Article(id=1200482409276895399, tenantId=1146029695717560320, journalId=1189987059142926344, issueId=1200482403828495344, articleNumber=null, orderNo=null, doi=10.19457/j.1001-2095.dqcd24775, pmid=null, cstr=null, oa=null, hot=null, price=null, onlineType=0, articleFormat=0, articleType=null, articleTypeStr=research-article, receivedDate=1668528000000, receivedDateStr=2022-11-16, revisedDate=1671811200000, revisedDateStr=2022-12-24, acceptedDate=null, acceptedDateStr=null, onlineDate=1764146910256, onlineDateStr=2025-11-26, pubDate=1718812800000, pubDateStr=2024-06-20, doiRegisterDate=null, doiRegisterDateStr=null, onlineIssueDate=1764146910256, onlineIssueDateStr=2025-11-26, onlineJustAcceptDate=null, onlineJustAcceptDateStr=null, onlineFirstDate=null, onlineFirstDateStr=null, sourceXml=null, magXml=null, createTime=1764146910256, creator=13701087609, updateTime=1764146910256, updator=13701087609, issue=Issue{id=1200482403828495344, tenantId=1146029695717560320, journalId=1189987059142926344, year='2024', volume='54', issue='6', pageStart='3', pageEnd='96', issueExtLink='null', onlineDate='null', pubDate='null', beforeIssueId=null, nextIssueId=null, price=null, status=1, issueComplete=1, articleOrder=1, issueType=-1, specialIssue=null, createTime=1764146908957, creator=13701087609, updateTime=1764224882080, updator=13701087609, preIssue=null, nextIssue=null, ext={EN=IssueExt(id=1200809446868898278, tenantId=1146029695717560320, journalId=1189987059142926344, issueId=1200482403828495344, language=EN, specialIssueTitle=, coverIllustrator=null, specialIssueEditor=, specialIssueAbout=), CN=IssueExt(id=1200809446868898279, tenantId=1146029695717560320, journalId=1189987059142926344, issueId=1200482403828495344, language=CN, specialIssueTitle=, coverIllustrator=null, specialIssueEditor=, specialIssueAbout=)}, issueFiles=null}, startPage=45, endPage=53, ext={EN=ArticleExt(id=1200482409625022644, articleId=1200482409276895399, tenantId=1146029695717560320, journalId=1189987059142926344, language=EN, title=New Energy Scenario Generation Based on SA/WGAN, columnId=null, journalTitle=Electric Drive, columnName=null, runingTitle=null, highlight=null, articleAbstract=

With the increased penetration rate of new energy year by year,it is difficult to accurately predict the randomness and fluctuation characteristics of its output,causing a severe challenge to the operation,planning and scheduling of electrical power system. Therefore,modeling for the uncertainty of new energy has attracted more and more attention. To obtain the time sequence characteristics of new energy output scenario more effectively,a new energy scenario generation method was proposed based on data drive,and combined self-attention mechanism with generative adversarial network discriminator with gradient penalty through applying the SA/WGAN model. Through building a deep learning model based on the combination of two models,effectively highlight the timing sequence characteristics of new energy output scenario and enhancing the nonlinear fitting capability in scenario generation. The example results show that,compared with the scenario generation results of original WGAN and WGAN-LSTM,the new energy generation scenario of proposed model can not only effectively improve the accuracy,but also possess the advantages of stable WGAN-GP training results and quick SA calculation speed,which can achieve a more efficient generation of scenarios that is close to the distribution of real new energy scenario.

, 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=Yuhao WANG, Haitao LIU, Kangkai ZHU, Cong ZHONG, Jiayi MA), CN=ArticleExt(id=1200482412548452684, articleId=1200482409276895399, tenantId=1146029695717560320, journalId=1189987059142926344, language=CN, title=基于SA/WGAN的新能源场景生成方法, columnId=1190325328129131131, journalTitle=电气传动, columnName=综合能源与现代电网, runingTitle=null, highlight=null, articleAbstract=

随着新能源渗透率逐年提高,其出力的随机性与波动特性难以准确预测,给电力系统的运行、规划和调度提出了严峻的挑战,因此新能源的不确定性建模受到越来越多的关注。为了更有效地获得新能源出力场景中的时序特征,提出了一种基于数据驱动的新能源场景生成方法,通过采用SA/WGAN 模型,把自注意力机制和带有梯度惩罚的生成对抗网络判别器结合,构建基于两种模型结合的深度学习模型,有效突显新能源出力场景中时序特性,增强场景生成中非线性拟合能力。算例结果表明,所提模型的新能源生成场景相较于原始WGAN和WGAN-LSTM的场景生成结果,可以有效提高精准度,同时具备了WGAN-GP训练结果稳定和SA计算速度快的优势,更高效地生成与真实新能源场景分布接近的场景。

, correspAuthors=null, authorNote=null, correspAuthorsNote=null, copyrightStatement=null, copyrightOwner=null, extLink=null, articleAbsUrl=null, sourceXml=CXCH0Z87MJYmdZcJ4LKMuw==, magXml=PFaq5WqGfqy++i/bmwTHWw==, pdfUrl=null, pdf=4UsYgTrgTArX7KQboMNVeQ==, pdfFileSize=5786583, pdfExtLink=null, richHtmlUrl=null, mobilePdfUrl=null, reviewReport=null, pdfFirstPage=null, abstractGraph=KjnKPQqZdNVYtldLtTK1JA==, abstractGraphContent=null, abstractVideo=null, citation=null, cebUrl=null, magXmlContent=bmfJcHQkKWuwvzJSt58AdA==, mapNumber=null, authorCompany=null, fund=null, authors=

王宇昊(1998—),男,硕士,主要研究方向为新能源优化及配置,Email:

, authorsList=王宇昊, 刘海涛, 朱康凯, 仲聪, 马佳伊)}, authors=[Author(id=1200501568186478986, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, orderNo=0, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=1521934803@qq.com, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1200501568282947983, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, authorId=1200501568186478986, language=EN, stringName=Yuhao WANG, firstName=Yuhao, middleName=null, lastName=WANG, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1 School of Electric Power Engineering,Nanjing Institute of Technology,Nanjing 210000,Jiangsu,China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1200501568350056850, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, authorId=1200501568186478986, language=CN, stringName=王宇昊, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1 南京工程学院 电力工程学院,江苏 南京 210000, bio={"content":"

王宇昊(1998—),男,硕士,主要研究方向为新能源优化及配置,Email:

"}, bioImg=null, bioContent=

王宇昊(1998—),男,硕士,主要研究方向为新能源优化及配置,Email:

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1200501567972569468, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, xref=1, ext=[AuthorCompanyExt(id=1200501567980958076, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, companyId=1200501567972569468, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 School of Electric Power Engineering,Nanjing Institute of Technology,Nanjing 210000,Jiangsu,China), AuthorCompanyExt(id=1200501567989346685, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, companyId=1200501567972569468, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 南京工程学院 电力工程学院,江苏 南京 210000)])]), Author(id=1200501568450720150, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, orderNo=1, 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=1200501568593326490, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, authorId=1200501568450720150, language=EN, stringName=Haitao LIU, firstName=Haitao, middleName=null, lastName=LIU, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1 School of Electric Power Engineering,Nanjing Institute of Technology,Nanjing 210000,Jiangsu,China
2 Jiangsu Collaborative Innovation Center of Smart Distribution Network,Nanjing 210000,Jiangsu,China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1200501568677212573, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, authorId=1200501568450720150, language=CN, stringName=刘海涛, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1 南京工程学院 电力工程学院,江苏 南京 210000
2 江苏省配电网智能技术与装备协同创新中心,江苏 南京 210000, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1200501567972569468, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, xref=1, ext=[AuthorCompanyExt(id=1200501567980958076, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, companyId=1200501567972569468, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 School of Electric Power Engineering,Nanjing Institute of Technology,Nanjing 210000,Jiangsu,China), AuthorCompanyExt(id=1200501567989346685, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, companyId=1200501567972569468, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 南京工程学院 电力工程学院,江苏 南京 210000)]), AuthorCompany(id=1200501568081621377, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, xref=2, ext=[AuthorCompanyExt(id=1200501568094204290, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, companyId=1200501568081621377, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 Jiangsu Collaborative Innovation Center of Smart Distribution Network,Nanjing 210000,Jiangsu,China), AuthorCompanyExt(id=1200501568098398595, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, companyId=1200501568081621377, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 江苏省配电网智能技术与装备协同创新中心,江苏 南京 210000)])]), Author(id=1200501568794653089, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, 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=1200501568937259431, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, authorId=1200501568794653089, language=EN, stringName=Kangkai ZHU, firstName=Kangkai, middleName=null, lastName=ZHU, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1 School of Electric Power Engineering,Nanjing Institute of Technology,Nanjing 210000,Jiangsu,China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1200501569046311342, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, authorId=1200501568794653089, language=CN, stringName=朱康凯, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1 南京工程学院 电力工程学院,江苏 南京 210000, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1200501567972569468, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, xref=1, ext=[AuthorCompanyExt(id=1200501567980958076, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, companyId=1200501567972569468, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 School of Electric Power Engineering,Nanjing Institute of Technology,Nanjing 210000,Jiangsu,China), AuthorCompanyExt(id=1200501567989346685, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, companyId=1200501567972569468, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 南京工程学院 电力工程学院,江苏 南京 210000)])]), Author(id=1200501569197306293, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, 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=1200501569327329724, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, authorId=1200501569197306293, language=EN, stringName=Cong ZHONG, firstName=Cong, middleName=null, lastName=ZHONG, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1 School of Electric Power Engineering,Nanjing Institute of Technology,Nanjing 210000,Jiangsu,China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1200501569457353152, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, authorId=1200501569197306293, language=CN, stringName=仲聪, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1 南京工程学院 电力工程学院,江苏 南京 210000, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1200501567972569468, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, xref=1, ext=[AuthorCompanyExt(id=1200501567980958076, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, companyId=1200501567972569468, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 School of Electric Power Engineering,Nanjing Institute of Technology,Nanjing 210000,Jiangsu,China), AuthorCompanyExt(id=1200501567989346685, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, companyId=1200501567972569468, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 南京工程学院 电力工程学院,江苏 南京 210000)])]), Author(id=1200501569537044931, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, 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=1200501569633513927, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, authorId=1200501569537044931, language=EN, stringName=Jiayi MA, firstName=Jiayi, middleName=null, lastName=MA, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1 School of Electric Power Engineering,Nanjing Institute of Technology,Nanjing 210000,Jiangsu,China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1200501569734177228, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, authorId=1200501569537044931, language=CN, stringName=马佳伊, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1 南京工程学院 电力工程学院,江苏 南京 210000, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1200501567972569468, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, xref=1, ext=[AuthorCompanyExt(id=1200501567980958076, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, companyId=1200501567972569468, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 School of Electric Power Engineering,Nanjing Institute of Technology,Nanjing 210000,Jiangsu,China), AuthorCompanyExt(id=1200501567989346685, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, companyId=1200501567972569468, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 南京工程学院 电力工程学院,江苏 南京 210000)])])], keywords=[Keyword(id=1200501569990029783, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, language=EN, orderNo=1, keyword=unsupervised learning), Keyword(id=1200501570078110172, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, language=EN, orderNo=2, keyword=self-attention (SA)), Keyword(id=1200501570199744992, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, language=EN, orderNo=3, keyword=generative adversarial networks(GAN)), Keyword(id=1200501570342351336, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, language=EN, orderNo=4, keyword=new energy), Keyword(id=1200501570472374763, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, language=EN, orderNo=5, keyword=scenario generation), Keyword(id=1200501570585620977, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, language=CN, orderNo=1, keyword=无监督学习), Keyword(id=1200501570862445050, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, language=CN, orderNo=2, keyword=自注意力机制), Keyword(id=1200501571013439999, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, language=CN, orderNo=3, keyword=生成对抗网络), Keyword(id=1200501571181212165, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, language=CN, orderNo=4, keyword=新能源), Keyword(id=1200501571315429895, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, language=CN, orderNo=5, keyword=场景生成)], refs=[Reference(id=1200501576638001795, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, doi=null, pmid=null, pmcid=null, year=2020, volume=68, issue=1, pageStart=396, pageEnd=400, url=null, language=null, rfNumber=[1], rfOrder=0, authorNames=CHEN X, ZHANG T, YE W, journalName=IEEE Transactions on Circuits and Systems II:Express Briefs, refType=null, unstructuredReference=CHEN X, ZHANG T, YE W, et al. Blockchain-based electric vehicle incentive system for renewable energy consumption[J]. IEEE Transactions on Circuits and Systems II:Express Briefs, 2020, 68(1):396-400., articleTitle=Blockchain-based electric vehicle incentive system for renewable energy consumption, refAbstract=null), Reference(id=1200501576734470790, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, doi=null, pmid=null, pmcid=null, year=2022, volume=42, issue=8, pageStart=79, pageEnd=88, url=null, language=null, rfNumber=[2], rfOrder=1, authorNames=徐艳春, 蒋伟俊, 汪平, journalName=电力自动化设备, refType=null, unstructuredReference=徐艳春, 蒋伟俊, 汪平, 等. 考虑暂态电压定稳定的含高渗透率风光的电网动态无功规划方法[J]. 电力自动化设备, 2022, 42(8):79-88., articleTitle=考虑暂态电压定稳定的含高渗透率风光的电网动态无功规划方法, refAbstract=null), Reference(id=1200501576839328394, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, doi=null, pmid=null, pmcid=null, year=2022, volume=42, issue=8, pageStart=79, pageEnd=88, url=null, language=null, rfNumber=[2], rfOrder=2, authorNames=XU Yanchun, JIANG Weijun, WANG Ping, journalName=Electric Power Automation Equipment, refType=null, unstructuredReference=XU Yanchun, JIANG Weijun, WANG Ping, et al. Dynamic reactive power planning method for power grid with high permeability wind power and photovoltaic considering transient voltage stability[J]. Electric Power Automation Equipment, 2022, 42 (8):79-88., articleTitle=Dynamic reactive power planning method for power grid with high permeability wind power and photovoltaic considering transient voltage stability, refAbstract=null), Reference(id=1200501576948380303, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, doi=null, pmid=null, pmcid=null, year=2021, volume=45, issue=5, pageStart=79, pageEnd=85, url=null, language=null, rfNumber=[3], rfOrder=3, authorNames=杨茂, 董昊, journalName=电力系统自动化, refType=null, unstructuredReference=杨茂, 董昊. 基于数值天气预报风速和蒙特卡洛法的短期风电功率区间预测[J]. 电力系统自动化, 2021, 45(5):79-85., articleTitle=基于数值天气预报风速和蒙特卡洛法的短期风电功率区间预测, refAbstract=null), Reference(id=1200501577057432211, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, doi=null, pmid=null, pmcid=null, year=2021, volume=45, issue=5, pageStart=79, pageEnd=85, url=null, language=null, rfNumber=[3], rfOrder=4, authorNames=YANG Mao, DONG Hao, journalName=Automation of Electric Power Systems, refType=null, unstructuredReference=YANG Mao, DONG Hao. Short-term wind power interval prediction based on wind speed of numerical weather prediction and Monte Carlo method[J]. Automation of Electric Power Systems, 2021, 45(5):79-85., articleTitle=Short-term wind power interval prediction based on wind speed of numerical weather prediction and Monte Carlo method, refAbstract=null), Reference(id=1200501577191649942, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, doi=null, pmid=null, pmcid=null, year=2022, volume=46, issue=9, pageStart=3304, pageEnd=3318, url=null, language=null, rfNumber=[4], rfOrder=5, authorNames=姚志力, 江斌开, 龚春阳, journalName=电网技术, refType=null, unstructuredReference=姚志力, 江斌开, 龚春阳, 等. 计及电动汽车充电负荷特性的区域多能源站规划设计[J]. 电网技术, 2022, 46(9):3304-3318., articleTitle=计及电动汽车充电负荷特性的区域多能源站规划设计, refAbstract=null), Reference(id=1200501577325867674, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, doi=null, pmid=null, pmcid=null, year=2022, volume=46, issue=9, pageStart=3304, pageEnd=3318, url=null, language=null, rfNumber=[4], rfOrder=6, authorNames=YAO Zhili, JIANG Binkai, GONG Chunyang, journalName=Power System Technology, refType=null, unstructuredReference=YAO Zhili, JIANG Binkai, GONG Chunyang, et al. Planning and design of regional multi-energy stations considering electric vehicle charging load characteristics[J]. Power System Technology, 2022, 46(9):3304-3318., articleTitle=Planning and design of regional multi-energy stations considering electric vehicle charging load characteristics, refAbstract=null), Reference(id=1200501577439113885, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, doi=null, pmid=null, pmcid=null, year=2022, volume=50, issue=9, pageStart=75, pageEnd=83, url=null, language=null, rfNumber=[5], rfOrder=7, authorNames=马燕峰, 骆泽榕, 赵书强, journalName=电力系统保护与控制, refType=null, unstructuredReference=马燕峰, 骆泽榕, 赵书强, 等. 基于改进蒙特卡洛混合抽样的含风光电力系统风险评估[J]. 电力系统保护与控制, 2022, 50(9):75-83., articleTitle=基于改进蒙特卡洛混合抽样的含风光电力系统风险评估, refAbstract=null), Reference(id=1200501577510417055, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, doi=null, pmid=null, pmcid=null, year=2022, volume=50, issue=9, pageStart=75, pageEnd=83, url=null, language=null, rfNumber=[5], rfOrder=8, authorNames=MA Yanfeng, LUO Zerong, ZHAO Shuqiang, journalName=Power System Protection and Control, refType=null, unstructuredReference=MA Yanfeng, LUO Zerong, ZHAO Shuqiang, et al. Risk assessment of a power system containing wind power and photovoltaic based on improved Monte Carlo mixed sampling[J]. Power System Protection and Control, 2022, 50(9):75-83., articleTitle=Risk assessment of a power system containing wind power and photovoltaic based on improved Monte Carlo mixed sampling, refAbstract=null), Reference(id=1200501577606886050, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, doi=null, pmid=null, pmcid=null, year=2018, volume=42, issue=3, pageStart=835, pageEnd=841, url=null, language=null, rfNumber=[6], rfOrder=9, authorNames=艾小猛, 塔伊尔江·巴合依, 杨立滨, journalName=电网技术, refType=null, unstructuredReference=艾小猛, 塔伊尔江·巴合依, 杨立滨, 等. 基于场景集的含风电电力系统旋转备用优化[J]. 电网技术, 2018, 42(3):835-841., articleTitle=基于场景集的含风电电力系统旋转备用优化, refAbstract=null), Reference(id=1200501577711743653, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, doi=null, pmid=null, pmcid=null, year=2018, volume=42, issue=3, pageStart=835, pageEnd=841, url=null, language=null, rfNumber=[6], rfOrder=10, authorNames=AI Xiaomeng, Tayierjiang B, YANG Libini, journalName=Power System Technology, refType=null, unstructuredReference=AI Xiaomeng, Tayierjiang B, YANG Libini, et al. Optimizing the spinning reserve in wind power system using scenario method[J]. Power System Technology, 2018, 42(3):835-841., articleTitle=Optimizing the spinning reserve in wind power system using scenario method, refAbstract=null), Reference(id=1200501577879515815, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, doi=null, pmid=null, pmcid=null, year=2018, volume=42, issue=5, pageStart=147, pageEnd=153, url=null, language=null, rfNumber=[7], rfOrder=11, authorNames=董雷, 孟天骄, 陈乃仕, journalName=电力系统自动化, refType=null, unstructuredReference=董雷, 孟天骄, 陈乃仕, 等. 采用马尔可夫链—多场景技术的交直流主动配电网优化调度[J]. 电力系统自动化, 2018, 42(5):147-153., articleTitle=采用马尔可夫链—多场景技术的交直流主动配电网优化调度, refAbstract=null), Reference(id=1200501577980179113, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, doi=null, pmid=null, pmcid=null, year=2018, volume=42, issue=5, pageStart=147, pageEnd=153, url=null, language=null, rfNumber=[7], rfOrder=12, authorNames=DONG Lei, MENG Tianjiao, CHEN Naishi, journalName=Automation of Electric Power Systems, refType=null, unstructuredReference=DONG Lei, MENG Tianjiao, CHEN Naishi, et al. Optimized scheduling of AC/DC hybrid active distribution network using Markov chains and multiple scenarios technique[J]. Automation of Electric Power Systems, 2018, 42(5):147-153., articleTitle=Optimized scheduling of AC/DC hybrid active distribution network using Markov chains and multiple scenarios technique, refAbstract=null), Reference(id=1200501578076648107, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, doi=null, pmid=null, pmcid=null, year=2022, volume=46, issue=2, pageStart=622, pageEnd=634, url=null, language=null, rfNumber=[8], rfOrder=13, authorNames=董锴, 蔡新雷, 崔艳林, journalName=电网技术, refType=null, unstructuredReference=董锴, 蔡新雷, 崔艳林, 等. 基于马尔科夫链的电动汽车聚合建模及多模式调频控制策略[J]. 电网技术, 2022, 46(2):622-634., articleTitle=基于马尔科夫链的电动汽车聚合建模及多模式调频控制策略, refAbstract=null), Reference(id=1200501578206671535, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, doi=null, pmid=null, pmcid=null, year=2022, volume=46, issue=2, pageStart=622, pageEnd=634, url=null, language=null, rfNumber=[8], rfOrder=14, authorNames=DONG Kai, CAI Xinlei, CUI Yanlin, journalName=Power System Technology, refType=null, unstructuredReference=DONG Kai, CAI Xinlei, CUI Yanlin, et al. Aggregation modeling based on Markov chain and multi-mode control strategies of aggregated electric vehicles for frequency regulation[J]. Power System Technology, 2022, 46(2):622-634., articleTitle=Aggregation modeling based on Markov chain and multi-mode control strategies of aggregated electric vehicles for frequency regulation, refAbstract=null), Reference(id=1200501579381076657, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, doi=null, pmid=null, pmcid=null, year=2018, volume=33, issue=3, pageStart=3265, pageEnd=3275, url=null, language=null, rfNumber=[9], rfOrder=15, authorNames=CHEN Y, WANG Y, KIRSCHEN D, journalName=IEEE Transactions on Power Systems, refType=null, unstructuredReference=CHEN Y, WANG Y, KIRSCHEN D, et al. Model-free renewable scenario generation using generative adversarial networks[J]. IEEE Transactions on Power Systems, 2018, 33(3):3265-3275., articleTitle=Model-free renewable scenario generation using generative adversarial networks, refAbstract=null), Reference(id=1200501579519488691, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, doi=null, pmid=null, pmcid=null, year=2017, volume=null, issue=null, pageStart=97, pageEnd=108, url=null, language=null, rfNumber=[10], rfOrder=16, authorNames=YU Yang, GONG Zhiqiang, ZHONG Ping, journalName=International Conference on Ima-ge and Graphics, refType=null, unstructuredReference=YU Yang, GONG Zhiqiang, ZHONG Ping, et al. Unsupervised representation learning with deep convolutional neural network for remote sensing images[C]// International Conference on Ima-ge and Graphics, 2017:97-108., articleTitle=Unsupervised representation learning with deep convolutional neural network for remote sensing images, refAbstract=null), Reference(id=1200501579615957686, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, doi=null, pmid=null, pmcid=null, year=2021, volume=45, issue=5, pageStart=1742, pageEnd=1751, url=null, language=null, rfNumber=[11], rfOrder=17, authorNames=顾洁, 刘书琪, 胡玉, journalName=电网技术, refType=null, unstructuredReference=顾洁, 刘书琪, 胡玉, 等. 基于深度卷积生成对抗网络场景生成的间歇式分布式电源优化配置[J]. 电网技术, 2021, 45(5):1742-1751., articleTitle=基于深度卷积生成对抗网络场景生成的间歇式分布式电源优化配置, refAbstract=null), Reference(id=1200501579695649464, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, doi=null, pmid=null, pmcid=null, year=2021, volume=45, issue=5, pageStart=1742, pageEnd=1751, url=null, language=null, rfNumber=[11], rfOrder=18, authorNames=GU Jie, LIU Shuqi, HU Yu, journalName=Power System Technology, refType=null, unstructuredReference=GU Jie, LIU Shuqi, HU Yu, et al. Optimal allocation of intermittent distributed generation based on deep convolutions generative adversarial network in scenario generation[J]. Power System Technology, 2021, 45(5):1742-1751., articleTitle=Optimal allocation of intermittent distributed generation based on deep convolutions generative adversarial network in scenario generation, refAbstract=null), Reference(id=1200501579800507068, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, doi=null, pmid=null, pmcid=null, year=2017, volume=null, issue=null, pageStart=5769, pageEnd=5779, url=null, language=null, rfNumber=[12], rfOrder=19, authorNames=GULRAJANI I, AHMED F, ARJOVSKY M, journalName=USA:NIPS, refType=null, unstructuredReference=GULRAJANI I, AHMED F, ARJOVSKY M, et al. Improved training of Wasserstein GANs[C]//In Proceedings of the 31st International Conference on Neural Information Processing Systems,Long Beach,CA, USA:NIPS, 2017:5769-5779., articleTitle=Improved training of Wasserstein GANs, refAbstract=null), Reference(id=1200501579896976062, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, doi=null, pmid=null, pmcid=null, year=2014, volume=null, issue=null, pageStart=2672, pageEnd=2680, url=null, language=null, rfNumber=[13], rfOrder=20, authorNames=GOODFELLOW I, POUGET-ABADIE J, MIRZA M, journalName=Proceedings of the 27th International Conference on Neural Information Processing Systems. Montreal, refType=null, unstructuredReference=GOODFELLOW I, POUGET-ABADIE J, MIRZA M, et al. Ge-nerative adversarial nets[C]// Proceedings of the 27th International Conference on Neural Information Processing Systems. Montreal, Canada: MIT Press, 2014:2672-2680., articleTitle=Ge-nerative adversarial nets, refAbstract=null), Reference(id=1200501579968279232, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, doi=null, pmid=null, pmcid=null, year=2023, volume=43, issue=3, pageStart=987, pageEnd=1004, url=null, language=null, rfNumber=[14], rfOrder=21, authorNames=邵振国, 张承圣, 陈飞雄, journalName=中国电机工程学报, refType=null, unstructuredReference=邵振国, 张承圣, 陈飞雄, 等. 生成对抗网络及其在电力系统中的应用综述[J]. 中国电机工程学报, 2023, 43(3):987-1004., articleTitle=生成对抗网络及其在电力系统中的应用综述, refAbstract=null), Reference(id=1200501580047971010, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, doi=null, pmid=null, pmcid=null, year=2023, volume=43, issue=3, pageStart=987, pageEnd=1004, url=null, language=null, rfNumber=[14], rfOrder=22, authorNames=SHAO Zhenguo, ZHANG Chengsheng, CHEN Feixiong, journalName=Proceedings of the CSEE, refType=null, unstructuredReference=SHAO Zhenguo, ZHANG Chengsheng, CHEN Feixiong, et al. A review on generative adversarial networks for power system applications[J]. Proceedings of the CSEE, 2023, 43(3):987-1004., articleTitle=A review on generative adversarial networks for power system applications, refAbstract=null), Reference(id=1200501580215743172, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, doi=null, pmid=null, pmcid=null, year=2022, volume=42, issue=8, pageStart=2889, pageEnd=2900, url=null, language=null, rfNumber=[15], rfOrder=23, authorNames=兰健, 郭庆来, 周艳真, journalName=中国电机工程学报, refType=null, unstructuredReference=兰健, 郭庆来, 周艳真, 等. 基于生成对抗网络和模型迁移的电力系统典型运行方式样本生成[J]. 中国电机工程学报, 2022, 42(8):2889-2900., articleTitle=基于生成对抗网络和模型迁移的电力系统典型运行方式样本生成, refAbstract=null), Reference(id=1200501580299629254, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, doi=null, pmid=null, pmcid=null, year=2022, volume=42, issue=8, pageStart=2889, pageEnd=2900, url=null, language=null, rfNumber=[15], rfOrder=24, authorNames=LAN Jian, GUO Qinglai, ZHOU Yanzhen, journalName=Proceedings of the CSEE, refType=null, unstructuredReference=LAN Jian, GUO Qinglai, ZHOU Yanzhen, et al. Generation of power system typical operation mode samples:a generation adversarial network and model-based transfer learning approach[J]. Proceedings of the CSEE, 2022, 42(8):2889-2900., articleTitle=Generation of power system typical operation mode samples:a generation adversarial network and model-based transfer learning approach, refAbstract=null), Reference(id=1200501580379321033, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=5999, pageEnd=6009, url=null, language=null, rfNumber=[16], rfOrder=25, authorNames=VASWANI A, SHAZEER N, PARMAR N, journalName=31st International Conference on Neural Information Processing Systems,Long Beach,USA, refType=null, unstructuredReference=VASWANI A, SHAZEER N, PARMAR N, et al. Attention is all you need[C]// 31st International Conference on Neural Information Processing Systems,Long Beach,USA,2017:5999-6009., articleTitle=Attention is all you need, refAbstract=null), Reference(id=1200501580459012809, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, doi=null, pmid=null, pmcid=null, year=2021, volume=42, issue=12, pageStart=1, pageEnd=8, url=null, language=null, rfNumber=[17], rfOrder=26, authorNames=朱庆, 郑红娟, 唐子逸, journalName=电力建设, refType=null, unstructuredReference=朱庆, 郑红娟, 唐子逸, 等. 基于生成对抗网络的综合能源负荷场景生成方法[J]. 电力建设, 2021, 42(12):1-8., articleTitle=基于生成对抗网络的综合能源负荷场景生成方法, refAbstract=null), Reference(id=1200501580744225482, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, doi=null, pmid=null, pmcid=null, year=2021, volume=42, issue=12, pageStart=1, pageEnd=8, url=null, language=null, rfNumber=[17], rfOrder=27, authorNames=ZHU Qing, ZHENG Hongjuan, TANG Ziyi, journalName=Electric Power Construction, refType=null, unstructuredReference=ZHU Qing, ZHENG Hongjuan, TANG Ziyi, et al. Load scenario generation of integrated energy system using generative adversarial networks[J]. Electric Power Construction, 2021, 42(12):1-8., articleTitle=Load scenario generation of integrated energy system using generative adversarial networks, refAbstract=null)], funds=[Fund(id=1200501576252125817, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, awardId=51777197, language=CN, fundingSource=国家自然科学基金(51777197), fundOrder=null, country=null), Fund(id=1200501576356983420, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, awardId=22KJA470005, language=CN, fundingSource=江苏省高校自然科学研究重大项目(22KJA470005), fundOrder=null, country=null)], companyList=[AuthorCompany(id=1200501567972569468, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, xref=1, ext=[AuthorCompanyExt(id=1200501567980958076, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, companyId=1200501567972569468, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 School of Electric Power Engineering,Nanjing Institute of Technology,Nanjing 210000,Jiangsu,China), AuthorCompanyExt(id=1200501567989346685, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, companyId=1200501567972569468, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 南京工程学院 电力工程学院,江苏 南京 210000)]), AuthorCompany(id=1200501568081621377, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, xref=2, ext=[AuthorCompanyExt(id=1200501568094204290, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, companyId=1200501568081621377, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 Jiangsu Collaborative Innovation Center of Smart Distribution Network,Nanjing 210000,Jiangsu,China), AuthorCompanyExt(id=1200501568098398595, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, companyId=1200501568081621377, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 江苏省配电网智能技术与装备协同创新中心,江苏 南京 210000)])], figs=[ArticleFig(id=1200501571596448275, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, language=EN, label=Fig.1, caption=Basic structure diagram of GAN model, figureFileSmall=N1WMPeNDqgW/kMyM8E9UvA==, figureFileBig=oIF3bhQmTag/LxWEH7YejQ==, tableContent=null), ArticleFig(id=1200501571709694486, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, language=CN, label=图1, caption=GAN模型基本结构图, figureFileSmall=N1WMPeNDqgW/kMyM8E9UvA==, figureFileBig=oIF3bhQmTag/LxWEH7YejQ==, tableContent=null), ArticleFig(id=1200501571860689437, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, language=EN, label=Fig.2, caption=Schematic diagram of self-attention calculation process, figureFileSmall=mJFsxkGBHnf1QILC6jYQqg==, figureFileBig=asVXvQSjWQ3/iLJ+A4MJ+A==, tableContent=null), ArticleFig(id=1200501571973935650, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, language=CN, label=图2, caption=自注意力机制计算过程示意图, figureFileSmall=mJFsxkGBHnf1QILC6jYQqg==, figureFileBig=asVXvQSjWQ3/iLJ+A4MJ+A==, tableContent=null), ArticleFig(id=1200501572066210342, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, language=EN, label=Fig.3, caption=Schematic diagram of self-attention model capturing global pairwise interaction relationships, figureFileSmall=+26wdxHxDzgA2sfM1jjUiw==, figureFileBig=1ucdbwZoU0Wn8y02PaaEkA==, tableContent=null), ArticleFig(id=1200501572141707816, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, language=CN, label=图3, caption=自注意力模型捕捉全局成对交互关系的示意图, figureFileSmall=+26wdxHxDzgA2sfM1jjUiw==, figureFileBig=1ucdbwZoU0Wn8y02PaaEkA==, tableContent=null), ArticleFig(id=1200501572393366062, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, language=EN, label=Fig.4, caption=SA/WGAN network structure diagram, figureFileSmall=qM31diWE/+WVOi5JxESt1A==, figureFileBig=eMn9ITNxRHqFZVrLWB7J/w==, tableContent=null), ArticleFig(id=1200501572565332531, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, language=CN, label=图4, caption=SA/WGAN网络结构示意图, figureFileSmall=qM31diWE/+WVOi5JxESt1A==, figureFileBig=eMn9ITNxRHqFZVrLWB7J/w==, tableContent=null), ArticleFig(id=1200501572728910392, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, language=EN, label=Fig.5, caption=Overall flow chart, figureFileSmall=gj2Ng5TapdVsx+Zb7iVPIA==, figureFileBig=GW8IhHhRPAf98yxVKHIfbQ==, tableContent=null), ArticleFig(id=1200501572812796477, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, language=CN, label=图5, caption=整体流程图, figureFileSmall=gj2Ng5TapdVsx+Zb7iVPIA==, figureFileBig=GW8IhHhRPAf98yxVKHIfbQ==, tableContent=null), ArticleFig(id=1200501572934431296, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, language=EN, label=Fig.6, caption=Pre-processed data, figureFileSmall=7Qran41P+7DbSoUOzd79CA==, figureFileBig=eDoFQ026xA4XYXQl9F4DfQ==, tableContent=null), ArticleFig(id=1200501573085426243, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, language=CN, label=图6, caption=预处理后的数据, figureFileSmall=7Qran41P+7DbSoUOzd79CA==, figureFileBig=eDoFQ026xA4XYXQl9F4DfQ==, tableContent=null), ArticleFig(id=1200501573202866760, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, language=EN, label=Fig.7, caption=Renderings with 0 iterations, figureFileSmall=blO3CXZxYHEQiDZ1lCI6Hg==, figureFileBig=ntiEXfP3ica0R5kZYVJ3Gg==, tableContent=null), ArticleFig(id=1200501573311918664, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, language=CN, label=图7, caption=迭代0次效果图, figureFileSmall=blO3CXZxYHEQiDZ1lCI6Hg==, figureFileBig=ntiEXfP3ica0R5kZYVJ3Gg==, tableContent=null), ArticleFig(id=1200501573387416139, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, language=EN, label=Fig.8, caption=Renderings with 1 200 iterations, figureFileSmall=dEFQhWGlamtg2g/rSraE4g==, figureFileBig=cNHSzaXAUZuqnHH37+MnLg==, tableContent=null), ArticleFig(id=1200501573500662349, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, language=CN, label=图8, caption=迭代1 200次效果图, figureFileSmall=dEFQhWGlamtg2g/rSraE4g==, figureFileBig=cNHSzaXAUZuqnHH37+MnLg==, tableContent=null), ArticleFig(id=1200501573584548432, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, language=EN, label=Fig.9, caption=Renderings with 4 000 iterations, figureFileSmall=I24c7PBcCeWVG2LkfFalgw==, figureFileBig=4vTFni8bp7e4eCZ9vTscYA==, tableContent=null), ArticleFig(id=1200501573706183250, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, language=CN, label=图9, caption=迭代4 000次效果图, figureFileSmall=I24c7PBcCeWVG2LkfFalgw==, figureFileBig=4vTFni8bp7e4eCZ9vTscYA==, tableContent=null), ArticleFig(id=1200501573798457941, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, language=EN, label=Fig.10, caption=Comparison chart of generated data and real data, figureFileSmall=IvYWeeiYpoBDTjDrpKui2Q==, figureFileBig=iuzkO4qgMPcW9B+SnmV4OQ==, tableContent=null), ArticleFig(id=1200501574993834584, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, language=CN, label=图10, caption=生成数据与真实数据比较图, figureFileSmall=IvYWeeiYpoBDTjDrpKui2Q==, figureFileBig=iuzkO4qgMPcW9B+SnmV4OQ==, tableContent=null), ArticleFig(id=1200501575098692187, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, language=EN, label=Fig.11, caption=Comparison chart of autocorrelation coefficient between generated data and real data, figureFileSmall=6BiFtna7rKhlp2TEeOKBRg==, figureFileBig=Fh17pB9tneu80lEurbBZ2A==, tableContent=null), ArticleFig(id=1200501575195161185, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, language=CN, label=图11, caption=生成数据与真实数据自相关系数对比图, figureFileSmall=6BiFtna7rKhlp2TEeOKBRg==, figureFileBig=Fh17pB9tneu80lEurbBZ2A==, tableContent=null), ArticleFig(id=1200501575337767522, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, language=EN, label=Fig.12, caption=Comparison chart of CDF between generated data and real data, figureFileSmall=X6OzYxTib3ludr6rS5+dSg==, figureFileBig=PVIie5vpWkpmz3fggr8Qpw==, tableContent=null), ArticleFig(id=1200501575434236519, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, language=CN, label=图12, caption=生成数据与真实数据累积概率密度函数对比图, figureFileSmall=X6OzYxTib3ludr6rS5+dSg==, figureFileBig=PVIie5vpWkpmz3fggr8Qpw==, tableContent=null), ArticleFig(id=1200501575509733993, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, language=EN, label=Fig.13, caption=Comparison of CDF of different scene generation methods, figureFileSmall=AFo9vY9GvqOnhBEvCqTh5A==, figureFileBig=iW7uWimGk6gn9uKSBmRc/w==, tableContent=null), ArticleFig(id=1200501575618785900, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, language=CN, label=图13, caption=不同场景生成方法累积概率密度函数对比图, figureFileSmall=AFo9vY9GvqOnhBEvCqTh5A==, figureFileBig=iW7uWimGk6gn9uKSBmRc/w==, tableContent=null), ArticleFig(id=1200501575723643503, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, language=EN, label=Tab.1, caption=

MSE comparison of different methods

, figureFileSmall=null, figureFileBig=null, tableContent=
方法 MSE
风电场景 光伏场景
WGAN-GP 0.004 95 0.003 49
WGAN-LSTM 0.002 68 0.003 24
SA/WGAN 0.002 33 0.001 82
), ArticleFig(id=1200501575920775792, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1200482409276895399, language=CN, label=表1, caption=

不同方法MSE对比

, figureFileSmall=null, figureFileBig=null, tableContent=
方法 MSE
风电场景 光伏场景
WGAN-GP 0.004 95 0.003 49
WGAN-LSTM 0.002 68 0.003 24
SA/WGAN 0.002 33 0.001 82
)], attaches=null, journal=Journal(id=1149695973913305096, delFlag=0, nameCn=电气传动, nameEn=Electric Drive, nameHistory1=null, nameHistory2=null, issn=1001-2095, eissn=null, cn=12-1067/TP, coden=null, periodic=0, language=CN, oaType=null, 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=MX9kQyxauhRIcWSHz1lsfg==, journalPrice=null, startedYear=null, abbrevIsoEn=null, journalRemark=null, publicationField=null, createdTime=1752038480501, updatedTime=1761735785885, createdBy=18614031015, updatedBy=13701087609, firstLetterCn=E, firstLetterEn=E, subjectCode=Engineering, subjectName=Engineering, subjectCodeEn=Engineering, subjectNameEn=null, picCn=MX9kQyxauhRIcWSHz1lsfg==, picEn=R8RQJTk995s1dk9HSLANAQ==, jcr=null, cjcr=null, exts=[JournalExt(id=1190369420775096487, language=CN, name=电气传动, nameHistory1=null, nameHistory2=null, managedBy=, sponsoredBy=, publishedBy=, editorOffice=, officeProv=null, officeCity=null, officeAddr=, officeZip=, editDirector=, officeDirector=null, officePhone=null, coverPicUrl=null, journalRemark=, submitArticleUrl=null, websiteUrl=, createdTime=1761735785907, updatedTime=1761735785907, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=, submissionAuthorUrl=https://tg.au365.cn/Login/index.html, submissionEditorUrl=https://tg.au365.cn/Login/index/cate/3.html, submissionReviewUrl=https://tg.au365.cn/Login/index/cate/2.html, submissionCeEditorUrl=, submissionAeEditorUrl=, option={"copyright":""}), JournalExt(id=1190369420829622440, language=EN, name=Electric Drive, nameHistory1=null, nameHistory2=null, managedBy=, sponsoredBy=, publishedBy=, editorOffice=, officeProv=null, officeCity=null, officeAddr=, officeZip=, editDirector=, officeDirector=null, officePhone=null, coverPicUrl=null, journalRemark=, submitArticleUrl=null, websiteUrl=, createdTime=1761735785920, updatedTime=1761735785920, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=, submissionAuthorUrl=https://tg.au365.cn/Login/index.html, submissionEditorUrl=https://tg.au365.cn/Login/index/cate/3.html, submissionReviewUrl=https://tg.au365.cn/Login/index/cate/2.html, submissionCeEditorUrl=, submissionAeEditorUrl=, option={"copyright":""})], databaseList=null, tenantJournalId=1189987059142926344, websiteList=[Website(id=1189988338468713189, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1189987059142926344, 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/dqcd/CN, language=CN, createTime=1761644928805, createBy=18614031015, updateTime=1761644944247, updateBy=18614031015, name=电气传动-中文, tplId=1146099689490845704, title=电气传动, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1190236467705254391, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189988338468713189, code=articleTextType, value=kx, createTime=1761704087425, updateTime=1761704087425, creator=18614031015, updator=18614031015), WebsiteProps(id=1190236467671699956, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189988338468713189, code=banner, value=null, createTime=1761704087417, updateTime=1761704087417, creator=18614031015, updator=18614031015), WebsiteProps(id=1190236467763974650, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189988338468713189, code=grayFlag, value=0, createTime=1761704087439, updateTime=1761704087439, creator=18614031015, updator=18614031015), WebsiteProps(id=1190236467663311347, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189988338468713189, code=logo, value=https://castjournals.cast.org.cn/joweb/dqcd/CN/file/pic?fileId=C+51VGe7It8gHMdS0MAlXg==, createTime=1761704087415, updateTime=1761704087415, creator=18614031015, updator=18614031015), WebsiteProps(id=1190236467784946172, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189988338468713189, code=minRunFlag, value=0, createTime=1761704087444, updateTime=1761704087444, creator=18614031015, updator=18614031015), WebsiteProps(id=1190236467692671478, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189988338468713189, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/dqcd/CN/file/pic, createTime=1761704087422, updateTime=1761704087422, creator=18614031015, updator=18614031015), WebsiteProps(id=1190236467776557563, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189988338468713189, code=silenceFlag, value=0, createTime=1761704087442, updateTime=1761704087442, creator=18614031015, updator=18614031015), WebsiteProps(id=1190236467680088565, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189988338468713189, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_cn_619/, createTime=1761704087419, updateTime=1761704087419, creator=18614031015, updator=18614031015), WebsiteProps(id=1190236467713643000, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189988338468713189, code=themeColor, value=null, createTime=1761704087427, updateTime=1761704087427, creator=18614031015, updator=18614031015), WebsiteProps(id=1190236467726225913, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189988338468713189, code=themeStyle, value=null, createTime=1761704087430, updateTime=1761704087430, creator=18614031015, updator=18614031015)]), Website(id=1189988338527433447, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1189987059142926344, 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/dqcd/EN, language=EN, createTime=1761644928819, createBy=18614031015, updateTime=1761644963810, updateBy=18614031015, name=电气传动-英文, tplId=1146101810881728533, title=Electric Drive, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1190236496637563393, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189988338527433447, code=articleTextType, value=kx, createTime=1761704094323, updateTime=1761704094323, creator=18614031015, updator=18614031015), WebsiteProps(id=1190236496612397566, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189988338527433447, code=banner, value=null, createTime=1761704094317, updateTime=1761704094317, creator=18614031015, updator=18614031015), WebsiteProps(id=1190236496666923524, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189988338527433447, code=grayFlag, value=0, createTime=1761704094330, updateTime=1761704094330, creator=18614031015, updator=18614031015), WebsiteProps(id=1190236496604008957, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189988338527433447, code=logo, value=https://castjournals.cast.org.cn/joweb/dqcd/EN/file/pic?fileId=C+51VGe7It8gHMdS0MAlXg==, createTime=1761704094315, updateTime=1761704094315, creator=18614031015, updator=18614031015), WebsiteProps(id=1190236496683700742, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189988338527433447, code=minRunFlag, value=0, createTime=1761704094334, updateTime=1761704094334, creator=18614031015, updator=18614031015), WebsiteProps(id=1190236496629174784, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189988338527433447, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/dqcd/EN/file/pic, createTime=1761704094321, updateTime=1761704094321, creator=18614031015, updator=18614031015), WebsiteProps(id=1190236496675312133, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189988338527433447, code=silenceFlag, value=0, createTime=1761704094332, updateTime=1761704094332, creator=18614031015, updator=18614031015), WebsiteProps(id=1190236496620786175, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189988338527433447, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_en_623/, createTime=1761704094319, updateTime=1761704094319, creator=18614031015, updator=18614031015), WebsiteProps(id=1190236496650146306, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189988338527433447, code=themeColor, value=null, createTime=1761704094326, updateTime=1761704094326, creator=18614031015, updator=18614031015), WebsiteProps(id=1190236496658534915, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189988338527433447, code=themeStyle, value=null, createTime=1761704094328, updateTime=1761704094328, creator=18614031015, updator=18614031015)])], journalTitle=电气传动, weixinUrl=null, journalUrl=https://www.au365.cn/, iacademicId=null, status=1, seqNo=null, journalTitleEn=Electric Drive, journalPhotoCn=MX9kQyxauhRIcWSHz1lsfg==, journalPhotoEn=R8RQJTk995s1dk9HSLANAQ==, journalFirstLetter=E, 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=, provinceCode=null, provinceName=null, collectFlag=false), detailUrlCn=https://castjournals.cast.org.cn/joweb/dqcd/CN/10.19457/j.1001-2095.dqcd24775, detailUrlEn=https://castjournals.cast.org.cn/joweb/dqcd/EN/10.19457/j.1001-2095.dqcd24775, pdfUrlCn=https://castjournals.cast.org.cn/joweb/dqcd/CN/PDF/10.19457/j.1001-2095.dqcd24775, pdfUrlEn=https://castjournals.cast.org.cn/joweb/dqcd/EN/PDF/10.19457/j.1001-2095.dqcd24775, aliStartDate=null, aliEndDate=null, collectionFlag=false, citedCount=null, citedUrl=null, reference=null)
收藏切换
基于SA/WGAN的新能源场景生成方法
收藏切换
PDF下载
王宇昊 1 , 刘海涛 1, 2 , 朱康凯 1 , 仲聪 1 , 马佳伊 1
电气传动 | 综合能源与现代电网 2024,54(6): 45-53
收起
收藏切换
电气传动 | 综合能源与现代电网 2024, 54(6): 45-53
基于SA/WGAN的新能源场景生成方法
全屏
王宇昊1 , 刘海涛1, 2, 朱康凯1, 仲聪1, 马佳伊1
作者信息
  • 1 南京工程学院 电力工程学院,江苏 南京 210000
  • 2 江苏省配电网智能技术与装备协同创新中心,江苏 南京 210000
  • 王宇昊(1998—),男,硕士,主要研究方向为新能源优化及配置,Email:

New Energy Scenario Generation Based on SA/WGAN
Yuhao WANG1 , Haitao LIU1, 2, Kangkai ZHU1, Cong ZHONG1, Jiayi MA1
Affiliations
  • 1 School of Electric Power Engineering,Nanjing Institute of Technology,Nanjing 210000,Jiangsu,China
  • 2 Jiangsu Collaborative Innovation Center of Smart Distribution Network,Nanjing 210000,Jiangsu,China
出版时间: 2024-06-20 doi: 10.19457/j.1001-2095.dqcd24775
文章导航
收藏切换

随着新能源渗透率逐年提高,其出力的随机性与波动特性难以准确预测,给电力系统的运行、规划和调度提出了严峻的挑战,因此新能源的不确定性建模受到越来越多的关注。为了更有效地获得新能源出力场景中的时序特征,提出了一种基于数据驱动的新能源场景生成方法,通过采用SA/WGAN 模型,把自注意力机制和带有梯度惩罚的生成对抗网络判别器结合,构建基于两种模型结合的深度学习模型,有效突显新能源出力场景中时序特性,增强场景生成中非线性拟合能力。算例结果表明,所提模型的新能源生成场景相较于原始WGAN和WGAN-LSTM的场景生成结果,可以有效提高精准度,同时具备了WGAN-GP训练结果稳定和SA计算速度快的优势,更高效地生成与真实新能源场景分布接近的场景。

无监督学习  /  自注意力机制  /  生成对抗网络  /  新能源  /  场景生成

With the increased penetration rate of new energy year by year,it is difficult to accurately predict the randomness and fluctuation characteristics of its output,causing a severe challenge to the operation,planning and scheduling of electrical power system. Therefore,modeling for the uncertainty of new energy has attracted more and more attention. To obtain the time sequence characteristics of new energy output scenario more effectively,a new energy scenario generation method was proposed based on data drive,and combined self-attention mechanism with generative adversarial network discriminator with gradient penalty through applying the SA/WGAN model. Through building a deep learning model based on the combination of two models,effectively highlight the timing sequence characteristics of new energy output scenario and enhancing the nonlinear fitting capability in scenario generation. The example results show that,compared with the scenario generation results of original WGAN and WGAN-LSTM,the new energy generation scenario of proposed model can not only effectively improve the accuracy,but also possess the advantages of stable WGAN-GP training results and quick SA calculation speed,which can achieve a more efficient generation of scenarios that is close to the distribution of real new energy scenario.

unsupervised learning  /  self-attention (SA)  /  generative adversarial networks(GAN)  /  new energy  /  scenario generation
王宇昊, 刘海涛, 朱康凯, 仲聪, 马佳伊. 基于SA/WGAN的新能源场景生成方法. 电气传动, 2024 , 54 (6) : 45 -53 . DOI: 10.19457/j.1001-2095.dqcd24775
Yuhao WANG, Haitao LIU, Kangkai ZHU, Cong ZHONG, Jiayi MA. New Energy Scenario Generation Based on SA/WGAN[J]. Electric Drive, 2024 , 54 (6) : 45 -53 . DOI: 10.19457/j.1001-2095.dqcd24775
高渗透率的新能源发电并网将给电网的安全稳定运行带来巨大的挑战[1]。风电、光伏等新能源由于其一次能源的随机性和波动性,导致其发电出力的不确定性。为了应对新能源的不确定性,当前电力系统的运行、调度、规划策略都要进行调整[2]。系统中新能源发电渗透率不断升高之后,其运行调度规划问题将由确定性优化问题转化为不确定性优化问题。目前基于场景分析法的随机优化研究方向主要集中在电网规划和电网调度运行上,并且已经有了一些进展。场景分析法中对新能源发电不确定性建模的精确程度将影响其优化结果的有效性。因此研究新能源场景生成方法可以制定高有效性的电力系统优化方案,对保障高渗透率电力系统的安全经济稳定运行有着重大意义。
较为常用的场景生成方法有抽样方法[3-6]、时间序列建模法[7-8]和人工智能的方法。基于抽样方法的主要有蒙特卡洛抽样法、拉丁超立方抽样法等;基于时间序列建模方法的主要有马尔可夫链法、基于自回归滑动平均模型的时间序列法;基于数据驱动的人工智能方法主要有生成对抗网络法(generative adversarial networks,GAN)。GAN可以基于原始数据,生成大量与原始数据特征接近的样本数据,建立场景集。由于GAN存在训练不稳定、梯度消失、模式崩溃等现象,近年来部分学者对GAN进行了一些改进。文献[9]使用引入Wasserstein距离的生成对抗网络(Wasserstein GAN,WGAN),提高了网络训练的稳定度和样本训练质量,可以更好地学习新能源出力的时间-空间相关性;文献[10]将卷积网络引入改进后的WGAN的网络结构中,提出了一种深度卷积生成对抗网络(deep convolution generative adversarial networks,DCGAN),通过卷积网络较强的特征提取能力来提升GAN生成样本质量;文献[11]参照文献[10]中DCGAN的网络结构,设计了适用于提取风光联合出力曲线特点的网络结构,通过采用改进的条件DCGAN模型生成引入月份标签的风光联合出力场景。但是上述方法还是存在着不能很好地捕捉真实场景的时序特性和非线性拟合能力不强的问题。
针对上述研究存在的问题,本文通过改进文献[12]提出的基于Wasserstein距离和梯度惩罚机制的生成对抗网络(Wasserstein GAN with gradient penalty,WGAN-GP)对新能源出力场景进行不确定性建模,克服了传统方法模型难以建立和出力场景不确定性描述困难的问题。通过采用带有位置编码的自注意力模块构建GAN的判别器,更有效地提取场景中的时序特征,提高了非线性拟合能力,使其能更快更稳定地生成与真实场景相接近的新能源出力场景。
生成对抗网络法最早由Ian J.Goodfellow等人于2014年在《Generative Adversarial Networks》中提出,GAN是一种无监督的神经网络,通过训练真实数据集不断地拟合模型,可以隐式地学习数据内在的概率分布,其基本结构如图1所示[13]。GAN因其可以在原始数据质量不高时,生成大量相似数据,提高生成数据质量被广泛应用于解决新能源场景生成、机组故障诊断等电力数据样本生成的问题。
图1中,生成器G用来捕获数据分布,判别器D用来判断生成样本来自真实数据的概率,两个模型同时训练。生成器G的训练目标是用生成数据去欺骗判别器D,将判别器D判别错误概率最大化,生成器和判别器的损失函数为
L o s s G = - E z p z ( z ) { D [ G ( z ) ] }
L o s s D = - { E x p d a t a ( x ) [ l g D ( x ) ] + E z p Z ( z ) { l g { 1 - D [ G ( z ) ] } } }
式中: E ( )为求期望函数; G ( ) D ( )分别为生成器和判别器函数; p z ( z ) p d a t a ( x )分别为输入噪声 z和样本输入 x的分布[14]
生成器和判别器的训练过程是最大最小值博弈游戏[15],其目标函数为
m i n G m a x D V ( D , G ) = E x p d a t a ( x ) [ l g D ( x ) ] + E z p Z ( z ) { l g { 1 - D [ G ( z ) ] } }
V ( D , G )是一个二分类的交叉熵函数。式(3)的最终目标是使生成样本概率分布 p G ( z ) p d a t a ( x )之间的JS距离最小。
任意函数GD的空间中,使得G重现训练数据的分布唯一解决方案是 D=0.5。在整个训练的过程中不需要任何马尔可夫链或展开的近似推理网络。且当生成器G和判别器D由多层感知器构成时,整个系统可以反向传播进行训练。
原始GAN采用JS距离来描述式(3)中判别器的目标函数,但是训练刚开始时,生成样本与输入样本的概率分布可能没有交集,导致JS距离为恒定值lg2,出现梯度消失的问题。因此文献[12]通过引入Wasserstein距离取代原始GAN中的JS距离,并通过梯度惩罚取代权重裁剪,通过限制梯度不超过 L i p s c h i t z常数 K 0,使训练更加稳定,解决了训练过程中可能出现的梯度消失或爆炸的问题。优化后的WGAN-GP模型的目标函数为
        m i n G m a x D V ( D , G ) = E x p d a t a ( x ) [ D ( x ) ] - E z p Z ( z ) [ D ( G ( z ) ] - λ E · [ | | D ( x ' ) - K 0 | | ] 2
其中
x ' = ε x + ( 1 - ε ) G ( z )
式中: ε为随机数; λ为梯度惩罚权重系数,一般取10。
但是由于改进后的WGAN-GP收敛速度慢,适用于对训练速度要求不高的场景。
基于卷积核的循环神经网络虽然在理论上可以建立更长距离的依赖关系,但是由于信息传递容量的问题以及可能出现梯度消失的现象,建立长距离的依赖关系实际上难以实现。
目前,主要通过增加网络的层数和使用全连接层来建立长距离依赖关系,但是这两种方法都存在缺陷。增加网络层数会导致网络学习速度变慢,全连接层又无法处理变长的输入序列,这两种方法都不能很好地直接建立长距离依赖关系。
文献[16]通过自注意力模型来进行文本表示,通过自注意力机制(self-attention,SA)来“动态”地生成输入向量序列的连接权重,自动赋权解决了全连接层无法处理变长输入序列的缺点,建立了直接的长距离依赖关系。
自注意力机制摆脱了传统的循环神经网络(recurrent neural networks,RNN)、卷积神经网络(convolutional neural networks,CNN)、长短期记忆神经网络(long short term memory,LSTM),可以较好地捕捉全局成对交互关系。设输入序列 X = [ x 1 , , x N ] R D x × N,自动赋权后的输出序列为 H = [ h 1 , , h N ] R D v × N,自注意力制计算过程的示意图如图2所示。图中, Q K V分别为每个输入映射到的3个不同空间:查询空间、键空间、值空间,如下式:
Q = W q X R D k × N         W q R D k × D x K = W k X R D k × N       W k R D k × D x V = W v X R D v × N       W v R D v × D x
式中: W q W k W v为外线性映射的参数矩阵; D x为输入矩阵的行数; N为输入矩阵的列数; D k为将输入映射到查询空间、键空间的行数; D v为将输入映射到值空间的行数。
将每个输入映射到查询空间、键空间、值空间3个不同的空间内,得到3个不同的向量:查询向量 q i R D k、键向量 k i R D k和值向量 v i R D v Q = [ q 1 , , q i , , q N ]为查询向量构成的矩阵, K = [ k 1 , , k i , , k N ]为键向量构成的矩阵, V = [ v 1 , , v i , , v N ]为值向量构成的矩阵。
用键值对格式来表示第 n个查询向量 q n Q,键值对中键用于计算注意力分布 α i,值用于计算聚合信息。用 ( K , V ) = [ ( k 1 , v 1 ) , , ( k N , v N ) ]来表示查询向量的 N组输入信息,输出的注意力函数 h n
$\begin{aligned} h_{n} & =\operatorname{att}\left[(\boldsymbol{K}, \boldsymbol{V}), \boldsymbol{q}_{n}\right] \\ & =\sum_{j=1}^{N} \boldsymbol{\alpha}_{n j} \boldsymbol{v}_{j} \\ & =\sum_{j=1}^{N} \operatorname{softmax}\left[s\left(\boldsymbol{k}_{j}, \boldsymbol{q}_{n}\right)\right] \boldsymbol{v}_{j} \end{aligned}$
式中:att(·)为注意力函数;s(·)为缩放点积模型形式的注意力打分函数; n j分别为输入、输出向量序列的位置, n [ 1 , N ] j [ 1 , N ] α n j为第 n个输出关注到第 j个输入的权重; s o f t m a x ( )为按列归一化函数。
若使用缩放点积来表示注意力打分函数,可以更高效地得到打分函数,其输出向量序列可以简化为
H = V s o f t m a x ( K T Q D k )
s o f t m a x ( )使其输出向量 H的每一列的值都在0到1之间。
由式(6)中注意力函数的计算过程可以显示出所有的输入向量都参与了运算,捕捉了全局成对交互关系。自注意力模型捕捉全局成对交互关系的示意图如图3所示。自注意力模型可以作为神经网络中的一层,既可以代替传统的RNN/CNN,也可以和RNN/CNN交替使用。
图4a是本文设计的生成器的结构示意图,图4b是基于自注意力模型的判别器结构示意图。用自注意力模型来改善传统WGAN-GP的判别器结构,提高了网络的收敛速度,可以更高效稳定地生成高质量的数据样本,并根据新能源出力数据特点,设计适用于新能源出力场景生成的网络结构,最终提出了本文所用的基于自注意力模型改进的WGAN-GP模型(improved WGAN-GP based on SA,SA/WGAN)。
本文采用融合自注意力机制和WGAN-GP的SA/WGAN模型对新能源出力场景进行场景生成。图5为本文提出的基于SA/WGAN的新能源出力场景生成方法的整体流程图。
在训练过程中,将输入噪声输入生成器得到生成数据,将生成数据引入由SA改进的判别器区分真实数据与生成数据。由于生成器与判别器的权重更新速率不匹配,所以选择判别器每更新5次,更新1次生成器,并不断重复以上过程,不断迭代。随着迭代的进行,生成器G不断调整网络的权值,不断优化,不断提高生成数据与真实数据的相似程度,判别器D也不断地提高判别真实数据与生成数据的能力,最终判别器D无法区分数据的来源是真实数据还是生成数据。两者不断博弈最终达到了纳什均衡,此时基于自注意力模型改进后的WGAN-GP仅需要在生成器输入高斯噪声即可以进行新能源出力场景生成。
本文选取的风电、光伏出力数据样本集来自甘肃宁夏某地。所选数据的采样时间间隔均为15 min,一天96个采样点。将样本集中的数据进行检测,找出因故障导致的功率缺失点,对功率缺失的点进行数据填充,并通过最大最小值归一化,使原始数据等比例缩放到0~1之间,最后将96个采样点数据扩展为121 ×1的输入序列,预处理后的数据如图6所示。
将预处理后的数据作为样本集,并在样本集中每4天选取1天数据作为测试集,其余作为训练集进行训练。图7~图9分别为本文所提方法在训练过程中迭代0次(即训练开始时生成器输入为高斯噪声的初始状态)、1 200次、4 000次生成不同时刻的新能源出力场景效果图。
图7~图9中曲线横坐标轴为时间变化趋势,纵坐标为归一化后的出力值。其中,黑色曲线为生成的光伏出力曲线,灰色曲线为生成的风电出力数据。训练刚开始时生成器G生成的数据是符合高斯分布的随机噪声,此时生成器G还没有学习到风电、光伏的出力特点。随着训练的进行,训练集迭代次数的不断增加,生成器G不断优化,开始学习数据的特点,生成器输出结果的质量不断提高,生成数据开始出现周期性变化。其中光伏出力数据特点学习较快,生成器G开始捕捉到光伏出力数据白天出力较多、夜晚出力为0的特点。最终当生成器G和判别器D达到纳什均衡时,风电、光伏出力数据较为光滑,很好地捕捉到了风电出力的波动性大及光伏出力受光照强度影响的特点,总体生成效果符合预期。从测试集中随机选择未被学习过的样本与生成数据进行对比,生成数据与真实数据比较如图10所示。
图10可以看出,本文方法的生成场景与真实场景基本一致,生成场景准确地捕捉了真实场景出力数据的峰谷值、波动性等特征。
为验证所用方法生成场景的有效性,本文从时序特性、概率分布特性两个方面进行场景生成质量评估。
为验证真实场景与本文所用方法生成场景的时序相关性,分别对风电、光伏真实场景与生成场景的自相关系数做对比。自相关系数可以反映原新能源出力序列与滞后 τ h的新能源出力序列的相关性,可应用于评估新能源场景生成中生成场景学习真实场景时序相关性的质量。图11a图11b分别为风电和光伏生成出力场景与真实出力场景在滞后时间 τ在0~7 h的自相关系数对比图。
由自相关系数对比图可以看出,生成场景与真实场景在滞后0~7 h的自相关系数基本一致,生成场景很好地学习了真实场景的时序特性。
将训练结果与风电、光伏的真实出力数据进行对比分析,并绘制累积概率密度函数图(cumulative distribution function,CDF)。CDF图可以直观地反映出力数据的概率分布特性,被广泛应用于场景生成质量评估。图12a图12b分别为风电和光伏生成数据与真实数据累积概率密度函数对比图。
图12可以看出,本文方法生成场景的累积概率密度曲线拟合度较高,很好地学习了出力数据的潜在概率分布特性。
将本文方法生成结果与原始WGAN-GP和文献[17]中所提的基于LSTM改进的WGAN(improved WGAN based on LSTM,WGAN-LSTM)比较,图13a图13b分别为各种方法下风电和光伏生成出力数据与真实数据对比的CDF图。
由于图13中几种方法的CDF曲线相近,因此引入均方误差(mean-square error,MSE)来比较各种方法生成结果与真实数据累积概率密度函数相近程度。表1为3种方法在生成风电出力和光伏出力时与真实场景的均方误差。
表1可知,本文方法所生成场景与真实场景累积概率密度函数的MSE分别为0.002 33 和0.001 82,优于原始WGAN-GP和WGAN-LSTM生成场景与真实场景累积概率密度函数的MSE。综上所述,所提的新能源场景生成方法的出力场景与真实场景更加接近,场景生成效果优于现有方法,更好地捕捉了真实出力场景的内部联系,在整体上更加符合真实场景的相关规律性。
本文提出了一种基于SA/WGAN的新能源场景生成方法,通过改进原始GAN,实现高效稳定地生成与真实场景相近的出力场景。通过对某地风光历史实测数据投入本文模型进行训练,并与原始WGAN-GP 和WGAN-LSTM的训练结果进行对比分析,证明本文方法具有以下优势:
1)使用基于Wasserstein距离和梯度惩罚函数的生成对抗网络(WGAN-GP)法对新能源场景进行不确定性建模,克服了原始 WGAN 训练困难、收敛速度慢、梯度消失等问题,并针对风电、光伏出力的不确定性问题,设计了适用于新能源出力场景生成的网络结构,提高了场景生成质量。
2)为了更有效地提取场景中的时序特征,采用带有位置编码的自注意力模块(SA)改进WGAN-GP的判别器结构,强化网络对历史数据时序特性的捕捉能力和模型非线性拟合能力,使模型生成结果更能表现新能源出力场景的时序特性和关联性。相比于传统序列神经网络CNN,RNN,LSTM等,自注意力模块的计算更具并行性,计算速度更快,进一步提高了收敛速度。
  • 国家自然科学基金(51777197)
  • 江苏省高校自然科学研究重大项目(22KJA470005)
参考文献 引证文献
排序方式:
[1]
CHEN X, ZHANG T, YE W, et al. Blockchain-based electric vehicle incentive system for renewable energy consumption[J]. IEEE Transactions on Circuits and Systems II:Express Briefs, 2020, 68(1):396-400.
[2]
徐艳春, 蒋伟俊, 汪平, 等. 考虑暂态电压定稳定的含高渗透率风光的电网动态无功规划方法[J]. 电力自动化设备, 2022, 42(8):79-88.
XU Yanchun, JIANG Weijun, WANG Ping, et al. Dynamic reactive power planning method for power grid with high permeability wind power and photovoltaic considering transient voltage stability[J]. Electric Power Automation Equipment, 2022, 42 (8):79-88.
[3]
杨茂, 董昊. 基于数值天气预报风速和蒙特卡洛法的短期风电功率区间预测[J]. 电力系统自动化, 2021, 45(5):79-85.
YANG Mao, DONG Hao. Short-term wind power interval prediction based on wind speed of numerical weather prediction and Monte Carlo method[J]. Automation of Electric Power Systems, 2021, 45(5):79-85.
[4]
姚志力, 江斌开, 龚春阳, 等. 计及电动汽车充电负荷特性的区域多能源站规划设计[J]. 电网技术, 2022, 46(9):3304-3318.
YAO Zhili, JIANG Binkai, GONG Chunyang, et al. Planning and design of regional multi-energy stations considering electric vehicle charging load characteristics[J]. Power System Technology, 2022, 46(9):3304-3318.
[5]
马燕峰, 骆泽榕, 赵书强, 等. 基于改进蒙特卡洛混合抽样的含风光电力系统风险评估[J]. 电力系统保护与控制, 2022, 50(9):75-83.
MA Yanfeng, LUO Zerong, ZHAO Shuqiang, et al. Risk assessment of a power system containing wind power and photovoltaic based on improved Monte Carlo mixed sampling[J]. Power System Protection and Control, 2022, 50(9):75-83.
[6]
艾小猛, 塔伊尔江·巴合依, 杨立滨, 等. 基于场景集的含风电电力系统旋转备用优化[J]. 电网技术, 2018, 42(3):835-841.
AI Xiaomeng, Tayierjiang B, YANG Libini, et al. Optimizing the spinning reserve in wind power system using scenario method[J]. Power System Technology, 2018, 42(3):835-841.
[7]
董雷, 孟天骄, 陈乃仕, 等. 采用马尔可夫链—多场景技术的交直流主动配电网优化调度[J]. 电力系统自动化, 2018, 42(5):147-153.
DONG Lei, MENG Tianjiao, CHEN Naishi, et al. Optimized scheduling of AC/DC hybrid active distribution network using Markov chains and multiple scenarios technique[J]. Automation of Electric Power Systems, 2018, 42(5):147-153.
[8]
董锴, 蔡新雷, 崔艳林, 等. 基于马尔科夫链的电动汽车聚合建模及多模式调频控制策略[J]. 电网技术, 2022, 46(2):622-634.
DONG Kai, CAI Xinlei, CUI Yanlin, et al. Aggregation modeling based on Markov chain and multi-mode control strategies of aggregated electric vehicles for frequency regulation[J]. Power System Technology, 2022, 46(2):622-634.
[9]
CHEN Y, WANG Y, KIRSCHEN D, et al. Model-free renewable scenario generation using generative adversarial networks[J]. IEEE Transactions on Power Systems, 2018, 33(3):3265-3275.
[10]
YU Yang, GONG Zhiqiang, ZHONG Ping, et al. Unsupervised representation learning with deep convolutional neural network for remote sensing images[C]// International Conference on Ima-ge and Graphics, 2017:97-108.
[11]
顾洁, 刘书琪, 胡玉, 等. 基于深度卷积生成对抗网络场景生成的间歇式分布式电源优化配置[J]. 电网技术, 2021, 45(5):1742-1751.
GU Jie, LIU Shuqi, HU Yu, et al. Optimal allocation of intermittent distributed generation based on deep convolutions generative adversarial network in scenario generation[J]. Power System Technology, 2021, 45(5):1742-1751.
[12]
GULRAJANI I, AHMED F, ARJOVSKY M, et al. Improved training of Wasserstein GANs[C]//In Proceedings of the 31st International Conference on Neural Information Processing Systems,Long Beach,CA, USA:NIPS, 2017:5769-5779.
[13]
GOODFELLOW I, POUGET-ABADIE J, MIRZA M, et al. Ge-nerative adversarial nets[C]// Proceedings of the 27th International Conference on Neural Information Processing Systems. Montreal, Canada: MIT Press, 2014:2672-2680.
[14]
邵振国, 张承圣, 陈飞雄, 等. 生成对抗网络及其在电力系统中的应用综述[J]. 中国电机工程学报, 2023, 43(3):987-1004.
SHAO Zhenguo, ZHANG Chengsheng, CHEN Feixiong, et al. A review on generative adversarial networks for power system applications[J]. Proceedings of the CSEE, 2023, 43(3):987-1004.
[15]
兰健, 郭庆来, 周艳真, 等. 基于生成对抗网络和模型迁移的电力系统典型运行方式样本生成[J]. 中国电机工程学报, 2022, 42(8):2889-2900.
LAN Jian, GUO Qinglai, ZHOU Yanzhen, et al. Generation of power system typical operation mode samples:a generation adversarial network and model-based transfer learning approach[J]. Proceedings of the CSEE, 2022, 42(8):2889-2900.
[16]
VASWANI A, SHAZEER N, PARMAR N, et al. Attention is all you need[C]// 31st International Conference on Neural Information Processing Systems,Long Beach,USA,2017:5999-6009.
[17]
朱庆, 郑红娟, 唐子逸, 等. 基于生成对抗网络的综合能源负荷场景生成方法[J]. 电力建设, 2021, 42(12):1-8.
ZHU Qing, ZHENG Hongjuan, TANG Ziyi, et al. Load scenario generation of integrated energy system using generative adversarial networks[J]. Electric Power Construction, 2021, 42(12):1-8.
2024年第54卷第6期
PDF下载
147
65
引用本文
BibTeX
文章信息
doi: 10.19457/j.1001-2095.dqcd24775
  • 接收时间:2022-11-16
  • 首发时间:2025-11-26
  • 出版时间:2024-06-20
补充材料
相关文章
文章信息
作者
出版历史
  • 收稿日期:2022-11-16
  • 修回日期:2022-12-24
基金
国家自然科学基金(51777197)
江苏省高校自然科学研究重大项目(22KJA470005)
作者信息
    1 南京工程学院 电力工程学院,江苏 南京 210000
    2 江苏省配电网智能技术与装备协同创新中心,江苏 南京 210000
参考文献
分享链接
https://castjournals.cast.org.cn/joweb/dqcd/CN/10.19457/j.1001-2095.dqcd24775
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