Article(id=1153695647673864667, tenantId=1146029695717560320, journalId=1146031654075715584, issueId=1153695641046864317, articleNumber=null, orderNo=null, doi=10.13234/j.issn.2095-2805.2024.5.67, pmid=null, cstr=null, oa=null, hot=null, price=null, onlineType=0, articleFormat=0, articleType=null, articleTypeStr=null, receivedDate=1712505600000, receivedDateStr=2024-04-08, revisedDate=1722355200000, revisedDateStr=2024-07-31, acceptedDate=1723305600000, acceptedDateStr=2024-08-11, onlineDate=1752992077033, onlineDateStr=2025-07-20, pubDate=1727625600000, pubDateStr=2024-09-30, doiRegisterDate=null, doiRegisterDateStr=null, onlineIssueDate=1752992077033, onlineIssueDateStr=2025-07-20, onlineJustAcceptDate=null, onlineJustAcceptDateStr=null, onlineFirstDate=null, onlineFirstDateStr=null, sourceXml=null, magXml=null, createTime=1752992077033, creator=13701087609, updateTime=1752992077033, updator=13701087609, issue=Issue{id=1153695641046864317, tenantId=1146029695717560320, journalId=1146031654075715584, year='2024', volume='22', issue='5', pageStart='1', pageEnd='330', issueExtLink='null', onlineDate='null', pubDate='null', beforeIssueId=null, nextIssueId=null, price=null, status=1, issueComplete=1, articleOrder=1, issueType=-1, specialIssue=0, createTime=1752992075453, creator=13701087609, updateTime=1753780969288, updator=13701087609, preIssue=null, nextIssue=null, ext={EN=IssueExt(id=1157004501661078352, tenantId=1146029695717560320, journalId=1146031654075715584, issueId=1153695641046864317, language=EN, specialIssueTitle=, coverIllustrator=, specialIssueEditor=, specialIssueAbout=), CN=IssueExt(id=1157004501661078353, tenantId=1146029695717560320, journalId=1146031654075715584, issueId=1153695641046864317, language=CN, specialIssueTitle=, coverIllustrator=, specialIssueEditor=, specialIssueAbout=)}, issueFiles=null}, startPage=67, endPage=73, ext={EN=ArticleExt(id=1153695648391090654, articleId=1153695647673864667, tenantId=1146029695717560320, journalId=1146031654075715584, language=EN, title=Insertion Loss Prediction of Differential-mode EMI Filter Based on Artificial Neural Networks, columnId=1153695644242923977, journalTitle=Journal of Power Supply, columnName=Special Issue on Electromagnetic Compatibility in Power Electronic Systems, runingTitle=null, highlight=null, articleAbstract=

In power electronic devices, high-speed switching will often lead to serious electromagnetic interference (EMI) problems, which seriously affects the reliability of power electronic systems. To solve these EMI problems, EMI filters are a common solution. The insertion loss is an evaluation index for the noise attenuation capability, and the accuracy of its model directly affects the parameter design accuracy of EMI filters. To improve the prediction accuracy of the EMI filter insertion loss model, accurately describe the system behavior and predict the filtering performance of the EMI filter, and improve the design efficiency of the EMI filter, the insertion loss of a single-stage differential-mode EMI filter is modeled using a back propagation neural network. The proposed neural network model has better practical application value than the ideal model and the behavioral model of a high-frequency circuit, aiming to provide guidance for the design and optimization of EMI filters. This model can quickly evaluate the actual insertion loss of EMI filters to improve their design efficiency.

, correspAuthors=null, 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=Rongliang CHEN, Haiyan LIANG, Yitao LIU), CN=ArticleExt(id=1153695725520146950, articleId=1153695647673864667, tenantId=1146029695717560320, journalId=1146031654075715584, language=CN, title=基于人工神经网络的差模EMI 滤波器插入损耗预测, columnId=1153695644381336012, journalTitle=电源学报, columnName=电力电子系统中的电磁兼容专辑, runingTitle=null, highlight=null, articleAbstract=

在电力电子设备中,高速开关经常会导致严重的电磁干扰EMI(electromagnetic interfere),严重影响电力电子系统的可靠性。为解决EMI问题,EMI滤波器是1种常用的解决方案。插入损耗作为噪声衰减能力的评价指标,其模型的准确性直接影响EMI滤波器的参数设计精度。为了提高EMI滤波器插入损耗模型的预测精度,首先准确描述系统行为并预测 EMI 滤波器滤波性能,提高EMI滤波器设计效率,然后利用反向传播神经网络对单级差模EMI滤波器的插入损耗进行建模。相较于理想模型和高频电路的行为模型,所提神经网络模型预测精度具有更好的实际应用价值,可以快速评估EMI滤波器的实际插入损耗,提高 EMI 滤波器设计效率,并为EMI滤波器的设计和优化提供指导。

, correspAuthors=null, authorNote=null, correspAuthorsNote=null, copyrightStatement=null, copyrightOwner=null, extLink=null, articleAbsUrl=null, sourceXml=1f6o728jDLp6sMtE7F7r4Q==, magXml=soOJNBZEF+v5ohTES1uD0g==, pdfUrl=null, pdf=w4Qieyb31MVhdm0lz70coA==, pdfFileSize=null, pdfExtLink=null, richHtmlUrl=null, mobilePdfUrl=null, reviewReport=null, pdfFirstPage=null, abstractGraph=null, abstractGraphContent=null, abstractVideo=null, citation=null, cebUrl=null, magXmlContent=MoWsjI08xSVww5TPXsooQA==, mapNumber=null, authorCompany=null, fund=null, authors=

陈荣亮(1999-),男,硕士。研究方向:电力电子电磁干扰抑制。E-mail: 2110296050@email.szu.edu.cn。

梁海燕(2001-),女,本科。研究方向:电力电子变换器磁性材料。E-mail: 2310295004@email.szu.edu.cn。

刘艺涛(1986-),男,中国电源学会高级会员,通信作者,博士,副教授。研究方向:电力电子系统电磁干扰/兼容。E-mail: liuyt@szu.edu.cn。

, authorsList=陈荣亮, 梁海燕, 刘艺涛)}, authors=[Author(id=1154032981908902108, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1153695647673864667, orderNo=0, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=2110296050@email.szu.edu.cn, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1154032981984399583, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1153695647673864667, authorId=1154032981908902108, language=EN, stringName=Rongliang CHEN, firstName=Rongliang, middleName=null, lastName=CHEN, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=College of Mechatronics and Control Engineering Shenzhen University Shenzhen 518060 China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1154032982068285667, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1153695647673864667, authorId=1154032981908902108, language=CN, stringName=陈荣亮, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=深圳大学 机电与控制工程学院 深圳 518060, bio={"img":"P9qd0onGywyaD45xsHNDBw==","content":"

陈荣亮(1999-),男,硕士。研究方向:电力电子电磁干扰抑制。E-mail: 2110296050@email.szu.edu.cn。

"}, bioImg=P9qd0onGywyaD45xsHNDBw==, bioContent=

陈荣亮(1999-),男,硕士。研究方向:电力电子电磁干扰抑制。E-mail: 2110296050@email.szu.edu.cn。

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1154032980868714679, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1153695647673864667, xref=null, ext=[AuthorCompanyExt(id=1154032980872908984, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1153695647673864667, companyId=1154032980868714679, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=College of Mechatronics and Control Engineering Shenzhen University Shenzhen 518060 China), AuthorCompanyExt(id=1154032981804044502, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1153695647673864667, companyId=1154032980868714679, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=深圳大学 机电与控制工程学院 深圳 518060)])]), Author(id=1154032982168948967, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1153695647673864667, orderNo=1, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=2310295004@email.szu.edu.cn, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1154032984152854811, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1153695647673864667, authorId=1154032982168948967, language=EN, stringName=Haiyan LIANG, firstName=Haiyan, middleName=null, lastName=LIANG, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=College of Mechatronics and Control Engineering Shenzhen University Shenzhen 518060 China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1154032984232546593, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1153695647673864667, authorId=1154032982168948967, language=CN, stringName=梁海燕, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=深圳大学 机电与控制工程学院 深圳 518060, bio={"content":"

梁海燕(2001-),女,本科。研究方向:电力电子变换器磁性材料。E-mail: 2310295004@email.szu.edu.cn。

"}, bioImg=null, bioContent=

梁海燕(2001-),女,本科。研究方向:电力电子变换器磁性材料。E-mail: 2310295004@email.szu.edu.cn。

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1154032980868714679, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1153695647673864667, xref=null, ext=[AuthorCompanyExt(id=1154032980872908984, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1153695647673864667, companyId=1154032980868714679, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=College of Mechatronics and Control Engineering Shenzhen University Shenzhen 518060 China), AuthorCompanyExt(id=1154032981804044502, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1153695647673864667, companyId=1154032980868714679, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=深圳大学 机电与控制工程学院 深圳 518060)])]), Author(id=1154032984324821287, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1153695647673864667, orderNo=2, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=liuyt@szu.edu.cn, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1154032984832332081, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1153695647673864667, authorId=1154032984324821287, language=EN, stringName=Yitao LIU, firstName=Yitao, middleName=null, lastName=LIU, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=College of Mechatronics and Control Engineering Shenzhen University Shenzhen 518060 China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1154032984891052339, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1153695647673864667, authorId=1154032984324821287, language=CN, stringName=刘艺涛, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=深圳大学 机电与控制工程学院 深圳 518060, bio={"content":"

刘艺涛(1986-),男,中国电源学会高级会员,通信作者,博士,副教授。研究方向:电力电子系统电磁干扰/兼容。E-mail: liuyt@szu.edu.cn。

"}, bioImg=null, bioContent=

刘艺涛(1986-),男,中国电源学会高级会员,通信作者,博士,副教授。研究方向:电力电子系统电磁干扰/兼容。E-mail: liuyt@szu.edu.cn。

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1154032980868714679, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1153695647673864667, xref=null, ext=[AuthorCompanyExt(id=1154032980872908984, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1153695647673864667, companyId=1154032980868714679, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=College of Mechatronics and Control Engineering Shenzhen University Shenzhen 518060 China), AuthorCompanyExt(id=1154032981804044502, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1153695647673864667, companyId=1154032980868714679, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=深圳大学 机电与控制工程学院 深圳 518060)])])], keywords=[Keyword(id=1154032985276928324, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1153695647673864667, language=EN, orderNo=1, keyword=Parasitic parameter), Keyword(id=1154032985348231494, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1153695647673864667, language=EN, orderNo=2, keyword=differential-mode electromagnetic interference (EMI) filter), Keyword(id=1154032985495032136, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1153695647673864667, language=EN, orderNo=3, keyword=insertion loss), Keyword(id=1154032985591501131, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1153695647673864667, language=EN, orderNo=4, keyword=artificial neural network), Keyword(id=1154032985650221390, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1153695647673864667, language=CN, orderNo=1, keyword=寄生参数), Keyword(id=1154032985738301777, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1153695647673864667, language=CN, orderNo=2, keyword=差模EMI滤波器), Keyword(id=1154032985801216339, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1153695647673864667, language=CN, orderNo=3, keyword=插入损耗), Keyword(id=1154032985901879638, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1153695647673864667, language=CN, orderNo=4, keyword=人工神经网络)], refs=[Reference(id=1154032989160853984, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1153695647673864667, doi=null, pmid=null, pmcid=null, year=2021, volume=68, issue=12, pageStart=12296, pageEnd=12309, url=null, language=null, rfNumber=[1], rfOrder=0, authorNames=Jiang Shiqi, Liu Yitao, Ye Xin, et al., journalName=IEEE Transactions on Industrial Electronics, refType=null, unstructuredReference=Jiang Shiqi, Liu Yitao, Ye Xin, et al.. Design of a fully integrated EMI filter for a single-phase grid-connected inverter[J]. IEEE Transactions on Industrial Electronics, 2021. 68(12): 12296-12309., articleTitle=Design of a fully integrated EMI filter for a single-phase grid-connected inverter, refAbstract=null), Reference(id=1154032989232157154, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1153695647673864667, doi=null, pmid=null, pmcid=null, year=2003, volume=null, issue=5, pageStart=17, pageEnd=27, url=null, language=null, rfNumber=[2], rfOrder=1, authorNames=曹才开, journalName=通信电源技术, refType=null, unstructuredReference=曹才开. 开关电源电磁干扰滤波器插入损耗的研究[J]. 通信电源技术, 2003. 5): 17-27., articleTitle=开关电源电磁干扰滤波器插入损耗的研究, refAbstract=null), Reference(id=1154032989299266020, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1153695647673864667, doi=null, pmid=null, pmcid=null, year=2003, volume=null, issue=5, pageStart=17, pageEnd=27, url=null, language=null, rfNumber=[2], rfOrder=2, authorNames=Cao Caikai, journalName=Telecom Power Technology, refType=null, unstructuredReference=Cao Caikai. Research of insert loss for electromagnetic interference filters of switching mode power supply[J]. Telecom Power Technology, 2003. 5): 17-27 (in Chinese)., articleTitle=Research of insert loss for electromagnetic interference filters of switching mode power supply, refAbstract=null), Reference(id=1154032989349597670, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1153695647673864667, doi=null, pmid=null, pmcid=null, year=2020, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[3], rfOrder=3, authorNames=陆德政, journalName=null, refType=null, unstructuredReference=陆德政. 基于神经网络的电磁干扰滤波器建模及设计研究[D]. 杭州: 浙江大学, 2020., articleTitle=基于神经网络的电磁干扰滤波器建模及设计研究, refAbstract=null), Reference(id=1154032989404123624, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1153695647673864667, doi=null, pmid=null, pmcid=null, year=2020, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[3], rfOrder=4, authorNames=Lu Dezheng, journalName=null, refType=null, unstructuredReference=Lu Dezheng. Research on modeling and design of electromagnetic interference filter based on neural network[D]. 杭州: 浙江大学, 2020. (in Chinese)., articleTitle=Research on modeling and design of electromagnetic interference filter based on neural network, refAbstract=null), Reference(id=1154032989462843884, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1153695647673864667, doi=null, pmid=null, pmcid=null, year=2021, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[4], rfOrder=5, authorNames=陈健, journalName=null, refType=null, unstructuredReference=陈健. 基于器件有限元建模的EMI滤波器高频特性研究[D]. 南京: 南京师范大学, 2021., articleTitle=基于器件有限元建模的EMI滤波器高频特性研究, refAbstract=null), Reference(id=1154032989630616048, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1153695647673864667, doi=null, pmid=null, pmcid=null, year=2022, volume=64, issue=4, pageStart=1086, pageEnd=1096, url=null, language=null, rfNumber=[5], rfOrder=6, authorNames=Negri S, Spadacini G, Grassi F, et al., journalName=IEEE Transactions on Electromagnetic Compatibility, refType=null, unstructuredReference=Negri S, Spadacini G, Grassi F, et al.. Prediction of EMI filter attenuation in power-electronic converters via circuit simulation[J]. IEEE Transactions on Electromagnetic Compatibility, 2022. 64(4): 1086-1096., articleTitle=Prediction of EMI filter attenuation in power-electronic converters via circuit simulation, refAbstract=null), Reference(id=1154032989756445172, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1153695647673864667, doi=null, pmid=null, pmcid=null, year=2021, volume=36, issue=4, pageStart=4633, pageEnd=4658, url=null, language=null, rfNumber=[6], rfOrder=7, authorNames=Zhao Shuai, Blaabjerg F, Wang Huai, journalName=IEEE Transactions on Power Electronics, refType=null, unstructuredReference=Zhao Shuai, Blaabjerg F, Wang Huai. An overview of artificial intelligence applications for power electronics[J]. IEEE Transactions on Power Electronics, 2021. 36(4): 4633-4658., articleTitle=An overview of artificial intelligence applications for power electronics, refAbstract=null), Reference(id=1154032989815165431, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1153695647673864667, doi=null, pmid=null, pmcid=null, year=2019, volume=34, issue=8, pageStart=7161, pageEnd=7171, url=null, language=null, rfNumber=[7], rfOrder=8, authorNames=Dragičević T, Wheeler P, Blaabjerg F, journalName=IEEE Transactions on Power Electronics, refType=null, unstructuredReference=Dragičević T, Wheeler P, Blaabjerg F. Artificial intelligence aided automated design for reliability of power electronic systems[J]. IEEE Transactions on Power Electronics, 2019. 34(8): 7161-7171., articleTitle=Artificial intelligence aided automated design for reliability of power electronic systems, refAbstract=null), Reference(id=1154032989873885691, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1153695647673864667, doi=null, pmid=null, pmcid=null, year=2022, volume=69, issue=11, pageStart=11144, pageEnd=11155, url=null, language=null, rfNumber=[8], rfOrder=9, authorNames=Li Xinze, Zhang Xin, Lin Fanfan, et al., journalName=IEEE Transactions on Industrial Electronics, refType=null, unstructuredReference=Li Xinze, Zhang Xin, Lin Fanfan, et al.. Artificial-intelligence-based design for circuit parameters of power converters[J]. IEEE Transactions on Industrial Electronics, 2022. 69(11): 11144-11155., articleTitle=Artificial-intelligence-based design for circuit parameters of power converters, refAbstract=null), Reference(id=1154032989974548991, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1153695647673864667, doi=null, pmid=null, pmcid=null, year=2018, volume=null, issue=null, pageStart=1702, pageEnd=1706, url=null, language=null, rfNumber=[9], rfOrder=10, authorNames=null, journalName=2018 International Conference on Advances in Computing, Communications and Informatics (ICACCI). Bangalore, India, refType=null, unstructuredReference=Performance enhancement of a series active power filter using Kalman filter based neural network control strategy[C]// 2018 International Conference on Advances in Computing, Communications and Informatics (ICACCI). Bangalore, India, 2018: 1702-1706., articleTitle=Performance enhancement of a series active power filter using Kalman filter based neural network control strategy, refAbstract=null), Reference(id=1154032990037463554, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1153695647673864667, doi=null, pmid=null, pmcid=null, year=2016, volume=null, issue=null, pageStart=1, pageEnd=6, url=null, language=null, rfNumber=[10], rfOrder=11, authorNames=null, journalName=2016 IEEE 7th Power India International Conference (PI-ICON). Bikaner, India, refType=null, unstructuredReference=Artificial neural network based three phase shunt active power filter[C]// 2016 IEEE 7th Power India International Conference (PI-ICON). Bikaner, India, 2016: 1-6., articleTitle=Artificial neural network based three phase shunt active power filter, refAbstract=null), Reference(id=1154032990129738249, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1153695647673864667, doi=null, pmid=null, pmcid=null, year=2020, volume=1, issue=null, pageStart=284, pageEnd=299, url=null, language=null, rfNumber=[11], rfOrder=12, authorNames=Guillod T, Papamanolis P, W Kolar J, journalName=IEEE Open Journal of Power Electronics, refType=null, unstructuredReference=Guillod T, Papamanolis P, W Kolar J. Artificial neural network (ANN) based fast and accurate inductor modeling and design[J]. IEEE Open Journal of Power Electronics, 2020. 1: 284-299., articleTitle=Artificial neural network (ANN) based fast and accurate inductor modeling and design, refAbstract=null), Reference(id=1154032990217818636, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1153695647673864667, doi=null, pmid=null, pmcid=null, year=2021, volume=36, issue=9, pageStart=9704, pageEnd=9710, url=null, language=null, rfNumber=[12], rfOrder=13, authorNames=Tang Yuanhong, Hu Weihao, Cao Di, et al., journalName=IEEE Transactions on Power Electronics, refType=null, unstructuredReference=Tang Yuanhong, Hu Weihao, Cao Di, et al.. Artificial intelligence-aided minimum reactive power control for the DAB converter based on harmonic analysis method[J]. IEEE Transactions on Power Electronics, 2021. 36(9): 9704-9710., articleTitle=Artificial intelligence-aided minimum reactive power control for the DAB converter based on harmonic analysis method, refAbstract=null)], funds=[Fund(id=1154032988913390040, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1153695647673864667, awardId=JCYJ20230808104910021, language=EN, fundingSource=Shenzhen Science and Technology Program(JCYJ20230808104910021), fundOrder=null, country=null), Fund(id=1154032988967915997, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1153695647673864667, awardId=JCYJ20230808104910021, language=CN, fundingSource=深圳市科技计划资助项目(JCYJ20230808104910021), fundOrder=null, country=null)], companyList=[AuthorCompany(id=1154032980868714679, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1153695647673864667, xref=null, ext=[AuthorCompanyExt(id=1154032980872908984, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1153695647673864667, companyId=1154032980868714679, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=College of Mechatronics and Control Engineering Shenzhen University Shenzhen 518060 China), AuthorCompanyExt(id=1154032981804044502, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1153695647673864667, companyId=1154032980868714679, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=深圳大学 机电与控制工程学院 深圳 518060)])], figs=[ArticleFig(id=1154032986862375283, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1153695647673864667, language=EN, label=Fig. 1, caption=Topology of differential-mode EMI filter, figureFileSmall=yPZmFmqyN88YqcYW94dMJQ==, figureFileBig=j+bV4jO/UiZtDySR5ldBpQ==, tableContent=null), ArticleFig(id=1154032986933678457, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1153695647673864667, language=CN, label=图1, caption=差模 EMI 滤波器拓扑, figureFileSmall=yPZmFmqyN88YqcYW94dMJQ==, figureFileBig=j+bV4jO/UiZtDySR5ldBpQ==, tableContent=null), ArticleFig(id=1154032987055313280, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1153695647673864667, language=EN, label=Fig. 2, caption=Measurement of differential-mode insertion loss of EMI filter, figureFileSmall=1NeiMz62HLkYm2gim5Rr3w==, figureFileBig=77xFqIkzd6l6Cbrr7jTTZg==, tableContent=null), ArticleFig(id=1154032987147587974, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1153695647673864667, language=CN, label=图2, caption=EMI 滤波器的差模插入损耗测量, figureFileSmall=1NeiMz62HLkYm2gim5Rr3w==, figureFileBig=77xFqIkzd6l6Cbrr7jTTZg==, tableContent=null), ArticleFig(id=1154032987227279755, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1153695647673864667, language=EN, label=Fig. 3, caption=Equivalent model of differential-mode EMI filter, figureFileSmall=iXHYyW+qeXiOG9t5XIDY+w==, figureFileBig=OyzJ2P6a+6YKrwoIidx+DQ==, tableContent=null), ArticleFig(id=1154032987311165838, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1153695647673864667, language=CN, label=图3, caption=差模 EMI 滤波器等效模型, figureFileSmall=iXHYyW+qeXiOG9t5XIDY+w==, figureFileBig=OyzJ2P6a+6YKrwoIidx+DQ==, tableContent=null), ArticleFig(id=1154032987403440529, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1153695647673864667, language=EN, label=Fig. 4, caption=Difference between theoretical and measured insertion loss curves, figureFileSmall=CqXTbNAA02zdte2AIn9YaA==, figureFileBig=3B/MEWk+Ip91Lb+I36K9Ug==, tableContent=null), ArticleFig(id=1154032987483132308, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1153695647673864667, language=CN, label=图4, caption=插入损耗理论与实测曲线区别, figureFileSmall=CqXTbNAA02zdte2AIn9YaA==, figureFileBig=3B/MEWk+Ip91Lb+I36K9Ug==, tableContent=null), ArticleFig(id=1154032987567018394, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1153695647673864667, language=EN, label=Fig. 5, caption=Differential-mode insertion loss data acquisition platform, figureFileSmall=4tLGj2fIpu/SmRQnjWyX+w==, figureFileBig=//KdvU+ZBR1KZiBkCwKksg==, tableContent=null), ArticleFig(id=1154032987642515870, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1153695647673864667, language=CN, label=图5, caption=差模插入损耗数据采集平台, figureFileSmall=4tLGj2fIpu/SmRQnjWyX+w==, figureFileBig=//KdvU+ZBR1KZiBkCwKksg==, tableContent=null), ArticleFig(id=1154032987764150694, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1153695647673864667, language=EN, label=Fig. 6, caption=ANN model of differential-mode EMI filter, figureFileSmall=0JNgNWhLXCDlGMuKAttuFg==, figureFileBig=JZW0Gqc9dyWpPHCAwDMhuQ==, tableContent=null), ArticleFig(id=1154032987877396906, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1153695647673864667, language=CN, label=图6, caption=差模 EMI 滤波器的 ANN 模型, figureFileSmall=0JNgNWhLXCDlGMuKAttuFg==, figureFileBig=JZW0Gqc9dyWpPHCAwDMhuQ==, tableContent=null), ArticleFig(id=1154032987952894381, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1153695647673864667, language=EN, label=Fig. 7, caption=Iterations of ANN training results, figureFileSmall=Y1O7xs4F5+GYri0MmJawCw==, figureFileBig=/HdkiCW0B+Tv5K4QLfTh0g==, tableContent=null), ArticleFig(id=1154032988015808943, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1153695647673864667, language=CN, label=图7, caption=ANN 训练结果迭代, figureFileSmall=Y1O7xs4F5+GYri0MmJawCw==, figureFileBig=/HdkiCW0B+Tv5K4QLfTh0g==, tableContent=null), ArticleFig(id=1154032988154220978, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1153695647673864667, language=EN, label=Fig. 8, caption=Worst fitting results, figureFileSmall=MdhN6U/A8fkrWO1MxDH8xQ==, figureFileBig=X1IFQxK95fJdA4ybRztFag==, tableContent=null), ArticleFig(id=1154032988212941237, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1153695647673864667, language=CN, label=图8, caption=拟合效果最差结果, figureFileSmall=MdhN6U/A8fkrWO1MxDH8xQ==, figureFileBig=X1IFQxK95fJdA4ybRztFag==, tableContent=null), ArticleFig(id=1154032988301021625, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1153695647673864667, language=EN, label=Fig. 9, caption=Root mean square error of training set, figureFileSmall=+7gLrHVcVAdarISzCDlVRQ==, figureFileBig=husWCRoyqivboC+HRQdAeg==, tableContent=null), ArticleFig(id=1154032988359741884, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1153695647673864667, language=CN, label=图9, caption=训练集均方根误差, figureFileSmall=+7gLrHVcVAdarISzCDlVRQ==, figureFileBig=husWCRoyqivboC+HRQdAeg==, tableContent=null), ArticleFig(id=1154032988410073535, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1153695647673864667, language=EN, label=Fig. 10, caption=Insertion loss prediction results corresponding to maximum and minimum ${R}^{2}$ after denormalization of validation set, figureFileSmall=UaIqRNN2JGcvwG68nVA1FQ==, figureFileBig=h4HSqM8/V0ClkQrX5zDVlQ==, tableContent=null), ArticleFig(id=1154032988485571014, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1153695647673864667, language=CN, label=图10, caption=验证集反归一化后 ${\mathbf{R}}^{2}$ 最大与最小对应的插入损耗预测结果, figureFileSmall=UaIqRNN2JGcvwG68nVA1FQ==, figureFileBig=h4HSqM8/V0ClkQrX5zDVlQ==, tableContent=null), ArticleFig(id=1154032988544291271, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1153695647673864667, language=EN, label=Fig. 11, caption=Mean square error of validation set, figureFileSmall=QKkpAl6/ANIa37hurhbsDg==, figureFileBig=KCLgcbS4TgU5rBjzjTyUXQ==, tableContent=null), ArticleFig(id=1154032988611400138, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1153695647673864667, language=CN, label=图11, caption=验证集均方误差, figureFileSmall=QKkpAl6/ANIa37hurhbsDg==, figureFileBig=KCLgcbS4TgU5rBjzjTyUXQ==, tableContent=null), ArticleFig(id=1154032988678509007, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1153695647673864667, language=EN, label=Fig. 12, caption=Total mean square error based on ANN, figureFileSmall=/n3fJm2ErnqzUXgRa4PXgA==, figureFileBig=UkIC+sEwQcgcjAyrp4JSvQ==, tableContent=null), ArticleFig(id=1154032988741423568, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1153695647673864667, language=CN, label=图12, caption=ANN 总均方误差, figureFileSmall=/n3fJm2ErnqzUXgRa4PXgA==, figureFileBig=UkIC+sEwQcgcjAyrp4JSvQ==, tableContent=null)], attaches=null, journal=Journal(id=1046111678587809797, delFlag=0, nameCn=电源学报, nameEn=Journal of Power Supply, nameHistory1=null, nameHistory2=null, issn=2095-2805, eissn=, cn=12-1420/TM, coden=null, periodic=bio-monthly, 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=Mx+A2dn+ULnPHuEAI1LruQ==, journalPrice=null, startedYear=null, abbrevIsoEn=J Power Supp, journalRemark=null, publicationField=null, createdTime=null, updatedTime=1759802942253, createdBy=null, updatedBy=18614031015, firstLetterCn=J, firstLetterEn=J, subjectCode=Engineering, subjectName=工程, subjectCodeEn=Engineering, subjectNameEn=null, picCn=Mx+A2dn+ULnPHuEAI1LruQ==, picEn=yHt2vwjzkDgqh+JDCfJKoQ==, jcr=null, cjcr=null, exts=[JournalExt(id=1162453073839375337, language=CN, name=电源学报, 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.jops.cn/EN/home, createdTime=1755080010137, updatedTime=1755080010137, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=http://www.jops.cn/CN/column/column7.shtml, submissionAuthorUrl=https://mc03.manuscriptcentral.com/jops, submissionEditorUrl=https://mc03.manuscriptcentral.com/jops, submissionReviewUrl=https://mc03.manuscriptcentral.com/jops, submissionCeEditorUrl=https://mc03.manuscriptcentral.com/jops, submissionAeEditorUrl=https://mc03.manuscriptcentral.com/jops, option={"copyright":""}), JournalExt(id=1162453073902289898, language=EN, name=Journal of Power Supply, 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.jops.cn/CN/home, createdTime=1755080010152, updatedTime=1755080010152, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=http://www.jops.cn/EN/column/column7.shtml, submissionAuthorUrl=https://mc03.manuscriptcentral.com/jops, submissionEditorUrl=https://mc03.manuscriptcentral.com/jops, submissionReviewUrl=https://mc03.manuscriptcentral.com/jops, submissionCeEditorUrl=https://mc03.manuscriptcentral.com/jops, submissionAeEditorUrl=https://mc03.manuscriptcentral.com/jops, option={"copyright":""})], databaseList=null, tenantJournalId=1146031654075715584, websiteList=[Website(id=1146832214672683008, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1146031654075715584, 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/dyxb/EN, language=EN, createTime=1751355707101, createBy=18614031015, updateTime=1753435268747, updateBy=18614031015, name=电源学报-英文站点, tplId=1146101810881728533, title=电源学报, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1155559379819679852, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1146832214672683008, code=articleTextType, value=kx, createTime=1753436425404, updateTime=1753436425404, creator=18614031015, updator=18614031015), WebsiteProps(id=1155559379798708329, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1146832214672683008, code=banner, value=null, createTime=1753436425399, updateTime=1753436425399, creator=18614031015, updator=18614031015), WebsiteProps(id=1155559379781931112, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1146832214672683008, code=logo, value=https://castjournals.cast.org.cn/joweb/kjdb/CN/file/pic?fileId=efYTu4aDDzS8GgTA1MjEKw==, createTime=1753436425396, updateTime=1753436425396, creator=18614031015, updator=18614031015), WebsiteProps(id=1155559379811291243, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1146832214672683008, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/kjdb/CN/file/pic, createTime=1753436425402, updateTime=1753436425402, creator=18614031015, updator=18614031015), WebsiteProps(id=1155559379802902634, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1146832214672683008, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_cn_619/, createTime=1753436425400, updateTime=1753436425400, creator=18614031015, updator=18614031015)]), Website(id=1148243202240405915, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1146031654075715584, 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/dyxb/CN, language=CN, createTime=1751692112741, createBy=18614031015, updateTime=1753435242839, updateBy=18614031015, name=电源学报-中文站点, tplId=1146099689490845704, title=电源学报, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1148618015060553758, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202240405915, code=articleTextType, value=kx, createTime=1751781475081, updateTime=1751781475081, creator=18614031015, updator=18614031015), WebsiteProps(id=1148618015035387931, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202240405915, code=banner, value=null, createTime=1751781475075, updateTime=1751781475075, creator=18614031015, updator=18614031015), WebsiteProps(id=1148618015022805018, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202240405915, code=logo, value=https://castjournals.cast.org.cn/joweb/kjdb/CN/file/pic?fileId=efYTu4aDDzS8GgTA1MjEKw==, createTime=1751781475072, updateTime=1751781475072, creator=18614031015, updator=18614031015), WebsiteProps(id=1148618015052165149, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202240405915, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/kjdb/CN/file/pic, createTime=1751781475079, updateTime=1751781475079, creator=18614031015, updator=18614031015), WebsiteProps(id=1148618015043776540, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202240405915, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_cn_619/, createTime=1751781475077, updateTime=1751781475077, creator=18614031015, updator=18614031015)])], journalTitle=电源学报, weixinUrl=null, journalUrl=http://www.jops.cn/CN/home, iacademicId=null, status=0, seqNo=null, journalTitleEn=Journal of Power Supply, journalPhotoCn=Mx+A2dn+ULnPHuEAI1LruQ==, journalPhotoEn=yHt2vwjzkDgqh+JDCfJKoQ==, journalFirstLetter=J, 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/dyxb/CN/10.13234/j.issn.2095-2805.2024.5.67, detailUrlEn=https://castjournals.cast.org.cn/joweb/dyxb/EN/10.13234/j.issn.2095-2805.2024.5.67, pdfUrlCn=https://castjournals.cast.org.cn/joweb/dyxb/CN/PDF/10.13234/j.issn.2095-2805.2024.5.67, pdfUrlEn=https://castjournals.cast.org.cn/joweb/dyxb/EN/PDF/10.13234/j.issn.2095-2805.2024.5.67, aliStartDate=null, aliEndDate=null, collectionFlag=false, citedCount=null, citedUrl=null, reference=null)
收藏切换
基于人工神经网络的差模EMI 滤波器插入损耗预测
收藏切换
PDF下载
陈荣亮 , 梁海燕 , 刘艺涛
电源学报 | 电力电子系统中的电磁兼容专辑 2024,22(5): 67-73
收起
收藏切换
电源学报 | 电力电子系统中的电磁兼容专辑 2024, 22(5): 67-73
基于人工神经网络的差模EMI 滤波器插入损耗预测
全屏
陈荣亮 , 梁海燕 , 刘艺涛
作者信息
  • 深圳大学 机电与控制工程学院 深圳 518060
  • 陈荣亮(1999-),男,硕士。研究方向:电力电子电磁干扰抑制。E-mail: 2110296050@email.szu.edu.cn。

    梁海燕(2001-),女,本科。研究方向:电力电子变换器磁性材料。E-mail: 2310295004@email.szu.edu.cn。

    刘艺涛(1986-),男,中国电源学会高级会员,通信作者,博士,副教授。研究方向:电力电子系统电磁干扰/兼容。E-mail: liuyt@szu.edu.cn。

Insertion Loss Prediction of Differential-mode EMI Filter Based on Artificial Neural Networks
Rongliang CHEN , Haiyan LIANG , Yitao LIU
Affiliations
  • College of Mechatronics and Control Engineering Shenzhen University Shenzhen 518060 China
出版时间: 2024-09-30 doi: 10.13234/j.issn.2095-2805.2024.5.67
文章导航
收藏切换

在电力电子设备中,高速开关经常会导致严重的电磁干扰EMI(electromagnetic interfere),严重影响电力电子系统的可靠性。为解决EMI问题,EMI滤波器是1种常用的解决方案。插入损耗作为噪声衰减能力的评价指标,其模型的准确性直接影响EMI滤波器的参数设计精度。为了提高EMI滤波器插入损耗模型的预测精度,首先准确描述系统行为并预测 EMI 滤波器滤波性能,提高EMI滤波器设计效率,然后利用反向传播神经网络对单级差模EMI滤波器的插入损耗进行建模。相较于理想模型和高频电路的行为模型,所提神经网络模型预测精度具有更好的实际应用价值,可以快速评估EMI滤波器的实际插入损耗,提高 EMI 滤波器设计效率,并为EMI滤波器的设计和优化提供指导。

寄生参数  /  差模EMI滤波器  /  插入损耗  /  人工神经网络

In power electronic devices, high-speed switching will often lead to serious electromagnetic interference (EMI) problems, which seriously affects the reliability of power electronic systems. To solve these EMI problems, EMI filters are a common solution. The insertion loss is an evaluation index for the noise attenuation capability, and the accuracy of its model directly affects the parameter design accuracy of EMI filters. To improve the prediction accuracy of the EMI filter insertion loss model, accurately describe the system behavior and predict the filtering performance of the EMI filter, and improve the design efficiency of the EMI filter, the insertion loss of a single-stage differential-mode EMI filter is modeled using a back propagation neural network. The proposed neural network model has better practical application value than the ideal model and the behavioral model of a high-frequency circuit, aiming to provide guidance for the design and optimization of EMI filters. This model can quickly evaluate the actual insertion loss of EMI filters to improve their design efficiency.

Parasitic parameter  /  differential-mode electromagnetic interference (EMI) filter  /  insertion loss  /  artificial neural network
陈荣亮, 梁海燕, 刘艺涛. 基于人工神经网络的差模EMI 滤波器插入损耗预测. 电源学报, 2024 , 22 (5) : 67 -73 . DOI: 10.13234/j.issn.2095-2805.2024.5.67
Rongliang CHEN, Haiyan LIANG, Yitao LIU. Insertion Loss Prediction of Differential-mode EMI Filter Based on Artificial Neural Networks[J]. Journal of Power Supply, 2024 , 22 (5) : 67 -73 . DOI: 10.13234/j.issn.2095-2805.2024.5.67
电力电子设备中的电磁干扰 EMI(electromagnetic interfere)主要为传导 EMI, 而实际情景中传导 EMI 的抑制主要通过无源 EMI 滤波器来实现${}^{\Pi }$。EMI 滤波器通常使用插入损耗 IL(insertion loss) 作为其设计依据, 但是由于无源器件的耦合效应和寄生参数的影响, 在高频段传统 IL 的预测误差较大。文献[2]使用了切比雪夫修正系数法和频段修正法改善 EMI 滤波器的插入损耗, 但在频率高于 1 MHz 后的预测误差较大;文献[3]提出 1 种循环神经网络的方法对 IL 进行预测, 但模型训练需要花费较长时间;文献[4]通过有限元仿真的方式建立了精确的 EMI 滤波器模型,但建模方式过于复杂且对计算机性能要求较高; 文献[5]通过电路仿真的形式建立 EMI 滤波器的行为模型, 实现了对共模 IL 的精确预测,但差模 IL 预测存在较大误差。机器学习 ML (machine learning)技术和计算机技术的发展一直促进着电力电子技术的进步[6]。在电力电子系统中,人工智能技术 AI(artificial intelligence)得到了广泛应用[7]。在功率变换器的参数设计中,机器学习技术可以加快设计过程同时提高其准确性[8]。文献[9-10]应用人工神经网络 ANN(artificial neural network)对有源 EMI 滤波器 AEF(active EMI filter)进行建模, 提高了 EMI 滤波器的电磁干扰抑制性能。
综上所述, 目前基于人工神经网络的元件建模[11]与控制器优化设计[12] 较为常见,而使用人工智能对无源 EMI 滤波器建模与设计方面的应用较少。无源 EMI 滤波器分为共模 EMI 滤波器和差模 EMI 滤波器, 本文针对差模 EMI 滤波器, 提出 1 种基于人工神经网络的差模 EMI 插入损耗预测方法,旨在提高其参数设计的准确性,为 EMI 滤波器的设计和优化提供指导。
本文以传统$\pi$ 型差模 EMI 滤波器原型为研究对象, 其拓扑结构如图1所示。该差模 EMI 滤波器主要由电容${C}_{\mathrm{X}1}\text{、}{C}_{\mathrm{X}2}$ 和电感${L}_{\mathrm{{DM}}}$ 组成。通常差模电感${L}_{\mathrm{{DM}}}$ 采用对称拆分的形式,可以提供部分共模干扰抑制能力且整体体积较小。EMI 滤波器的特性是用插入损耗来描述的, 因此设计 EMI 滤波器首先要计算其插入损耗。EMI 滤波器的差模插入损耗测量, 如图2所示。
图2中, 设插入待测 EMI 滤波器前的测量仪器读数为${U}_{1}$,插入待测 EMI 滤波器后的测量仪器读数为${U}_{2}$,则该滤波器的插入损耗 IL 为
$\mathrm{{IL}}= {20}\lg {U}_{1}- {20}\lg {U}_{2}= {20}\lg \left(\frac{{U}_{1}}{{U}_{2}}\right)$
本文采用的测量仪器为矢量网络分析仪 VNA (vector network analyzer), 直接对 EMI 滤波器端口参数进行测量来表征 EMI 滤波器的插入损耗。当电路尺寸与工作波长相比拟时,一般采用分布式参数对端口网络进行描述,其中散射$S$ (scattering) 是分布式参数的 1 种。将差模 EMI 滤波器看作二端口网络, 则插入损耗又可以定义为滤波器接入前、 后从噪声源传输到负载的功率之比,结合参数$S$, 滤波器的插入损耗 IL 可以表示为
$\mathrm{{IL}}= {10}\lg \left(\frac{{\left|{S}_{21}\right|}^{2}}{1 -{\left|{S}_{11}\right|}^{2}}\right)$
式中:${S}_{11}$ 为输入端口电压的反射系数,即端口 2 匹配时,端口 1 的反射系数;${S}_{21}$ 为正向电压增益,即端口 2 匹配时, 端口 1 到端口 2 的正向传输系数。 差模 EMI 滤波器可以简化为如图3所示的 CLC 电路来分别进行参数设计。
根据图3电路的频率响应可以得到差模 EMI 滤波器抑制噪声所需的截止频率${f}_{\mathrm{c}\text{-DM, req }}$,且应满足
${f}_{\mathrm{c}- \mathrm{{DM}},\text{ req }}= \min \left({{f}_{\mathrm{T}}\times {10}^{-\frac{{V}_{\mathrm{{DM}},\text{ req }}}{60}}}\right)$
式中:${f}_{\mathrm{T}}$ 为通过 LISN 测试得到的 EMI 不符合电磁兼容标准的最大的频率;${V}_{\mathrm{{DM}},\text{ req }}$${f}_{\mathrm{T}}$ 所需要的衰减量(dB)加上$6\mathrm{{dB}}\mu \mathrm{V}$ 的裕量。差模$\mathrm{{EMI}}$ 滤波器的截止频率计算公式为
${f}_{\mathrm{c},\mathrm{{DM}}}\approx \frac{1}{2\pi }\sqrt[3]{\frac{1}{2{L}_{\mathrm{{DM}}}{C}_{\mathrm{X}1}{C}_{\mathrm{X}2}{R}_{\mathrm{{LISN}},\mathrm{{DM}}}}}$
式中,${R}_{\mathrm{{LISN}},\mathrm{{DM}}}$ 为 LISN 的差模阻抗。
根据式(4)即可确定差模 EMI 滤波器的截止频率大小, 进而确定滤波器的参数。此外, 根据图3(a)可以推导出未插入无源 EMI 滤波器时差模输出的电压表达式, 即
${v}_{\mathrm{{DMI}}}= \frac{{i}_{s\mathrm{{DM}}}{Z}_{s\mathrm{{DM}}}}{{Z}_{s\mathrm{{DM}}}+ {R}_{\mathrm{{LISN}},\mathrm{{DM}}}}{R}_{\mathrm{{LISN}},\mathrm{{DM}}}\approx {i}_{s\mathrm{{DM}}}{R}_{\mathrm{{LISN}},\mathrm{{DM}}}$
式中:${i}_{s\mathrm{{DM}}}$ 为差模噪声电流;${Z}_{s\mathrm{{DM}}}$ 为差模噪声阻抗, 通常视为无穷大。
根据图3(b)可以推导出插入无源 EMI 滤波器后差模输出的电压,并使用$\mathrm{s}$ 域(复频域)的复变量$s$ 表示为
${v}_{\mathrm{{DM}}2}= {i}_{s\mathrm{{DM}}}{R}_{\mathrm{{LISN}},\mathrm{{DM}}}/\left\lbrack {2{L}_{\mathrm{{DM}}}{C}_{\mathrm{X}1}{C}_{\mathrm{X}2}{R}_{\mathrm{{LISN}},\mathrm{{DM}}}{s}^{3}+ }\right.\\\left.{2{L}_{\mathrm{{DM}}}{C}_{\mathrm{X}2}{s}^{2}+ {R}_{\mathrm{{LISN}},\mathrm{{DM}}}\left({{C}_{\mathrm{{Xl}}}+ {C}_{\mathrm{X}2}}\right) s + 1}\right\rbrack $
结合式(5)~式(6)可以得到差模 EMI 滤波器的插入损耗增益$s$ 域的表达式${G}_{\mathrm{{DM}}}\left( s\right)$,即
${G}_{\mathrm{{DM}}}\left( s\right)= \frac{{v}_{\mathrm{{DM1}}}\left( s\right)}{{v}_{\mathrm{{DM2}}}\left( s\right)} = 2{L}_{\mathrm{{DM}}}{C}_{\mathrm{X}1}{C}_{\mathrm{X}2}{R}_{\mathrm{{LISN}},\mathrm{{DM}}}{s}^{3}+ \\ 2{L}_{\mathrm{{DM}}}{C}_{\mathrm{X}2}{s}^{2}+ {R}_{\mathrm{{LISN}},\mathrm{{DM}}}\left({{C}_{\mathrm{X}1}+ {C}_{\mathrm{X}2}}\right) s + 1 $
差模 EMI 滤波器的插入损耗如图4所示, 在高频段由于寄生参数和元器件耦合效应的影响, 实际测量的 IL 与理论公式(7)所预测的结果有较大的偏差。
根据第 1 节所述, 无源 EMI 滤波器的特性可以通过插入损耗来描述, 但是在其公式推导过程中采用了较多如式(4)~式(5)所示的近似处理,降低了对无源 EMI 滤波器建模的准确性。
单纯通过公式推导和计算获得的插入损耗与实际的测量值会有较大偏差, 而较大的滤波器插入损耗误差将直接造成工程师为了获得满足电磁干扰抑制性能要求的参数而进行反复迭代计算,工作效率低下且对工程师的经验依赖性较大。因此, 为了提高工作效率并降低 EMI 滤波器设计过程中对工程师经验的依赖性, 更好地预测无源 EMI 滤波器的实际插入损耗, 降低预测值与实际值的误差, 为 EMI 滤波器的设计提供精确的指导并减少设计时间,本文将差模 EMI 滤波器看作 1 个黑箱,通过使用人工神经网络对差模 EMI 滤波器的插入损耗 IL 进行建模, 实现 IL 的精确预测, 为后续的 EMI 滤波器设计提供可靠的性能指标。
实现差模 EMI 滤波器 IL 人工神经网络模型建立的步骤为
步骤 1 选取合适的输入和输出变量: 将差模 EMI 滤波器看作 1 个黑箱, 研究输入、输出变量之间的相互关系。选取合适的滤波器参数作为神经网络的输入、输出变量。在训练和预测过程中, 这些变量将作为神经网络的输入和输出数据。
步骤 2 样本数据集的建立: 样本数据集用于训练和验证神经网络。根据步骤 1 确定的输入、输出变量在实际的系统中使用测量仪器采集数据。
步骤 3 神经网络的结构选取: 神经网络的结构包括层数、每层神经元数、使用的激活函数类型, 需要根据拟合非线性函数的特点及输入、输出参数的个数进行确定。
步骤 4 训练神经网络:在训练过程中,神经网络从样本数据集中选取一部分作为训练数据集并使用训练数据集进行学习, 优化其内部参数, 如权重和偏差等,以最小化预测误差。
步骤 5 神经网络的预测: 将样本数据集中除训练集外的其他数据作为神经网络模型的拟合效果验证数据集, 用于测试神经网络的拟合性能, 并对预测结果进行分析。
结合式(1)和式(7)可以得到共模和差模插入损耗的表达式${\mathrm{{IL}}}_{\mathrm{{DM}}}$,即
${\mathrm{{IL}}}_{\mathrm{{DM}}}= {20}\lg \left({\begin{Vmatrix}{{G}_{\mathrm{{DM}}}\left( s\right)}\end{Vmatrix}!}\right)= f\left({{L}_{\mathrm{{DM}}},{C}_{\mathrm{X}1},{C}_{\mathrm{X}2}, f}\right)$
根据式 (8) 可知,${\mathrm{{IL}}}_{\mathrm{{DM}}}$${L}_{\mathrm{{DM}}}\text{、}{C}_{\mathrm{X}1}\text{、}{C}_{\mathrm{X}2}$$f$ 的函数, 因此可推断出:无源差模 EMI 滤波器的插入损耗主要受电感、电容和频率的影响。然而,插入损耗的考虑一般是在 1 个固定的频率范围, 相当于 1 个常量,因此本文选择无源 EMI 滤波器的电容${C}_{\mathrm{X}1}\text{、}{C}_{\mathrm{X}2}$ 、 和电感${L}_{\mathrm{{DM}}}$ 作为$\mathrm{{ANN}}$ 的输入变量,而输出变量则为插入损耗${\mathrm{{IL}}}_{\mathrm{{DM}}}$
针对差模 EMI 滤波器的输入变量${C}_{\mathrm{X}1}$${C}_{\mathrm{X}2}$, 本文设计 5 种候选方案,分别为 0.10、0.22、0.47、${1.00}\text{、}{2.00\mu }\mathrm{F}$,并针对差模电感${L}_{\mathrm{{DM}}}$ 设计 3 种选择方案,分别为${30}\text{、}{70}\text{、}{100\mu }\mathrm{H}$。为了更好地表征差模插入损耗曲线, 本文将输出变量定义为频率范围${100}\mathrm{{kHz}}\sim {30}\mathrm{{MHz}}$,并选择该频率范围内差模插入损耗曲线中的 101 个特征点作为神经网络的输出变量。
差模 EMI 滤波器插入损耗采集平台如图5所示, 将差模 EMI 滤波器看作 1 个二端口网络, 通过矢量网络分析仪 Bode 100 按照图5中方式进行链接。通过对 3 个输入变量${C}_{\mathrm{X}1}\text{、}{C}_{\mathrm{X}2}\text{、}{L}_{\mathrm{{DM}}}$ 候选方案进行不同方案间的配对和组合, 可以得出 75 种组合结果。测量过程中,环境干扰或线路接触不良会导致所采集的插入损耗曲线数据存在较大的尖峰干扰, 与其余大部分正常数据特征存在明显差异, 因此需要剔除错误数据。本文通过对测量曲线的评估发现, 有 15 组数据存在异常, 剔除这 15 组异常数据,最终确定本文差模 EMI 滤波器的 ANN 样本数据集大小为 60。
根据 2.1 节分析确定本文模型输入变量的个数为 3 个, 输出变量为插入损耗的 101 个特征点, 因此确定本文所提差模神经网络输入层和输出层的数量分别为 3 个和 101 个。隐含层可以采用多层,也可以采用单层。多隐含层由多个隐含层组成, 同单隐含层相比,多隐含层泛化能力强、预测精度高, 但训练时间较长。隐含层层数的选择需要综合考虑网络的精度和训练时间, 而本文所针对的映射问题较为简单, 因此采用仅有 3 层的浅层神经网络结构, 隐含层个数为 10, 其结构如图6所示。
为了避免梯度消失和梯度爆炸问题, 增强神经网络的表示能力与学习能力, 可以对神经元的激活函数进行设置, 常用的激活函数有 Sigmoid、Tansig、 Purelin、ReLU 等。本文所提出的隐含层和输出层采用的激活函数分别为 Tansig 和 Purelin, 其激活函数的表达式为
$\left\{\begin{array}{ll} y =\frac{{e}^{x}- {e}^{-x}}{{e}^{x}+ {e}^{-x}}& \text{ Tansig }\\ y = x &\text{ Pure lin }\end{array}\right.$
根据 2.3 节确定的神经网络, 本文通过使用 MATLAB 2021a 编程将其实现,编程所使用的电脑配置为 Intel(R) Core(TM) i7-4790 CPU、16GB RAM。 本文采用的浅层神经网络类型为反向传播 BP (back propagation) 网络, 是 1 个 3-10-101 的 3 层结构, 如图6所示, 采用 Levenberg-Marquardt 算法对其进行训练。在电脑运行内存较充足的条件下, 该算法可以更快地训练好神经网络, 设置最大的训练迭代次数为 1000,训练停止目标为$1 \times {10}^{-5}$。此外, 神经网络的学习率是其重要的超参数, 在梯度下降法中学习率的取值非常关键, 取值过大则不会收敛, 取值过小则收敛速度较慢。通过对不同的学习率进行训练发现, 学习率取值为 0.1 时能达到较好的训练效果。本文通过编程对所提神经网络模型进行训练,训练过程总共迭代了 12 次,耗时$3\mathrm{\;s}$,在第 6 次时达到了最佳性能,如图7所示,之后如果继续迭代将可能导致过拟合现象发生。为了避免过拟合现象的发生,本文采用了 1 种提前停止策略。
为了更好地反应模型对数据的拟合度, 本文使用决定系数${R}^{2}$ 作为评价指标,其表达式为
${R}^{2}= 1 -\frac{\mathop{\sum }\limits_{{i = 1}}^{n}{\left({y}_{i}- {\widehat{y}}_{i}\right)}^{2}}{\mathop{\sum }\limits_{{i = 1}}^{n}{\left({y}_{i}- {\bar{y}}_{i}\right)}^{2}}$
式中:${y}_{i}$ 为真实值;${\widehat{y}}_{i}$ 为预测值;${\bar{y}}_{i}$ 为真实值的平均值。
${R}^{2}$ 的取值范围为$\left\lbrack {0,1}\right\rbrack$,其值越接近于 1 则拟合效果越好。将训练后的数据反归一化为真实数值后,计算得出 ANN 训练的${R}^{2}= {0.986578}$,未归一化前的${R}^{2}= {0.918988}$图8为神经网络预测模型拟合效果最差的结果,分析可知:本文所提神经网络模型预测出的结果与真实的测量结果之间的拟合程度良好, 并且在高频段时其拟合效果更好, 很好地克服了传统插入损耗预测方法在高频段预测精度差的缺陷。整体训练数据的拟合效果如图9所示,${R}^{2}$ 可达 0.94 以上,说明该 ANN 训练数据的拟合效果较好。
本文通过使用 5 组未参与训练的数据来验证所训练神经网络的有效性, 验证集误差采用均方根误差 RMSE(root mean square error)进行评价, 拟合曲线如图10所示。可见:数据进行反归一化前的最大 RMSE=0.6123,最小 RMSE=0.4862,均小于 1 ;数据进行反归一化后的最大 RMSE=11.5961, 最小 RMSE=8.4762。通过分析归一化前的数据可知,本文所提 ANN 模型的误差均较小; 归一化后的数据虽然从数据层面上看较大,但是真实的非线性拟合效果较好。
验证集的均方误差 MSE(mean square error)如图11所示,可见系数$R$ 可达 0.97,拟合效果较好; ANN 模型训练集与验证集的总 MSE 如图12所示, 可见其 MSE=0.97012,说明整体的训练效果明显。
本文所提使用人工神经网络预测差模 EMI 滤波器插入损耗的方法克服了传统方法依赖插入损耗推导和参数$S$ 描述的局限性,且传统方法在高频下往往具有较大的预测误差。基于 ANN 的方法可以准确预测${100}\mathrm{{kHz}}\sim {30}\mathrm{{MHz}}$ 频率范围内的插入损耗, 并且通过使用 ANN 模型能够实现零成本的扩数据库;通过将数据库与抑制 EMI 噪声所需要的衰减需求进行匹配可以快速找到合适的 EMI 滤波器参数, 为 EMI 滤波器的参数设计与优化提供有价值的指导。
值得注意的是, 本文中的 ANN 建模专门针对具有 CLC 拓扑的差模 EMI 滤波器, 并且建立的模型仅适用于此类特定的无源器件。
  • 深圳市科技计划资助项目(JCYJ20230808104910021)
参考文献 引证文献
排序方式:
[1]
Jiang Shiqi, Liu Yitao, Ye Xin, et al.. Design of a fully integrated EMI filter for a single-phase grid-connected inverter[J]. IEEE Transactions on Industrial Electronics, 2021. 68(12): 12296-12309.
[2]
曹才开. 开关电源电磁干扰滤波器插入损耗的研究[J]. 通信电源技术, 2003. 5): 17-27.
Cao Caikai. Research of insert loss for electromagnetic interference filters of switching mode power supply[J]. Telecom Power Technology, 2003. 5): 17-27 (in Chinese).
[3]
陆德政. 基于神经网络的电磁干扰滤波器建模及设计研究[D]. 杭州: 浙江大学, 2020.
Lu Dezheng. Research on modeling and design of electromagnetic interference filter based on neural network[D]. 杭州: 浙江大学, 2020. (in Chinese).
[4]
陈健. 基于器件有限元建模的EMI滤波器高频特性研究[D]. 南京: 南京师范大学, 2021.
[5]
Negri S, Spadacini G, Grassi F, et al.. Prediction of EMI filter attenuation in power-electronic converters via circuit simulation[J]. IEEE Transactions on Electromagnetic Compatibility, 2022. 64(4): 1086-1096.
[6]
Zhao Shuai, Blaabjerg F, Wang Huai. An overview of artificial intelligence applications for power electronics[J]. IEEE Transactions on Power Electronics, 2021. 36(4): 4633-4658.
[7]
Dragičević T, Wheeler P, Blaabjerg F. Artificial intelligence aided automated design for reliability of power electronic systems[J]. IEEE Transactions on Power Electronics, 2019. 34(8): 7161-7171.
[8]
Li Xinze, Zhang Xin, Lin Fanfan, et al.. Artificial-intelligence-based design for circuit parameters of power converters[J]. IEEE Transactions on Industrial Electronics, 2022. 69(11): 11144-11155.
[9]
Performance enhancement of a series active power filter using Kalman filter based neural network control strategy[C]// 2018 International Conference on Advances in Computing, Communications and Informatics (ICACCI). Bangalore, India, 2018: 1702-1706.
[10]
Artificial neural network based three phase shunt active power filter[C]// 2016 IEEE 7th Power India International Conference (PI-ICON). Bikaner, India, 2016: 1-6.
[11]
Guillod T, Papamanolis P, W Kolar J. Artificial neural network (ANN) based fast and accurate inductor modeling and design[J]. IEEE Open Journal of Power Electronics, 2020. 1: 284-299.
[12]
Tang Yuanhong, Hu Weihao, Cao Di, et al.. Artificial intelligence-aided minimum reactive power control for the DAB converter based on harmonic analysis method[J]. IEEE Transactions on Power Electronics, 2021. 36(9): 9704-9710.
2024年第22卷第5期
PDF下载
353
141
引用本文
BibTeX
文章信息
doi: 10.13234/j.issn.2095-2805.2024.5.67
  • 接收时间:2024-04-08
  • 首发时间:2025-07-20
  • 出版时间:2024-09-30
补充材料
相关文章
文章信息
作者
出版历史
  • 收稿日期:2024-04-08
  • 修回日期:2024-07-31
  • 录用日期:2024-08-11
基金
Shenzhen Science and Technology Program(JCYJ20230808104910021)
深圳市科技计划资助项目(JCYJ20230808104910021)
作者信息
    深圳大学 机电与控制工程学院 深圳 518060
参考文献
分享链接
https://castjournals.cast.org.cn/joweb/dyxb/CN/10.13234/j.issn.2095-2805.2024.5.67
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