, figureFileBig=null, tableContent=
算法 指标 泄漏位置 $/\mathrm{m}$ 泄漏孔径 $/\%$
CNN-LSTM MAE 2.028 000 0.000 621
MAPE 2.112 000 3.092 000
RMSE 2.848 000 0.000 734
LSTM MAE 5.959 000 0.000 706
MAPE 5.446 000 3.987 000
RMSE 6.965 000 0.000 759
), ArticleFig(id=1156986821679866421, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156986625868784228, language=CN, label=表2, caption=CNN-LSTM 和 LSTM 评估结果, figureFileSmall=null, figureFileBig=null, tableContent=
算法 指标 泄漏位置 $/\mathrm{m}$ 泄漏孔径 $/\%$
CNN-LSTM MAE 2.028 000 0.000 621
MAPE 2.112 000 3.092 000
RMSE 2.848 000 0.000 734
LSTM MAE 5.959 000 0.000 706
MAPE 5.446 000 3.987 000
RMSE 6.965 000 0.000 759
)], attaches=null, journal=Journal(id=1146119176004939786, delFlag=0, nameCn=科学技术与工程, nameEn=Science Technology and Engineering, nameHistory1=null, nameHistory2=null, issn=1671-1815, eissn=, cn=11-4688/T, coden=null, periodic=4, language=CN, oaType=是, ccby=null, superviseOffice=null, ownerOffice=null, pubOffice=null, editorOffice=null, officeType=null, aims=null, clcCode=null, officeProv=null, officeCity=null, officeAddr=null, officeZip=null, officeEmail=null, officePhone=null, editDirector=null, officeDirector=null, officeDirectorPhone=null, officeStaffNum=null, officeEmpNum=null, coverPicUrl=UKU/O7GSka5polgCTkbIIw==, journalPrice=null, startedYear=null, abbrevIsoEn=Sci Technol Eng, journalRemark=null, publicationField=null, createdTime=null, updatedTime=1754445529766, createdBy=null, updatedBy=13701087609, firstLetterCn=S, firstLetterEn=S, subjectCode=Natural Sciences, subjectName=自然科学, subjectCodeEn=Natural Sciences, subjectNameEn=null, picCn=UKU/O7GSka5polgCTkbIIw==, picEn=5hwlULoNwcbj3xUmVi9MAQ==, jcr=null, cjcr=null, exts=[JournalExt(id=1159791870395564357, language=CN, name=科学技术与工程, nameHistory1=null, nameHistory2=null, managedBy=, sponsoredBy=, publishedBy=, editorOffice=, officeProv=null, officeCity=null, officeAddr=, officeZip=, editDirector=null, officeDirector=null, officePhone=null, coverPicUrl=null, journalRemark=, submitArticleUrl=null, websiteUrl=http://www.stae.com.cn/jsygc/home, createdTime=1754445529793, updatedTime=1754445529793, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=http://www.stae.com.cn/jsygc/site/menus/20090429150146001, submissionAuthorUrl=http://www.stae.com.cn/jsygc/author/login, submissionEditorUrl=http://www.stae.com.cn/jsygc/editor/login, submissionReviewUrl=http://www.stae.com.cn/jsygc/reviewer/login, submissionCeEditorUrl=, submissionAeEditorUrl=, option={"copyright":""}), JournalExt(id=1159791870441701702, language=EN, name=Science Technology and Engineering, nameHistory1=null, nameHistory2=null, managedBy=, sponsoredBy=, publishedBy=, editorOffice=, officeProv=null, officeCity=null, officeAddr=, officeZip=, editDirector=null, officeDirector=null, officePhone=null, coverPicUrl=null, journalRemark=, submitArticleUrl=null, websiteUrl=http://www.stae.com.cn/jsygc/home, createdTime=1754445529804, updatedTime=1754445529804, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=, submissionAuthorUrl=http://www.stae.com.cn/jsygc/author/login, submissionEditorUrl=http://www.stae.com.cn/jsygc/editor/login, submissionReviewUrl=http://www.stae.com.cn/jsygc/reviewer/login, submissionCeEditorUrl=, submissionAeEditorUrl=, option={"copyright":""})], databaseList=null, tenantJournalId=1146123166801305609, websiteList=[Website(id=1148243202391400884, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1146123166801305609, journalNameCn=null, journalNameEn=null, grayFlag=null, tenantId=1146029695717560320, platformId=null, journalGroupId=null, journalGroupNameCn=null, journalGroupNameEn=null, type=1, domain=https://castjournals.cast.org.cn/joweb/kxjsygc/CN, language=CN, createTime=1751692112777, createBy=18614031015, updateTime=1753520965431, updateBy=18614031015, name=科学技术与工程-中文站点, tplId=1146099689490845704, title=科学技术与工程, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1148622798802673703, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202391400884, code=articleTextType, value=kx, createTime=1751782615614, updateTime=1751782615614, creator=18614031015, updator=18614031015), WebsiteProps(id=1148622798781702180, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202391400884, code=banner, value=null, createTime=1751782615609, updateTime=1751782615609, creator=18614031015, updator=18614031015), WebsiteProps(id=1148622798769119267, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202391400884, code=logo, value=https://castjournals.cast.org.cn/joweb/kjdb/CN/file/pic?fileId=j86gbwi+p0Idkyl5SzIlmQ==, createTime=1751782615606, updateTime=1751782615606, creator=18614031015, updator=18614031015), WebsiteProps(id=1148622798794285094, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202391400884, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/kjdb/CN/file/pic, createTime=1751782615612, updateTime=1751782615612, creator=18614031015, updator=18614031015), WebsiteProps(id=1148622798790090789, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202391400884, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_cn_619/, createTime=1751782615611, updateTime=1751782615611, creator=18614031015, updator=18614031015)]), Website(id=1155914124811976731, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1146123166801305609, journalNameCn=null, journalNameEn=null, grayFlag=null, tenantId=1146029695717560320, platformId=null, journalGroupId=null, journalGroupNameCn=null, journalGroupNameEn=null, type=1, domain=https://castjournals.cast.org.cn/joweb/kxjsygc/EN, language=EN, createTime=1753521003206, createBy=18614031015, updateTime=1753521003206, updateBy=18614031015, name=科学技术与工程-英文站点, tplId=1146101810881728533, title=Science Technology and Engineering, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1155914371227308235, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155914124811976731, code=articleTextType, value=kx, createTime=1753521061952, updateTime=1753521061952, creator=18614031015, updator=18614031015), WebsiteProps(id=1155914371210531016, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155914124811976731, code=banner, value=null, createTime=1753521061947, updateTime=1753521061947, creator=18614031015, updator=18614031015), WebsiteProps(id=1155914371202142407, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155914124811976731, code=logo, value=https://castjournals.cast.org.cn/joweb/kjdb/CN/file/pic?fileId=j86gbwi+p0Idkyl5SzIlmQ==, createTime=1753521061945, updateTime=1753521061945, creator=18614031015, updator=18614031015), WebsiteProps(id=1155914371223113930, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155914124811976731, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/kjdb/CN/file/pic, createTime=1753521061950, updateTime=1753521061950, creator=18614031015, updator=18614031015), WebsiteProps(id=1155914371218919625, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155914124811976731, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_cn_619/, createTime=1753521061949, updateTime=1753521061949, creator=18614031015, updator=18614031015)])], journalTitle=科学技术与工程, weixinUrl=null, journalUrl=null, iacademicId=null, status=0, seqNo=null, journalTitleEn=Science Technology and Engineering, journalPhotoCn=UKU/O7GSka5polgCTkbIIw==, journalPhotoEn=5hwlULoNwcbj3xUmVi9MAQ==, journalFirstLetter=S, journalRecommend=null, journalNew=null, journalCollection=null, jcrJf=null, cjcrJf=null, jcrJfStr=null, cjcrJfStr=null, submissionFirstDecision=null, sciSubjectClassification=null, casSubjectClassification=null, citeScore=null, totalCitationFrequency=null, icpCode=null, psCode=null, advertisingLicenseCode=null, copyrightInformation=null, country=null, option=null, provinceCode=null, provinceName=null, collectFlag=false), detailUrlCn=https://castjournals.cast.org.cn/joweb/kxjsygc/CN/10.12404/j.issn.1671-1815.2402034, detailUrlEn=https://castjournals.cast.org.cn/joweb/kxjsygc/EN/10.12404/j.issn.1671-1815.2402034, pdfUrlCn=https://castjournals.cast.org.cn/joweb/kxjsygc/CN/PDF/10.12404/j.issn.1671-1815.2402034, pdfUrlEn=https://castjournals.cast.org.cn/joweb/kxjsygc/EN/PDF/10.12404/j.issn.1671-1815.2402034, aliStartDate=null, aliEndDate=null, collectionFlag=false, citedCount=null, citedUrl=null, reference=null)
收藏切换
煤矿井下供水管道泄漏孔径识别与定位
收藏切换
PDF下载
杜京义 1 , 陈镇 1, * , 张嘉伟 1 , 李晨 2 , 高瑞 1 , 王鹏 3
科学技术与工程 | 自动化技术、计算机技术 2025,25(8): 3296-3303
收起
收藏切换
科学技术与工程 | 自动化技术、计算机技术 2025, 25(8): 3296-3303
煤矿井下供水管道泄漏孔径识别与定位
全屏
杜京义1 , 陈镇1, * , 张嘉伟1, 李晨2, 高瑞1, 王鹏3
作者信息
  • 1 西安科技大学 电气与控制工程学院 西安 710054
  • 2 西安科技大学 通信与信息工程学院 西安 710054
  • 3 西安科技大学 机械工程学院 西安 710054
  • 杜京义(1965—),男,汉族,山东淄博人,博士,教授。研究方向:检测技术及其自动化。E-mail:517571853@qq.com。

通讯作者:

*陈镇(1999—),男,汉族,安徽六安人,硕士研究生。研究方向:模式识别与嵌入式系统。E-mail:1745293122@qq.com。
Identification and Location of Leakage Aperture of Underground Water Supply Pipeline in Coal Mine
Jing-yi DU1 , Zhen CHEN1, * , Jia-wei ZHANG1, Chen LI2, Rui GAO1, Peng WANG3
Affiliations
  • 1 College of Electrical and Control Engineering Xi'an University of Science and Technology Xi'an 710054 China
  • 2 College of Communication and Information Engineering Xi'an University of Science and Technology Xi'an 710054 China
  • 3 College of Mechanical Engineering Xi'an University of Science and Technology Xi'an 710054 China
出版时间: 2025-03-18 doi: 10.12404/j.issn.1671-1815.2402034
文章导航
收藏切换

为快速识别煤矿井下泄漏点的位置及泄漏孔径,利用供水管道泄漏时产生的压力及流量信号,提出一种泄漏孔径识别与定位模型。首先利用模态能量熵和遗传算法结合包络熵对变分模态分解(variational mode decomposition,VMD) 进行参数优化,再使用VMD 对压力信号进行降噪处理;采用卷积神经网络(convolutional neural networks, CNN) 提取压力及流量信号的深层特征序列,长短时记忆网络(long short- term memory, LSTM) 提取深层特征序列的时序特征,进行泄漏孔径识别与定位。实验结果表明:经过参数优化的变分模态分解,相较卡尔曼滤波、均值滤波、低通滤波在均方根误差(root mean square error,RMSE) 、平均绝对误差(mean absolute error, MAE) 、信噪比(signal to noise ratio, SNR) 、归一化互相关系数(normalized cross correlation, NCC) 上均有提高,表明其能够有效降低噪声成分,保留有效信号;CNN-LSTM 相较LSTM,在泄漏点定位中,MAE降低了65. 97% ,平均绝对百分比误差( mean absolute percentage error, MAPE) 降低了61. 22% ,RMSE 降低了59. 11% 。在泄漏孔径识别中,MAE 降低了12. 04% ,MAPE 降低了22. 45% ,RMSE 降低了3. 29% ,证明CNN-LSTM 可以充分利用管道压力及流量信号的空间及时间特征进行泄漏位置及孔径的识别,其检测效果相较LSTM 更加准确和稳定。

变分模态分解(VMD)  /  卷积神经网络(CNN)  /  长短时记忆网络(LSTM)  /  模态能量熵  /  遗传算法(GA)  /  包络熵

In order to quickly identify the location of the leakage point and the leak aperture in the coal mine, a model was proposed for identifying and locating the leak aperture by using the pressure and flow signals generated when the water supply pipeline leaked. Modal energy entropy and genetic algorithm combined with envelope entropy were used to optimize the parameters of variational mode decomposition (VMD), and then VMD was used to denoise the pressure signal. Convolutional neural network (CNN) was used to extract the deep feature sequence of pressure and flow signal, and the long short-term memory network (LSTM) was used to extract the time sequence of deep feature sequence to identify and locate the leak aperture. The experimental results show that compared with Kalman filter, mean value filter and low-pass filter, the variational modal decomposition with optimized parameters has higher root-mean-square error (RMSE), mean absolute error (MAE), signal-to-noise ratio (SNR) and normalized cross correlation (NCC), which indicates that it can effectively reduce noise components and retain effective signals. Compared with LSTM, the MAE, mean absolute percentage error (MAPE) and RMSE of CNN-LSTM in leak location decrease by 65.97%, 61.22% and 59.11%. In the identification of leak aperture, MAE decreases by 12.04%, MAPE decreases by 22.45%, and RMSE decreases by 3.29%, which proves that CNN-LSTM can make full use of the spatial and temporal characteristics of pipeline pressure and flow signals to identify the leak location and aperture, and its detection effect is more accurate and stable than LSTM.

variational mode decomposition (VMD)  /  convolutional neural networks (CNN)  /  long short-term memory (LSTM)  /  modal energy entropy  /  genetic algorithm (GA)  /  envelope entropy
杜京义, 陈镇, 张嘉伟, 李晨, 高瑞, 王鹏. 煤矿井下供水管道泄漏孔径识别与定位. 科学技术与工程, 2025 , 25 (8) : 3296 -3303 . DOI: 10.12404/j.issn.1671-1815.2402034
Jing-yi DU, Zhen CHEN, Jia-wei ZHANG, Chen LI, Rui GAO, Peng WANG. Identification and Location of Leakage Aperture of Underground Water Supply Pipeline in Coal Mine[J]. Science Technology and Engineering, 2025 , 25 (8) : 3296 -3303 . DOI: 10.12404/j.issn.1671-1815.2402034
煤矿井下供水管道泄漏是较为常见的煤矿事故类型, 由于其长期处于湿度较高的恶劣环境, 易产生腐蚀破裂。传统检漏方式为人工巡检,存在时间长、人力成本高、危险性大等问题,且其难以捕捉突发问题。因此, 供水管网泄漏孔径识别与定位研究对井工煤矿安全作业具有重要意义, 能有效提高生产的安全性和效率。
负压波检测法作为管道泄漏检测技术中最成熟的方法, 其易于在井下布置, 能快速检测出管道泄漏, 但由于电磁干扰等环境噪声的影响, 采集到的压力信号中往往存在着大量噪声,对泄漏检测带来很大的影响。随着越来越多的学者投入研究, 大量管道信号去噪方法不断提出。Dragomiretskiy 等[1] 提出变分模态分解(variational mode decomposition, VMD), 其可以避免经验模态分解 (empirical mode decomposition, EMD) 存在的收敛不佳及模态混叠的问题。王秀芳等[2] 提出通过评估 VMD 相邻模态能量值来选取有效模态重构信号的方法, 识别了强噪声干扰的管道小泄漏信号。然而, 变分模态分解需要预先设定参数。在 VMD 的参数优化研究中, 唐贵基等 [3] 采用粒子群优化算法来优化 VMD 的参数设置,增强了对微弱特征频率的提取能力。张萍等[4] 运用鲸鱼优化算法(whale optimization algorithm, WOA) 对 VMD 进行参数寻优, 显著提高了轴承故障检测的准确率。
传统的泄漏检测法利用传感器采集信号,通过支持向量机(support vector machine, SVM)[5] 、反向传播(back propagation, BP) 神经网络[6] 等机器学习算法进行泄漏监测。但在这类浅层网络难以满足精准的泄漏孔径识别及定位的要求。近年来, 深度学习算法因其挖掘深层特征的能力在多种领域中应用,刘源[7] 基于一维卷积神经网络实现了泄漏孔径识别。尽管上述方法在提取信号特征上取得了一定成效, 但主要集中于数据的空间分布特性, 无法利用信号的时间分布特性[8-13]
针对管道信号滤波中变分模态分解的参数寻优,检测模型存在未同时考虑多种信号之间的空间及时间特征等问题。鉴于此, 提出一种基于 VMD- 卷积神经网络 (convolutional neural networks, CNN)- 长短时记忆网络 (long short- term memory, LSTM) 的煤矿井下供水管道的泄漏孔径识别与定位方法。该方法通过变分模态分解对压力信号进行处理, 消除环境噪声影响, 并利用模态能量熵确定最优模态数, 通过遗传算法基于包络熵优化惩罚系数及噪声容限, 提升信号分解效果, 将重构的压力信号和流量信号通过卷积神经网络提取其深层特征,再利用长短时记忆网络对卷积神经网络提取的特征序列进行时序特征提取, 实现煤矿井下供水管道的泄漏孔径识别与定位。
针对供水管道泄漏产生的压力及流量信号的特点, 设计基于变分模态分解和时空网络 (VMD-CNN-LSTM) 的泄漏孔径识别与定位模型, 如图1 所示。VMD-CNN-LSTM 的模型架构主要包括 4 层:输入层、CNN 层、LSTM 层和输出层。在输入层中, 首先对 VMD 进行参数优化,再使用 VMD 将压力信号拆分为多个模态分量, 降低压力信号中的噪声分量。将流量信号与重构后的压力信号融合, 作为 CNN 层输入数据。在 CNN 层,输入矩阵和特定卷积核进行一维卷积池化操作,提取泄漏信号中压力和流量这两种特征序列之间的深层特征, 并使用 Dropout 层,避免过拟合效应。在 LSTM 层中,将 CNN 层输出的特征序列作为输入, 利用 LSTM 提取特征序列的时序特征。在输出层中, 输出泄漏位置及泄漏孔径。
压力信号在采集过程中由于现场施工、电磁干扰等环境噪声的影响, 严重影响了信号质量。VMD 算法噪声鲁棒性强, 特别适应于非线性非平稳信号分析处理,采用变分模态分解处理管道压力信号, 以达到更好的管道泄漏孔径识别与定位效果。
VMD 是在 Wiener 滤波和 Hibert 变换的基础上优化的一种算法, 其旨在将输入信号分解为若干个模态信号, 其具有特定的带宽稀疏特性, 可组成重构信号。其主要步骤如下。
步骤 1 构建约束变分问题。
$\left\{\begin{array}{l}\mathop{\min }\limits_{\left|{u}_{k},{w}_{k}\right|}\left({\mathop{\sum }\limits_{{k = 1}}^{K}{\begin{Vmatrix}{\partial }_{t}\left\{\left\lbrack \delta \left(t\right)+ \frac{\mathrm{j}}{\pi t}\right\rbrack *{u}_{k}\left(t\right)\right\}{\mathrm{e}}^{-\mathrm{j}{w}_{k}t}\end{Vmatrix}}_{2}^{2}}\right)\\\text{ s.t.}\mathop{\sum }\limits_{{k = 1}}^{K}{u}_{k}= f\left(t\right)\end{array}\right.$
式 (1) 中:${u}_{k}\left(t\right)$ 为原信号经变分模态分解后的第$k$ 个分量;$K$ 为模态总个数;$t$ 为时间;$f\left(t\right)$ 为原始信号;${w}_{k}$ 为有限带宽;${\partial }_{t}$$t$ 的偏导;${\delta }_{t}$ 为狄拉克函数;* 为卷积运算符。
步骤 2 约束变分问题求解。
为求解式 (1), Dragomiretskiy 等[1] 采用二次惩罚函数与拉格朗日乘子$\lambda \left(t\right)$,将其转化为无约束优化问题。经变换后的具有增广拉格朗日函数的方程为
$ L\left({\left\{{u}_{k}\right\},\left\{{w}_{k}\right\},\lambda }\right)= \\\alpha \mathop{\sum }\limits_{{k = 1}}^{K}{\begin{Vmatrix}{\partial }_{t}\left\{{\left\lbrack \delta \left(t\right)+ \frac{\mathrm{j}}{\pi t}\right\rbrack }^{* }{u}_{k}\left(t\right)\right\}{\mathrm{e}}^{-\mathrm{j}{w}_{k}t}\end{Vmatrix}}_{2}^{2}+ \\{\begin{Vmatrix}f\left(t\right)- \mathop{\sum }\limits_{{k = 1}}^{K}{u}_{k}\left(t\right)\end{Vmatrix}}_{2}^{2}+ \\\langle \lambda \left(t\right), f\left(t\right)- \mathop{\sum }\limits_{{k = 1}}^{K}{u}_{k}\left(t\right)\rangle $
式 (2) 中:$\lambda$ 为拉格朗日算子;$\alpha$ 为惩罚系数;$L$ 为拉格朗日变换;$\langle \rangle$ 为内积运算。
VMD 分解需要预先确定模态数, 即分解的成分个数。在模态数不合适的情况下, 很难较好的分解信号, 得不到有效的固有模态函数 (intrinsic mode function, IMF) 分量, 然而由于实际管道采集到的压力信号的特性复杂多样, 确定合适的模态数是一个十分困难的任务。当选取较少模态数量时, VMD 可能过滤掉负压波信号中的关键信息,影响网络检测的准确度。相反,当选取较多模态数量时,相邻模态分量的中心频率会比较接近, 可能引起模态混叠,引入额外的噪声干扰;同时,惩罚系数$\alpha$ 也会影响变分模态分解的效果,$\alpha$ 的取值直接影响着各 IMF 分量的带宽, 同时惩罚系数设置的不合理会显著提高算法运行时间; 噪声容限$\lambda$ 也会对分解效果产生显著影响, 噪声容限过大, 会使得模态的平滑度过高, 导致模态之间的细节信息丢失, 而过小的噪声容限会使模态中引入大量噪声成分, 导致模态质量下降。为提升管道压力信号去噪效果, 对变分模态分解的模态数及惩罚系数和噪声容限进行参数优化。
基于模态能量差值的最优模态数选取法。为解决最优模态数选取的问题, 避免压力信号被过分解或欠分解, 使用模态能量差值法选取最优模态数, 其主要思想为: 通过计算不同模态数的总能量, 评估相邻模态的增量来选取最优模态数。当发生过分解时, 会引入噪声, 该模态数下的总能量会突增,因此选择该模态相邻的模态数为最优模态数, 可避免过分解, 通过不断选取直到发生过分解, 也可避免欠分解问题。
$\left\{\begin{array}{l}{E}_{k}^{l}= \sqrt{\frac{\mathop{\sum }\limits_{{i = 1}}^{n}{x}_{l}^{2}\left(i\right)}{n}},\;l = 1,2,\cdots , k \\{E}_{k}= \mathop{\sum }\limits_{{l = 1}}^{k}{E}_{k}^{l}\\{\theta }_{k, k - 1}= \frac{\left|{E}_{k}- {E}_{k - 1}\right|}{n}\end{array}\right.$
式 (3) 中:${E}_{k}^{l}$ 为模态数为$k$ 时第$l$ 个模态的能量;${x}_{l}^{2}\left(i\right)$ 为第$l$ 个模态在第$i$ 个数据点上的值;${E}_{k}$ 为模态数为$k$ 时所有模态的总能量;${\theta }_{k, k - 1}$ 用于评估相邻模态的能量变化比例。
基于遗传算法的最优惩罚系数及噪声容限选取, 主要包括遗传算法和适应度函数, 具体如下。
(1)遗传算法。利用遗传算法, 选取最优的惩罚系数和噪声容限,以达到最优的分解效果。其主要工作流程如图2 所示。
(2)适应度函数。适应度在遗传算法中扮演者至关重要的角色, 它用于评估个体在特定问题内的表现,染色体的适应度通常是通过定义的目标函数计算得出的。
经过变分模态分解处理后的压力信号, 若各模态分量中噪声成分较多而信号特征不明显,那么相应的 IMF 分解结果将展现较低的稀疏度, 其包络熵也会显著增加。因此, 包络熵可作为重构信号质量的评估指标,引入包络熵作为遗传算法寻优的适应函数, 以最小化包络熵作为寻优目标, 获得最优的惩罚系数及噪声容限,以抑制各分量的噪声信号, 保留负压波信号特征。
将包络信号进行处理,得到概率分布序列${p}_{j}$,概率分布序列${p}_{j}$ 熵值即包络熵。零均值信号$x\left(j\right)(j =$ $1,2,\cdots , N)$ 的包络熵${E}_{p}$,可表示为
$\left\{\begin{array}{l}{E}_{p}= -\mathop{\sum }\limits_{{j = 1}}^{N}{p}_{j}\lg {p}_{j}\\{p}_{j}= \frac{a\left(j\right)}{\mathop{\sum }\limits_{{j = 1}}^{N}a\left(j\right)} \end{array}\right.$
式 (4) 中:$a\left(j\right)$ 为信号$x\left(j\right)$ 的包络信号;${p}_{j}$$a\left(j\right)$ 的归一化形式。
由于是利用管首管尾的压力和流量信号进行泄漏孔径识别和定位,共有 4 组时序信号,而直接使用时序模型难以挖掘其相互间的深层特征, 提高检测的鲁棒性和精确性, 因此将压力和流量在泄漏信号中的深层特征序列通过 CNN 网络进行提取。CNN 层主要由卷积层、池化层和 Dropout 层组成。
将流量序列与经过变分模态分解重构后的压力序列作为输入的特征矩阵, 卷积层将特征矩阵进行卷积得到特征向量, 随后输入到池化层进行最大池化操作,将池化后的特征向量输入 Dropout 层,减少每次迭代中对局部特征的依赖。
利用 LSTM 对卷积神经网络提取的特征序列进行时序特征提取, 以深入挖掘泄漏信号在时序上的内在关联。
LSTM 通过门结构管理和调节信息, 其激活函数的反向误差可随迭层数的增加而向下传递, 有效解决了传统循环神经网络的长期依赖问题。其隐含层包含遗忘门、输入门和输出门, 能接受上一层神经元的输出, 并保留过去的重要信息。具体的 3 种门结构如图3 所示。
(1)遗忘门。主要作用是信息筛选,将没有意义的信息丢弃。遗忘门接收上一时刻隐含层的输出${\mathbf{h}}_{t - 1}$ 和当前时刻的输入${\mathbf{x}}_{t}$,并将其通过激活函数。遗忘门的计算公式为
${\mathbf{f}}_{t}= \sigma \left({{\mathbf{W}}_{f}\left\lbrack {{\mathbf{h}}_{t - 1},{\mathbf{x}}_{t}}\right\rbrack +{\mathbf{b}}_{f}}\right)$
式 (5) 中:${\mathbf{W}}_{f}$${\mathbf{h}}_{t - 1}$${\mathbf{x}}_{t}$ 到遗忘门的权重矩阵;${\mathbf{b}}_{f}$ 为偏置。
(2)输入门。主要作用是更新状态,它将前一时刻隐藏状态的信息${\mathbf{h}}_{t - 1}$ 和当前输入的信息${\mathbf{x}}_{t}$ 传递到激活函数$\sigma$ 中,决定更新哪些信息。$\tanh$ 层接收${\mathbf{h}}_{t - 1}$${\mathbf{x}}_{t}$,生成候选状态向量${\widetilde{\mathbf{C}}}_{t}$,遗忘的信息${f}_{t}$ 与上一时刻的状态${\mathbf{C}}_{t - 1}$ 相乘,加入新的信息${\mathbf{i}}_{t}* {\widetilde{\mathbf{C}}}_{t}$, 得到新的状态。涉及的计算公式为
$\left\{\begin{array}{l}{\mathbf{i}}_{t}= \sigma \left({{\mathbf{W}}_{i}\left\lbrack {{\mathbf{h}}_{t - 1},{\mathbf{x}}_{t}}\right\rbrack +{b}_{t}}\right)\\{\widetilde{\mathbf{C}}}_{t}= \tanh \left({{\mathbf{W}}_{c}\left\lbrack {{\mathbf{h}}_{t - 1},{\mathbf{x}}_{t}}\right\rbrack +{\mathbf{b}}_{c}}\right)\\{\mathbf{C}}_{c}= {\mathbf{f}}_{c}{\mathbf{C}}_{c - 1}+ {\mathbf{i}}_{c}{\widetilde{\mathbf{C}}}_{1}\end{array}\right.$
式(6)中:${\mathbf{W}}_{i}\text{、}{\mathbf{W}}_{c}$ 分别为对应${\mathbf{h}}_{t - 1}$${\mathbf{x}}_{t}$ 的权重矩阵;${\mathbf{b}}_{t}$${\mathbf{b}}_{c}$ 为对应偏置; tanh 为激活函数;${\mathbf{i}}_{t}$ 为输入门。
(3)输出门。主要作用是决定下一时刻的隐藏状态。通过将前一时刻的隐藏状态${\mathbf{h}}_{t - 1}$ 与当前时刻的输入${\mathbf{x}}_{t}$ 传递到激活函数$\sigma$,然后将新的状态${\mathbf{C}}_{t}$ 传递给 tanh 层,将它们的输出相乘,确定隐藏状态${\mathbf{h}}_{t}$ 中应包含的信息。输出门${\mathbf{O}}_{t}$ 的计算公式为
$\left\{\begin{array}{l}{\mathbf{O}}_{t}= \sigma \left({{\mathbf{W}}_{o}\left\lbrack {{\mathbf{h}}_{t - 1},{\mathbf{x}}_{t}}\right\rbrack +{\mathbf{b}}_{o}}\right)\\{\mathbf{h}}_{t}= {\mathbf{O}}_{t}\tanh \left({\mathbf{C}}_{t}\right)\end{array}\right.$
式(7)中:${\mathbf{W}}_{o}$ 为前一时刻的隐藏状态${\mathbf{h}}_{t - 1}$ 与当前输入${\mathbf{x}}_{t}$ 到输出门激活函数$\sigma$ 的权重;${\mathbf{b}}_{o}$ 为与之对应的偏置。
针对煤矿井下供水管道的泄漏问题,采用 Flowmaster仿真软件对不同泄漏位置、泄漏孔径的瞬态变化规律进行仿真, 其计算模型如图4 所示。
管道为总长${800}\mathrm{\;m}$,内径${0.1}\mathrm{\;m}$ 的不可压道管壁的绝对粗糙度设置为${0.025}\mathrm{\;{mm}}$,通过阀门控制器 12 控制阀门 7 来模拟不同的泄漏更好的仿真泄漏, 将水库 9 的水位设置为$0\mathrm{\;m}$,表面压力设置为大气压,管道内水击波传播速度设置为${1000}\mathrm{\;m}/\mathrm{s}$,通过控制管道 2 和管道 3 的长度来模拟不同位置的泄漏。
为了模拟现实环境中采集到的负压波信号, 利用 Flowmaster 仿真得到的泄漏产生的负压波信号, 并向其中加入${40}\mathrm{\;{dB}}$ 的高斯白噪声,以模拟现实环境中引入的噪声信号, 仿真信号可表示为
$\left\{\begin{array}{l} x\left(t\right)= s\left(t\right)+ n\left(t\right)\\ n\left(t\right)= {A}^{* }\eta \left(t\right)\end{array}\right.$
式 (8) 中:$x\left(t\right)$ 为带噪负压波信号;$s\left(t\right)$ 为仿真得到的纯净负压波信号;$n\left(t\right)$ 为高斯白噪声;$A$ 为标准差;$\eta \left(t\right)$ 为在时间$t$ 上的标准正态分布。
图5 为纯净负压波信号及添加${40}\mathrm{\;{dB}}$ 的高斯白噪声信号。
通过模态能量差值法选取最优模态数,图6 为通过固定惩罚系数及噪声容限, 不断迭代模态数得到的能量变化比迭代图,可以看出,在模态数$K$ 为 4 时,总能量突增,判断其发生过分解,故选取最优模态数为 4。
固定模态数为 4 , 利用遗传算法选取最优惩罚系数及噪声容限,得到最优惩罚系数 Alpha 为 1717, 噪声容限为 40.383, 其对应的适应函数 (即包络熵) 为 26.6。
利用最优的模态数、惩罚系数、噪声容限对带噪压力信号进变分模态分解得。图7 为变分模态分解的各 IMF 分量, 可见各分量均未出现噪声干扰, 说明其没有出现欠分解而引入噪声。
图8 为变分模态分解的 IMF 分量频谱。图9 为变分模态分解的信号重构结果, 对比纯净信号, 可见变分模态分解很好地保留了信号的有效信息, 并滤除了信号中混杂的噪声信号。
将 VMD 变分模态分解的效果与卡尔曼滤波、 低通滤波等算法比较, 通过均方根误差 (root mean square error, RMSE)、平均绝对误差 (mean absolute error, MAE)、信噪比 (signal to noise ratio, SNR)、归一化互相关系数(normalized cross correlation, NCC) 进行比较评估。
表1 为各类滤波算法的评估结果, 可见经过参数优化的变分模态分解的均方误差和平均绝对误差相比其他算法有着显著的优势, 相较于整体效果较优的低通滤波, 变分模态分解的均方误差降低了 54%, 其平均绝对误差降低了 22.9%, 这表明其成功减小了滤波后信号与纯净信号之间的数值误差; 同时变分模态分解的信噪比也是最高的, 相较于低通滤波提高了${8.2}\%$,表明其能够有效减少噪声成分, 保留有效信号; 其归一化互相关系数相较于其他算法亦有一些提高, 表明其滤波后的信号与纯净信号具有更高的相似性。为直观的比较各个算法的效果, 将每项指标标准化后生成的雷达图如图10 所示。
图10 可知, 经过参数寻优的变分模态分解的信噪比及归一化互相关系数是所有算法中最高的, 其均方根误差、平均绝对误差是其中最低的。
本实验采用操作系统为 64 位 Windows11 的计算机硬件平台,其处理器为 Intel i9-13900HX CPU, 运行内存为 16 GB,显卡为 NVIDIA RTX40608 GB; 本实验采用的图形处理加速器为 CUDA11.2 和 cuDNN8.2.1,编程语言为 Python3.8, 深度学习框架采用 Tensorflow-gpu2.6。
为训练和评估 CNN-LSTM 进行泄漏孔径识别与定位的效果, 制作了不同泄漏位置及泄漏孔径的数据集, 原始数据来源于 Flowmaster 仿真输出, 通过 Flowmaster 进行不同泄漏工况的瞬态仿真, 获取了 100 组实验数据, 其压力及流量数据采样率为 100 个$/\mathrm{s}$,时长为${20}\mathrm{\;s}$。划分数据集时,随机筛选出 80 组作为训练集, 剩余为测试集。
采用 Huber 损失对管道泄漏位置及泄漏孔径进行训练, Batchsize 为 4096 , 模型迭代次数为 200 , 初始学习率为 0.01 , 泄漏位置及泄漏孔径的训练损失曲线如图11 所示。
图11 可知, 在第 125 次迭代时, CNN-LSTM 模型的两个目标的损失函数开始收敛, 并且最终损失在${1.2}\times {10}^{-4}$ 以下。
图12 为测试集的检测结果与实际值的比较, 可见CNN-LSTM能比较好的检测泄漏位置及泄漏孔径。
为验证 CNN-LSTM 模型改进的有效性, 利用测试集检测结果与实际值的 MAE、平均绝对百分比误差 (mean absolute percent error, MAPE)、RMSE 进行评估, 与 LSTM 模型检测的结果进行对比实验。如表2 所示, 可见 CNN-LSTM 在泄漏位置及泄漏孔径的三类评价指标上均优于 LSTM, 其中在泄漏位置检测上,检测的单位为$\mathrm{m}$,其 MAE 降低了 65.97%, MAPE 降低了 61.22%, RMSE 降低了 59.11%; 在泄漏孔径检测中,单位为$\%$,即泄漏孔径相对于管道直径的百分比,其 MAE 降低了 12.04%, MAPE 降低了 22.45%, RMSE 降低了 3.29%。进一步证明了 CNN-LSTM 模型在泄漏孔径识别与定位中优于常规的 LSTM 模型, 从 4 项评估指标可见 CNN-LSTM 可以充分利用泄漏过程中的相关数据进行泄漏位置及泄漏孔径的检测,其检测效果更加准确和稳定。
(1) 在压力信号的去噪上, 提出利用模态能量熵变化比进行变分模态分解的最优模态数选取, 并利用遗传算法结合包络熵选择最优惩罚系数与噪声容限进行变分模态分解, 有效消除了压力信号中的噪声,相较其他滤波算法有较大提升。
(2)在泄漏孔径识别与定位中,提出使用 CNN-LSTM 提取压力和流量信号在空间和时间维度上的特征。经过实验, CNN-LSTM 能比较好的检测泄漏位置及泄漏孔径, 相较 LSTM 在准确度上有较大提升。
  • 陕西省自然科学基础研究计划(2023-JC-YB-362)
参考文献 引证文献
排序方式:
[1]
Dragomiretskiy K, Zosso D. Variational mode decomposition[J]. IEEE Transactions on Signal Process, 2014. 62(3): 531-544.
[2]
王秀芳, 朱道鸿, 葛延良. 基于VMD-EV的天然气管道小泄漏信号去噪研究[J]. 压力容器, 2019. 36(3): 69-73.
Wang Xiufang, Zhu Daohong, Ge Yanliang. Research on denoising small leakage signal of natural gas pipeline based on VMD-EV[J]. Pressure Vessel Technology, 2019. 36(3): 69-73.
[3]
唐贵基, 王晓龙. 参数优化变分模态分解方法在滚动轴承早期故障诊断中的应用[J]. 西安交通大学学报, 2015. 49(5): 73-81.
Tang Guiji, Wang Xiaolong. Early fault diagnosis of rolling bearing with IVMD fusion singular value difference spectrum[J]. Journal of Xi'an Jiaotong University, 2015. 49(5): 73-81.
[4]
张萍, 张文海, 赵新贺, 等. WOA-VMD算法在轴承故障诊断中的应用[J]. 噪声与振动控制, 2021. 41(4): 86-93.
Zhang Ping, Zhang Wenhai, Zhao Xinhe, et al. Application of WOA-VMD algorithm in bearing fault diagnosis[J]. Noise and Vibration Control, 2021. 41(4): 86-93.
[5]
Qin W, HC, Liu CL, et al. Remaining useful life prediction for lithium-ion batteries using particle filter and artificial neural network[J]. Industrial Management & Data Systems, 2020. 120(2): 312-328.
[6]
Kuang YT, Singh R, Singh S, et al. A novel macroeconomic forecasting model based on revised multimedia assisted BP neural network model and ant Colony algorithmk[J]. Multimedia Tools and Applications, 2017. 76(18): 18749-18770.
[7]
刘源. 基于1DCNN的天然气管道泄漏孔径实时识别[D]. 大庆: 东北石油大学, 2023.
Liu Yuan. Real-time identification of natural gas pipeline leak aperture based on 1DCNN[D]. Daqing: Northeast Petroleum University, 2023.
[8]
罗伟, 张奎, 杨艳. 输油管道泄漏信号去噪的研究[J]. 科学技术与工程, 2011. 11(7): 1601-1604.
Luo Wei, Zhang Kui, Yang Yan. Research on denoising of oil pipeline leakage signal[J]. Science Technology and Engineering, 2011. 11(7): 1601-1604.
[9]
纪健, 傅晓宁, 王毅, 等. 气液两相流管道泄漏流场特性数值模拟分析[J]. 科学技术与工程, 2021. 21(1): 144-151.
Ji Jian, Fu Xiaoning, Wang Yi et al. Numerical simulation analysis of leakage flow field characteristics in gas-liquid two-phase flow pipeline[J]. Science Technology and Engineering, 2021. 21(1): 144-151.
[10]
王俊, 封辉, 高琦, 等. 埋地管道泄漏数值模拟分析[J]. 科学技术与工程, 2020. 20(33): 13660-13666.
Wang Jun, Feng Hui, Gao Qi, et al. Numerical simulation analysis of buried pipeline leakage[J]. Science Technology and Engineering, 2020. 20(33): 13660-13666.
[11]
李宏, 申瑞琦. 基于改进的经验模态分解方法的热网管道泄漏检测方法[J]. 科学技术与工程, 2014. 14(26): 224-227.
Li Hong, Shen Ruiqi. Pipeline leakage detection method of heat supply network based on improved empirical mode decomposition method[J]. Science Technology and Engineering, 2014. 14(26): 224-227.
[12]
汪建新, 王恩浩. 管道泄漏源定位中最优小波基变换[J]. 科学技术与工程, 2022. 22(12): 4797-4803.
Wang Jianxin, Wang Enhao. Optimal wavelet basis transformation in pipeline leakage source location[J]. Science Technology and Engineering, 2022. 22(12): 4797-4803.
[13]
马广兴, 曲波, 常琛, 等. 基于CNN的供热管道泄漏识别方法研究[J]. 电子测量技术, 2022. 45(16): 34-41.
Ma Guangxing, Qu Bo, Chang Chen, et al. Research on leakage identification method of Heating pipeline based on CNN[J]. Electronic Measurement Technology, 2022. 45(16): 34-41.
2025年第25卷第8期
PDF下载
334
130
引用本文
BibTeX
文章信息
doi: 10.12404/j.issn.1671-1815.2402034
  • 接收时间:2024-03-21
  • 首发时间:2025-07-29
  • 出版时间:2025-03-18
补充材料
相关文章
文章信息
作者
出版历史
  • 收稿日期:2024-03-21
  • 修回日期:2024-12-16
基金
陕西省自然科学基础研究计划(2023-JC-YB-362)
作者信息
    1 西安科技大学 电气与控制工程学院 西安 710054
    2 西安科技大学 通信与信息工程学院 西安 710054
    3 西安科技大学 机械工程学院 西安 710054

通讯作者:

*陈镇(1999—),男,汉族,安徽六安人,硕士研究生。研究方向:模式识别与嵌入式系统。E-mail:1745293122@qq.com。
参考文献
分享链接
https://castjournals.cast.org.cn/joweb/kxjsygc/CN/10.12404/j.issn.1671-1815.2402034
分享至
全文二维码

扫描看全文

引用本文
BibTeX
本文的引用情况
2种不同金属材料的力学参数

Family
属数
Number of
genus
种数
Number of
species
占总种数比例
Percentage of
total species (%)

Genus
种数
Number of
species
占总种数比例
Percentage of total
species (%)
鹅膏菌科Amanitaceae 2 11 5.26 鹅膏菌属 Amanita 10 4.78
小菇科 Mycenaceae 2 12 5.74 丝盖伞属 Inocybe 5 2.39
多孔菌科 Polyporaceae 8 14 6.70 蜡蘑属 Laccaria 5 2.39
红菇科 Russulaceae 3 23 11.00 小皮伞属 Marasmius 6 2.87
小菇属 Mycena 11 5.26
光柄菇属 Pluteus 5 2.39
红菇属 Russula 17 8.13
栓菌属 Trametes 5 2.39
关闭全屏