Article(id=1154037274888102695, tenantId=1146029695717560320, journalId=1146031654075715584, issueId=1154037268550509325, articleNumber=null, orderNo=null, doi=10.13234/j.issn.2095-2805.2024.4.133, pmid=null, cstr=null, oa=null, hot=null, price=null, onlineType=0, articleFormat=0, articleType=null, articleTypeStr=null, receivedDate=1640534400000, receivedDateStr=2021-12-27, revisedDate=1648051200000, revisedDateStr=2022-03-24, acceptedDate=1648137600000, acceptedDateStr=2022-03-25, onlineDate=1753073527309, onlineDateStr=2025-07-21, pubDate=1722268800000, pubDateStr=2024-07-30, doiRegisterDate=null, doiRegisterDateStr=null, onlineIssueDate=1753073527309, onlineIssueDateStr=2025-07-21, onlineJustAcceptDate=null, onlineJustAcceptDateStr=null, onlineFirstDate=null, onlineFirstDateStr=null, sourceXml=null, magXml=null, createTime=1753073527309, creator=13701087609, updateTime=1753073527309, updator=13701087609, issue=Issue{id=1154037268550509325, tenantId=1146029695717560320, journalId=1146031654075715584, year='2024', volume='22', issue='4', pageStart='1', pageEnd='338', issueExtLink='null', onlineDate='null', pubDate='null', beforeIssueId=null, nextIssueId=null, price=null, status=1, issueComplete=1, articleOrder=1, issueType=-1, specialIssue=0, createTime=1753073525798, creator=13701087609, updateTime=1753780979931, updator=13701087609, preIssue=null, nextIssue=null, ext={EN=IssueExt(id=1157004546338804561, tenantId=1146029695717560320, journalId=1146031654075715584, issueId=1154037268550509325, language=EN, specialIssueTitle=, coverIllustrator=, specialIssueEditor=, specialIssueAbout=), CN=IssueExt(id=1157004546338804562, tenantId=1146029695717560320, journalId=1146031654075715584, issueId=1154037268550509325, language=CN, specialIssueTitle=, coverIllustrator=, specialIssueEditor=, specialIssueAbout=)}, issueFiles=null}, startPage=133, endPage=142, ext={EN=ArticleExt(id=1154037275785683755, articleId=1154037274888102695, tenantId=1146029695717560320, journalId=1146031654075715584, language=EN, title=Ultra-short-term Wind Speed Prediction for Multiple Wind Farms Based on Aggregated Spatio-temporal Graph Convolutional Networks, columnId=1152281492550987902, journalTitle=Journal of Power Supply, columnName=Renewable Energy System, runingTitle=null, highlight=null, articleAbstract=

In a certain environment where regional wind farms distribute irregularly, the traditional convolutional neural network prediction method cannot reflect the distribution states or influence relationship of regional wind farms, and it is difficult to accurately predict the wind speed. First, to solve this problem, the technology of graph convolutional networks is used for feature modeling, and the connected graph and weight matrix are established according to the topology of multiple wind farms and the cross-correlation coefficient of wind speed in each region. Second, depending on the time dynamic characteristics of wind speed at wind farms, an improved parallel convolution structure is used to obtain the correlation between wind speed series in multiple time periods at the same wind farm. Third, based on the spatial correlation and delay effect of wind speed at wind farms, the spatio-temporal characteristics of wind speed in different regions are aggregated by using a second-order aggregation method. Finally, the verification of data from one regional wind farm shows that the proposed method can extract the spatio-temporal characteristics and improve the performance of ultra short-term wind speed prediction for multiple wind farms on 0-4 h prediction scale.

, 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=Chenxiao XU, Chenggang CUI, Weimin GUO, Ning YANG, Bei LIU, Qingye MENG), CN=ArticleExt(id=1154037342701609040, articleId=1154037274888102695, tenantId=1146029695717560320, journalId=1146031654075715584, language=CN, title=基于聚合时空图卷积网络的多风场超短期风速预测, columnId=1149829992055595012, journalTitle=电源学报, columnName=新能源系统, runingTitle=null, highlight=null, articleAbstract=

在一定环境内区域风电场呈不规则分布的条件下,传统卷积神经网络预测方法无法体现出各区域风场的分布状态和影响关系,难以实现对风速的准确预测。针对此问题,采用图卷积网络进行特征建模,并根据多风场的拓扑结构和各区域风场风速的互相关系数建立连通图和权重矩阵。其次,依赖风场风速的时间动态特征,采用改进并列式卷积结构获取同一风场下多时间段的风速序列相关性。再次,利用风场风速的空间相关性和延时效应,采用二阶聚合方法将不同区域内风速的时空特征聚合。最后,经某区域风场数据验证表明,在0~4h预测尺度下该方法在多风场超短期风速预测中具有提取时空特征并提升预测性能的效果。

, correspAuthors=null, authorNote=null, correspAuthorsNote=null, copyrightStatement=null, copyrightOwner=null, extLink=null, articleAbsUrl=null, sourceXml=8Qt5gdy5LO54UiY+dDDDtw==, magXml=8i4mOPMO5enw0yh6p/2DEw==, pdfUrl=null, pdf=b3dKB06d3NLRveIThhA/ew==, pdfFileSize=null, pdfExtLink=null, richHtmlUrl=null, mobilePdfUrl=null, reviewReport=null, pdfFirstPage=null, abstractGraph=null, abstractGraphContent=null, abstractVideo=null, citation=null, cebUrl=null, magXmlContent=TizK0eiwbZ4lGkUexBVwqg==, mapNumber=null, authorCompany=null, fund=null, authors=

徐辰晓(1996-),男,硕士研究生。研究方向:人工智能在风力发电预测中的应用。E-mail: 694190047@qq.com。

崔承刚(1981-),男,通信作者,博士,副教授。研究方向:深度学习与强化学习。E-mail: cgcui@shiep.edu.cn。

郭为民(1972-),男,教授级高级工程师。研究方向:火电厂智能规划建设。E-mail: GUOWEIMIN8@crpower.com.cn。

杨宁(1976-),男,博士研究生,教授。研究方向:基于强化学习的功率优化。E-mail:yangning@shiep.edu.cn。

刘备(1993-),男,硕士研究生,工程师。研究方向:超临界火电机组主蒸汽温度控制算法。E-mail : 377179519@qq.com。

孟青叶(1991-),女,硕士研究生,工程师。研究方向:数据驱动的变速变桨距风力发电机组功率优化控制。E-mail: M_qingye@163.com。

, authorsList=徐辰晓, 崔承刚, 郭为民, 杨宁, 刘备, 孟青叶)}, authors=[Author(id=1154048171547288507, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, orderNo=0, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=694190047@qq.com, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1154048171601814461, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, authorId=1154048171547288507, language=EN, stringName=Chenxiao XU, firstName=Chenxiao, middleName=null, lastName=XU, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1 College of Automation Engineering Shanghai University of Electrical Power Shanghai 200090 China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1154048171652146110, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, authorId=1154048171547288507, language=CN, stringName=徐辰晓, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1 上海电力大学 自动化工程学院 上海 200090, bio={"img":"mTWagUR/TBKQYbnwehLCng==","content":"

徐辰晓(1996-),男,硕士研究生。研究方向:人工智能在风力发电预测中的应用。E-mail: 694190047@qq.com。

"}, bioImg=mTWagUR/TBKQYbnwehLCng==, bioContent=

徐辰晓(1996-),男,硕士研究生。研究方向:人工智能在风力发电预测中的应用。E-mail: 694190047@qq.com。

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1154048171404682164, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, xref=1, ext=[AuthorCompanyExt(id=1154048171408876469, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, companyId=1154048171404682164, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 College of Automation Engineering Shanghai University of Electrical Power Shanghai 200090 China), AuthorCompanyExt(id=1154048171413070774, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, companyId=1154048171404682164, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 上海电力大学 自动化工程学院 上海 200090)])]), Author(id=1154048171719254977, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, orderNo=1, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=cgcui@shiep.edu.cn, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1154048171790558149, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, authorId=1154048171719254977, language=EN, stringName=Chenggang CUI, firstName=Chenggang, middleName=null, lastName=CUI, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1 College of Automation Engineering Shanghai University of Electrical Power Shanghai 200090 China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1154048171857667015, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, authorId=1154048171719254977, language=CN, stringName=崔承刚, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1 上海电力大学 自动化工程学院 上海 200090, bio={"content":"

崔承刚(1981-),男,通信作者,博士,副教授。研究方向:深度学习与强化学习。E-mail: cgcui@shiep.edu.cn。

"}, bioImg=null, bioContent=

崔承刚(1981-),男,通信作者,博士,副教授。研究方向:深度学习与强化学习。E-mail: cgcui@shiep.edu.cn。

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1154048171404682164, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, xref=1, ext=[AuthorCompanyExt(id=1154048171408876469, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, companyId=1154048171404682164, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 College of Automation Engineering Shanghai University of Electrical Power Shanghai 200090 China), AuthorCompanyExt(id=1154048171413070774, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, companyId=1154048171404682164, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 上海电力大学 自动化工程学院 上海 200090)])]), Author(id=1154048171903804362, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, orderNo=2, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=GUOWEIMIN8@crpower.com.cn, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1154048171962524621, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, authorId=1154048171903804362, language=EN, stringName=Weimin GUO, firstName=Weimin, middleName=null, lastName=GUO, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2 Rundian Energy Science and Technology Co., Ltd Zhengzhou 450052 China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1154048172017050575, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, authorId=1154048171903804362, language=CN, stringName=郭为民, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2 润电能源科学技术有限公司 郑州 450052, bio={"content":"

郭为民(1972-),男,教授级高级工程师。研究方向:火电厂智能规划建设。E-mail: GUOWEIMIN8@crpower.com.cn。

"}, bioImg=null, bioContent=

郭为民(1972-),男,教授级高级工程师。研究方向:火电厂智能规划建设。E-mail: GUOWEIMIN8@crpower.com.cn。

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1154048171496956855, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, xref=2, ext=[AuthorCompanyExt(id=1154048171505345464, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, companyId=1154048171496956855, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 Rundian Energy Science and Technology Co., Ltd Zhengzhou 450052 China), AuthorCompanyExt(id=1154048171509539769, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, companyId=1154048171496956855, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 润电能源科学技术有限公司 郑州 450052)])]), Author(id=1154048172084159443, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, orderNo=3, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=yangning@shiep.edu.cn, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1154048172142879701, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, authorId=1154048172084159443, language=EN, stringName=Ning YANG, firstName=Ning, middleName=null, lastName=YANG, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1 College of Automation Engineering Shanghai University of Electrical Power Shanghai 200090 China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1154048172193211350, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, authorId=1154048172084159443, language=CN, stringName=杨宁, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1 上海电力大学 自动化工程学院 上海 200090, bio={"content":"

杨宁(1976-),男,博士研究生,教授。研究方向:基于强化学习的功率优化。E-mail:yangning@shiep.edu.cn。

"}, bioImg=null, bioContent=

杨宁(1976-),男,博士研究生,教授。研究方向:基于强化学习的功率优化。E-mail:yangning@shiep.edu.cn。

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1154048171404682164, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, xref=1, ext=[AuthorCompanyExt(id=1154048171408876469, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, companyId=1154048171404682164, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 College of Automation Engineering Shanghai University of Electrical Power Shanghai 200090 China), AuthorCompanyExt(id=1154048171413070774, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, companyId=1154048171404682164, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 上海电力大学 自动化工程学院 上海 200090)])]), Author(id=1154048172239348696, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, orderNo=4, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=377179519@qq.com, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1154048172289680346, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, authorId=1154048172239348696, language=EN, stringName=Bei LIU, firstName=Bei, middleName=null, lastName=LIU, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2 Rundian Energy Science and Technology Co., Ltd Zhengzhou 450052 China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1154048172335817691, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, authorId=1154048172239348696, language=CN, stringName=刘备, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2 润电能源科学技术有限公司 郑州 450052, bio={"content":"

刘备(1993-),男,硕士研究生,工程师。研究方向:超临界火电机组主蒸汽温度控制算法。E-mail : 377179519@qq.com。

"}, bioImg=null, bioContent=

刘备(1993-),男,硕士研究生,工程师。研究方向:超临界火电机组主蒸汽温度控制算法。E-mail : 377179519@qq.com。

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1154048171496956855, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, xref=2, ext=[AuthorCompanyExt(id=1154048171505345464, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, companyId=1154048171496956855, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 Rundian Energy Science and Technology Co., Ltd Zhengzhou 450052 China), AuthorCompanyExt(id=1154048171509539769, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, companyId=1154048171496956855, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 润电能源科学技术有限公司 郑州 450052)])]), Author(id=1154048172381955037, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, orderNo=5, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=M_qingye@163.com, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1154048172432286687, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, authorId=1154048172381955037, language=EN, stringName=Qingye MENG, firstName=Qingye, middleName=null, lastName=MENG, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2 Rundian Energy Science and Technology Co., Ltd Zhengzhou 450052 China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1154048172474229728, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, authorId=1154048172381955037, language=CN, stringName=孟青叶, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2 润电能源科学技术有限公司 郑州 450052, bio={"content":"

孟青叶(1991-),女,硕士研究生,工程师。研究方向:数据驱动的变速变桨距风力发电机组功率优化控制。E-mail: M_qingye@163.com。

"}, bioImg=null, bioContent=

孟青叶(1991-),女,硕士研究生,工程师。研究方向:数据驱动的变速变桨距风力发电机组功率优化控制。E-mail: M_qingye@163.com。

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1154048171496956855, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, xref=2, ext=[AuthorCompanyExt(id=1154048171505345464, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, companyId=1154048171496956855, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 Rundian Energy Science and Technology Co., Ltd Zhengzhou 450052 China), AuthorCompanyExt(id=1154048171509539769, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, companyId=1154048171496956855, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 润电能源科学技术有限公司 郑州 450052)])])], keywords=[Keyword(id=1154048173342450657, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, language=EN, orderNo=1, keyword=Wind speed prediction), Keyword(id=1154048173409559522, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, language=EN, orderNo=2, keyword=aggregated spatio-temporal graph convolutional networks), Keyword(id=1154048173464085475, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, language=EN, orderNo=3, keyword=spatio-temporal correlation), Keyword(id=1154048173510222820, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, language=CN, orderNo=1, keyword=风速预测), Keyword(id=1154048173547971557, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, language=CN, orderNo=2, keyword=聚合时空图卷积网络), Keyword(id=1154048173598303206, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, language=CN, orderNo=3, keyword=时空相关性)], refs=[Reference(id=1154048177104740363, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, doi=null, pmid=null, pmcid=null, year=2021, volume=54, issue=5, pageStart=1, pageEnd=6, url=null, language=null, rfNumber=[1], rfOrder=0, authorNames=谭显东, 刘俊, 徐志成, journalName=中国电力, refType=null, unstructuredReference=谭显东, 刘俊, 徐志成, 等. “双碳”目标下“十四五”电力供需形势[J]. 中国电力, 2021. 54(5): 1-6., articleTitle=“双碳”目标下“十四五”电力供需形势, refAbstract=null), Reference(id=1154048177171849228, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, doi=null, pmid=null, pmcid=null, year=2021, volume=54, issue=5, pageStart=1, pageEnd=6, url=null, language=null, rfNumber=[1], rfOrder=1, authorNames=Tan Xiandong, Liu Jun, Xu Zhicheng, journalName=Electric Power, refType=null, unstructuredReference=Tan Xiandong, Liu Jun, Xu Zhicheng, et al. Power supply and demand balance during the 14th five-year plan period under the goal of carbon emission peak and carbon neutrality[J]. Electric Power, 2021. 54(5): 1-6 (in Chinese)., articleTitle=Power supply and demand balance during the 14th five-year plan period under the goal of carbon emission peak and carbon neutrality, refAbstract=null), Reference(id=1154048177226375181, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, doi=null, pmid=null, pmcid=null, year=2016, volume=37, issue=2, pageStart=171, pageEnd=null, url=null, language=null, rfNumber=[2], rfOrder=2, authorNames=本刊讯, journalName=电力与能源, refType=null, unstructuredReference=本刊讯. 国家电网公司在京发布《国家电网公司促进新能源发展白皮书(2016)》[J]. 电力与能源, 2016. 37(2): 171., articleTitle=国家电网公司在京发布《国家电网公司促进新能源发展白皮书(2016)》, refAbstract=null), Reference(id=1154048177272512526, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, doi=null, pmid=null, pmcid=null, year=2016, volume=37, issue=2, pageStart=171, pageEnd=null, url=null, language=null, rfNumber=[2], rfOrder=3, authorNames=State grid corporation of China, journalName=Power & Energy, refType=null, unstructuredReference=State grid corporation of China. released the white paper on state grid corporation's promotion of new energy development (2016) in Beijing[J]. Power & Energy, 2016. 37(2): 171. (in Chinese)., articleTitle=released the white paper on state grid corporation's promotion of new energy development (2016) in Beijing, refAbstract=null), Reference(id=1154048177318649871, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, doi=null, pmid=null, pmcid=null, year=2021, volume=14, issue=7, pageStart=1894, pageEnd=null, url=null, language=null, rfNumber=[3], rfOrder=4, authorNames=Peng Xiaosheng, Cheng Kai, Lang Jianxun, journalName=Energies, refType=null, unstructuredReference=Peng Xiaosheng, Cheng Kai, Lang Jianxun, et al. Short-term wind power prediction for wind farm clusters based on SFFS feature selection and BLSTM deep learning[J]. Energies, 2021. 14(7): 1894., articleTitle=Short-term wind power prediction for wind farm clusters based on SFFS feature selection and BLSTM deep learning, refAbstract=null), Reference(id=1154048177364787216, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, doi=null, pmid=null, pmcid=null, year=2017, volume=41, issue=1, pageStart=79, pageEnd=85, url=null, language=null, rfNumber=[4], rfOrder=5, authorNames=崔嘉, 杨俊友, 杨理践, journalName=电网技术, refType=null, unstructuredReference=崔嘉, 杨俊友, 杨理践, 等. 基于改进 CFD与小波混合神经网络组合的风电场功率预测方法[J]. 电网技术, 2017. 41(1): 79-85., articleTitle=基于改进 CFD与小波混合神经网络组合的风电场功率预测方法, refAbstract=null), Reference(id=1154048177415118865, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, doi=null, pmid=null, pmcid=null, year=2017, volume=41, issue=1, pageStart=79, pageEnd=85, url=null, language=null, rfNumber=[4], rfOrder=6, authorNames=Cui Jia, Yang Junyou, Yang Lijian, journalName=Power System Technology, refType=null, unstructuredReference=Cui Jia, Yang Junyou, Yang Lijian, et al. New method of combined wind power forecasting based on improved CFD and wavelet-HNN model[J]. Power System Technology, 2017. 41(1): 79-85 (in Chinese)., articleTitle=New method of combined wind power forecasting based on improved CFD and wavelet-HNN model, refAbstract=null), Reference(id=1154048177515782162, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, doi=null, pmid=null, pmcid=null, year=2021, volume=220, issue=null, pageStart=119692, pageEnd=null, url=null, language=null, rfNumber=[5], rfOrder=7, authorNames=Peng Xiaosheng, Wang Hongyu, Lang Jianxun, journalName=Energy, refType=null, unstructuredReference=Peng Xiaosheng, Wang Hongyu, Lang Jianxun, et al. EALSTM-QR: Interval wind-power prediction model based on numerical weather prediction and deep learning[J]. Energy, 2021. 220: 119692., articleTitle=EALSTM-QR: Interval wind-power prediction model based on numerical weather prediction and deep learning, refAbstract=null), Reference(id=1154048177561919507, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, doi=null, pmid=null, pmcid=null, year=2021, volume=58, issue=4, pageStart=121, pageEnd=127, url=null, language=null, rfNumber=[6], rfOrder=8, authorNames=徐鹏超, 李琰, 赵艳雷, journalName=电测与仪表, refType=null, unstructuredReference=徐鹏超, 李琰, 赵艳雷. 基于自适应变异蝙蝠优化 BP 神经网络的短期风电功率预测[J]. 电测与仪表, 2021. 58(4): 121-127., articleTitle=基于自适应变异蝙蝠优化 BP 神经网络的短期风电功率预测, refAbstract=null), Reference(id=1154048177603862548, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, doi=null, pmid=null, pmcid=null, year=2021, volume=58, issue=4, pageStart=121, pageEnd=127, url=null, language=null, rfNumber=[6], rfOrder=9, authorNames=Xu Pengchao, Li Yan, Zhao Yanlei, journalName=Electrical Measurement & Instrumentation, refType=null, unstructuredReference=Xu Pengchao, Li Yan, Zhao Yanlei. Short-term wind power prediction based on BP neural network with adaptive mutation bat optimization algorithm[J]. Electrical Measurement & Instrumentation, 2021. 58(4): 121-127 (in Chinese)., articleTitle=Short-term wind power prediction based on BP neural network with adaptive mutation bat optimization algorithm, refAbstract=null), Reference(id=1154048177654194197, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, doi=null, pmid=null, pmcid=null, year=2020, volume=11, issue=1, pageStart=509, pageEnd=523, url=null, language=null, rfNumber=[7], rfOrder=10, authorNames=Zhu Qiaomu, Chen Jinfu, Shi Dongyuan, journalName=IEEE Transactions on Sustainable Energy, refType=null, unstructuredReference=Zhu Qiaomu, Chen Jinfu, Shi Dongyuan, et al. Learning temporal and spatial correlations jointly: A unified framework for wind speed prediction[J]. IEEE Transactions on Sustainable Energy, 2020. 11(1): 509-523., articleTitle=Learning temporal and spatial correlations jointly: A unified framework for wind speed prediction, refAbstract=null), Reference(id=1154048177704525846, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, doi=null, pmid=null, pmcid=null, year=2021, volume=45, issue=1, pageStart=28, pageEnd=35, url=null, language=null, rfNumber=[8], rfOrder=11, authorNames=凡航, 张雪敏, 梅生伟, journalName=电力系统自动化, refType=null, unstructuredReference=凡航, 张雪敏, 梅生伟, 等. 基于时空神经网络的风电场超短期风速预测模型[J]. 电力系统自动化, 2021. 45(1): 28-35., articleTitle=基于时空神经网络的风电场超短期风速预测模型, refAbstract=null), Reference(id=1154048177763246103, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, doi=null, pmid=null, pmcid=null, year=2021, volume=45, issue=1, pageStart=28, pageEnd=35, url=null, language=null, rfNumber=[8], rfOrder=12, authorNames=Fan Hang, Zhang Xuemin, Mei Shengwei, journalName=Automation of Electric Power Systems, refType=null, unstructuredReference=Fan Hang, Zhang Xuemin, Mei Shengwei, et al. Ultra-short-term wind speed prediction model for wind farms based on spatiotemporal neural network[J]. Automation of Electric Power Systems, 2021. 45(1): 28-35 (in Chinese)., articleTitle=Ultra-short-term wind speed prediction model for wind farms based on spatiotemporal neural network, refAbstract=null), Reference(id=1154048177813577752, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, doi=null, pmid=null, pmcid=null, year=2015, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[9], rfOrder=13, authorNames=null, journalName=null, refType=null, unstructuredReference=Q/GDW 10588–2015. 风电功率预测功能规范国家电网有限公司. [S]. 北京: 中国电力出版社, 2015., articleTitle=null, refAbstract=null), Reference(id=1154048177868103705, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, doi=null, pmid=null, pmcid=null, year=2015, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[9], rfOrder=14, authorNames=null, journalName=null, refType=null, unstructuredReference=Q/GDW 10588-2015. Function specification of wind power forecastingState Grid Corporation of China. [S]. Beijing: China Electric Power Press, 2015. (in Chinese)., articleTitle=null, refAbstract=null), Reference(id=1154048177922629658, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, doi=null, pmid=null, pmcid=null, year=2018, volume=10, issue=5, pageStart=null, pageEnd=null, url=10. 1063/1.5042795, language=null, rfNumber=[10], rfOrder=15, authorNames=Yang Mao, Chen Xinxin, Huang Binyang, journalName=Journal of Renewable and Sustainable Energy, refType=null, unstructuredReference=Yang Mao, Chen Xinxin, Huang Binyang. Ultra-short-term multi-step wind power prediction based on fractal scaling factor transformation[J]. Journal of Renewable and Sustainable Energy, 2018. 10(5):, articleTitle=Ultra-short-term multi-step wind power prediction based on fractal scaling factor transformation, refAbstract=null), Reference(id=1154048177981349915, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, doi=null, pmid=null, pmcid=null, year=2018, volume=33, issue=3, pageStart=3276, pageEnd=3284, url=null, language=null, rfNumber=[11], rfOrder=16, authorNames=Yan Jie, Zhang Hao, Liu Yongqian, journalName=IEEE Transactions on Power Systems, refType=null, unstructuredReference=Yan Jie, Zhang Hao, Liu Yongqian, et al. Forecasting the high penetration of wind power on multiple scales using multi-to-multi mapping[J]. IEEE Transactions on Power Systems, 2018. 33(3): 3276-3284., articleTitle=Forecasting the high penetration of wind power on multiple scales using multi-to-multi mapping, refAbstract=null), Reference(id=1154048178031681564, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, doi=null, pmid=null, pmcid=null, year=2015, volume=35, issue=18, pageStart=4581, pageEnd=4590, url=null, language=null, rfNumber=[12], rfOrder=17, authorNames=李湃, 管晓宏, 吴江, journalName=中国电机工程学报, refType=null, unstructuredReference=李湃, 管晓宏, 吴江. 基于大气动力模型的多风电场出力场景生成方法[J]. 中国电机工程学报, 2015. 35(18): 4581-4590., articleTitle=基于大气动力模型的多风电场出力场景生成方法, refAbstract=null), Reference(id=1154048178077818909, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, doi=null, pmid=null, pmcid=null, year=2015, volume=35, issue=18, pageStart=4581, pageEnd=4590, url=null, language=null, rfNumber=[12], rfOrder=18, authorNames=Li Pai, Guan Xiaohong, Wu Jiang, journalName=Proceedings of the CSEE, refType=null, unstructuredReference=Li Pai, Guan Xiaohong, Wu Jiang. Wind power scenario generation for multiple wind farms based on atmospheric dynamic model[J]. Proceedings of the CSEE, 2015. 35(18): 4581-4590 (in Chinese)., articleTitle=Wind power scenario generation for multiple wind farms based on atmospheric dynamic model, refAbstract=null), Reference(id=1154048178136539166, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, doi=null, pmid=null, pmcid=null, year=2021, volume=null, issue=null, pageStart=713, pageEnd=718, url=null, language=null, rfNumber=[13], rfOrder=19, authorNames=null, journalName=2021 3rd Asia Energy and Electrical Engineering Symposium (AEEES). Chengdu, China, refType=null, unstructuredReference=Regional wind power forecasting based on hierarchical clustering and upscaling method[C]// 2021 3rd Asia Energy and Electrical Engineering Symposium (AEEES). Chengdu, China, 2021: 713-718., articleTitle=Regional wind power forecasting based on hierarchical clustering and upscaling method, refAbstract=null), Reference(id=1154048178186870815, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, doi=null, pmid=null, pmcid=null, year=2020, volume=254, issue=null, pageStart=119993, pageEnd=null, url=null, language=null, rfNumber=[14], rfOrder=20, authorNames=Lu Peng, Ye Lin, Zhong Wuzhi, journalName=Journal of Cleaner Production, refType=null, unstructuredReference=Lu Peng, Ye Lin, Zhong Wuzhi, et al. A novel spatio-temporal wind power forecasting framework based on multi-output support vector machine and optimization strategy[J]. Journal of Cleaner Production, 2020. 254: 119993., articleTitle=A novel spatio-temporal wind power forecasting framework based on multi-output support vector machine and optimization strategy, refAbstract=null), Reference(id=1154048178233008160, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, doi=null, pmid=null, pmcid=null, year=2017, volume=41, issue=10, pageStart=161, pageEnd=169, url=null, language=null, rfNumber=[15], rfOrder=21, authorNames=薛禹胜, 陈宁, 王树民, journalName=电力系统自动化, refType=null, unstructuredReference=薛禹胜, 陈宁, 王树民, 等. 关于利用空间相关性预测风速的评述[J]. 电力系统自动化, 2017. 41(10): 161-169., articleTitle=关于利用空间相关性预测风速的评述, refAbstract=null), Reference(id=1154048178283339809, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, doi=null, pmid=null, pmcid=null, year=2017, volume=41, issue=10, pageStart=161, pageEnd=169, url=null, language=null, rfNumber=[15], rfOrder=22, authorNames=Xue Yusheng, Chen Ning, Wang Shumin, journalName=Automation of Electric Power Systems, refType=null, unstructuredReference=Xue Yusheng, Chen Ning, Wang Shumin, et al. Review on wind speed prediction based on spatial correlation[J]. Automation of Electric Power Systems, 2017. 41(10): 161-169 (in Chinese)., articleTitle=Review on wind speed prediction based on spatial correlation, refAbstract=null), Reference(id=1154048178342060066, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, doi=null, pmid=null, pmcid=null, year=2014, volume=38, issue=14, pageStart=126, pageEnd=135, url=null, language=null, rfNumber=[16], rfOrder=23, authorNames=叶林, 赵永宁, journalName=电力系统自动化, refType=null, unstructuredReference=叶林, 赵永宁. 基于空间相关性的风电功率预测研究综述[J]. 电力系统自动化, 2014. 38(14): 126-135., articleTitle=基于空间相关性的风电功率预测研究综述, refAbstract=null), Reference(id=1154048178404974627, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, doi=null, pmid=null, pmcid=null, year=2014, volume=38, issue=14, pageStart=126, pageEnd=135, url=null, language=null, rfNumber=[16], rfOrder=24, authorNames=Ye Lin, Zhao Yongning, journalName=Automation of Electric Power Systems, refType=null, unstructuredReference=Ye Lin, Zhao Yongning. A review on wind power prediction based on spatial correlation approach[J]. Automation of Electric Power Systems, 2014. 38(14): 126-135 (in Chinese)., articleTitle=A review on wind power prediction based on spatial correlation approach, refAbstract=null), Reference(id=1154048178467889188, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[17], rfOrder=25, authorNames=Mohandes M A, Rehman S, Rahman S M, journalName=International Journal of Energy Research, refType=null, unstructuredReference=Mohandes M A, Rehman S, Rahman S M. Spatial estimati-on of wind speed[J]. International Journal of Energy Research,, articleTitle=Spatial estimati-on of wind speed, refAbstract=null), Reference(id=1154048178522415141, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, doi=null, pmid=null, pmcid=null, year=2015, volume=39, issue=8, pageStart=5, pageEnd=11, url=null, language=null, rfNumber=[18], rfOrder=26, authorNames=郁琛, 薛禹胜, 文福拴, journalName=电力系统自动化, refType=null, unstructuredReference=郁琛, 薛禹胜, 文福拴, 等. 按时序特征优化模型后在线选配的超短期风电预测[J]. 电力系统自动化, 2015. 39(8): 5-11., articleTitle=按时序特征优化模型后在线选配的超短期风电预测, refAbstract=null), Reference(id=1154048178572746790, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, doi=null, pmid=null, pmcid=null, year=2015, volume=39, issue=8, pageStart=5, pageEnd=11, url=null, language=null, rfNumber=[18], rfOrder=27, authorNames=Yu Chen, Xue Yusheng, Wen Fushuan, journalName=Automation of Electric Power Systems, refType=null, unstructuredReference=Yu Chen, Xue Yusheng, Wen Fushuan, et al. An ultra-short-term wind power prediction method using "offline classification and optimization, online model matching" based on time series features[J]. Automation of Electric Power Systems, 2015. 39(8): 5-11 (in Chinese)., articleTitle=An ultra-short-term wind power prediction method using "offline classification and optimization, online model matching" based on time series features, refAbstract=null), Reference(id=1154048178627272743, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, doi=null, pmid=null, pmcid=null, year=2011, volume=36, issue=8, pageStart=2078, pageEnd=2086, url=null, language=null, rfNumber=[19], rfOrder=28, authorNames=Adaramola M S, Krogstad P Å, journalName=Renewable Energy, refType=null, unstructuredReference=Adaramola M S, Krogstad P Å. Experimental investigation of wake effects on wind turbine performance[J]. Renewable Energy, 2011. 36(8): 2078-2086., articleTitle=Experimental investigation of wake effects on wind turbine performance, refAbstract=null), Reference(id=1154048178673410088, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, doi=null, pmid=null, pmcid=null, year=2021, volume=45, issue=22, pageStart=65, pageEnd=73, url=null, language=null, rfNumber=[20], rfOrder=29, authorNames=王晨, 寇鹏, 王若谷, journalName=电力系统自动化, refType=null, unstructuredReference=王晨, 寇鹏, 王若谷, 等. 利用多空间尺度下时空相关性的点云分布多风机风速预测[J]. 电力系统自动化, 2021. 45(22): 65-73., articleTitle=利用多空间尺度下时空相关性的点云分布多风机风速预测, refAbstract=null), Reference(id=1154048178719547433, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, doi=null, pmid=null, pmcid=null, year=2021, volume=45, issue=22, pageStart=65, pageEnd=73, url=null, language=null, rfNumber=[20], rfOrder=30, authorNames=Wang Chen, Kou Peng, Wang Ruogu, journalName=Automation of Electric Power Systems, refType=null, unstructuredReference=Wang Chen, Kou Peng, Wang Ruogu, et al. Wind speed forecasting for multiple wind turbines with point cloud distribution using spatio-temporal correlation on multiple spatial scale[J]. Automation of Electric Power Systems, 2021. 45(22): 65-73 (in Chinese)., articleTitle=Wind speed forecasting for multiple wind turbines with point cloud distribution using spatio-temporal correlation on multiple spatial scale, refAbstract=null), Reference(id=1154048178782461994, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, doi=null, pmid=null, pmcid=null, year=2021, volume=233, issue=null, pageStart=113917, pageEnd=null, url=null, language=null, rfNumber=[21], rfOrder=31, authorNames=Liu Mingde, Ding Lin, Bai Yulong, journalName=Energy Conversion and Management, refType=null, unstructuredReference=Liu Mingde, Ding Lin, Bai Yulong. Application of hybrid model based on empirical mode decomposition, novel recurrent neural networks and the ARIMA to wind speed prediction[J]. Energy Conversion and Management, 2021. 233: 113917., articleTitle=Application of hybrid model based on empirical mode decomposition, novel recurrent neural networks and the ARIMA to wind speed prediction, refAbstract=null), Reference(id=1154048178836987947, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, doi=null, pmid=null, pmcid=null, year=2019, volume=39, issue=7, pageStart=2093, pageEnd=2106, url=null, language=null, rfNumber=[22], rfOrder=32, authorNames=陈金富, 朱乔木, 石东源, journalName=中国电机工程学报, refType=null, unstructuredReference=陈金富, 朱乔木, 石东源, 等. 利用时空相关性的多位置多步风速预测模型[J]. 中国电机工程学报, 2019. 39(7): 2093-2106., articleTitle=利用时空相关性的多位置多步风速预测模型, refAbstract=null), Reference(id=1154048178891513900, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, doi=null, pmid=null, pmcid=null, year=2019, volume=39, issue=7, pageStart=2093, pageEnd=2106, url=null, language=null, rfNumber=[22], rfOrder=33, authorNames=Chen Jinfu, Zhu Qiaomu, Shi Dongyuan, journalName=Proceedings of the CSEE, refType=null, unstructuredReference=Chen Jinfu, Zhu Qiaomu, Shi Dongyuan, et al. A multi-step wind speed prediction model for multiple sites leveraging spatio-temporal correlation[J]. Proceedings of the CSEE, 2019. 39(7): 2093-2106 (in Chinese)., articleTitle=A multi-step wind speed prediction model for multiple sites leveraging spatio-temporal correlation, refAbstract=null), Reference(id=1154048178941845549, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, doi=null, pmid=null, pmcid=null, year=2021, volume=175, issue=null, pageStart=718, pageEnd=730, url=null, language=null, rfNumber=[23], rfOrder=34, authorNames=Zheng Ling, Zhou Bin, Or S W, journalName=Renewable Energy, refType=null, unstructuredReference=Zheng Ling, Zhou Bin, Or S W, et al. Spatio-temporal wind speed prediction of multiple wind farms using capsule net work[J]. Renewable Energy, 2021. 175: 718-730., articleTitle=Spatio-temporal wind speed prediction of multiple wind farms using capsule net work, refAbstract=null), Reference(id=1154048178992177198, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, doi=null, pmid=null, pmcid=null, year=2020, volume=18, issue=2, pageStart=24, pageEnd=31, url=null, language=null, rfNumber=[24], rfOrder=35, authorNames=焦小敏, 耿华, 马少康, journalName=电源学报, refType=null, unstructuredReference=焦小敏, 耿华, 马少康, 等. 基于数据驱动的多风电机组协同控制方法[J]. 电源学报, 2020. 18(2): 24-31., articleTitle=基于数据驱动的多风电机组协同控制方法, refAbstract=null), Reference(id=1154048179076063279, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, doi=null, pmid=null, pmcid=null, year=2020, volume=18, issue=2, pageStart=24, pageEnd=31, url=null, language=null, rfNumber=[24], rfOrder=36, authorNames=Jiao Xiaomin, Geng Hua, Ma Shaokang, journalName=Journal of Power Supply, refType=null, unstructuredReference=Jiao Xiaomin, Geng Hua, Ma Shaokang, et al. Data-driven cooperative control method for multiple wind turbines[J]. Journal of Power Supply, 2020. 18(2): 24-31 (in Chinese)., articleTitle=Data-driven cooperative control method for multiple wind turbines, refAbstract=null), Reference(id=1154048179130589232, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, doi=null, pmid=null, pmcid=null, year=2016, volume=null, issue=null, pageStart=null, pageEnd=null, url=arXiv: 1609.02907, language=null, rfNumber=[25], rfOrder=37, authorNames=Kipf T N, Welling M, journalName=ArXiv e-Prints, refType=null, unstructuredReference=Kipf T N, Welling M. Semi-supervised classification with graph convolutional networks[J]. ArXiv e-Prints, 2016., articleTitle=Semi-supervised classification with graph convolutional networks, refAbstract=null)], funds=[Fund(id=1154048176915996681, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, awardId=51607111, language=EN, fundingSource=National Natural Science Foundation of China(51607111), fundOrder=null, country=null), Fund(id=1154048176974716938, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, awardId=51607111, language=CN, fundingSource=国家自然科学基金资助项目(51607111), fundOrder=null, country=null)], companyList=[AuthorCompany(id=1154048171404682164, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, xref=1, ext=[AuthorCompanyExt(id=1154048171408876469, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, companyId=1154048171404682164, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 College of Automation Engineering Shanghai University of Electrical Power Shanghai 200090 China), AuthorCompanyExt(id=1154048171413070774, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, companyId=1154048171404682164, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 上海电力大学 自动化工程学院 上海 200090)]), AuthorCompany(id=1154048171496956855, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, xref=2, ext=[AuthorCompanyExt(id=1154048171505345464, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, companyId=1154048171496956855, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 Rundian Energy Science and Technology Co., Ltd Zhengzhou 450052 China), AuthorCompanyExt(id=1154048171509539769, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, companyId=1154048171496956855, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 润电能源科学技术有限公司 郑州 450052)])], figs=[ArticleFig(id=1154048175590597620, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, language=EN, label=Fig. 1, caption=Distribution in spatial structure of one wind farm, figureFileSmall=T4x6dCVnU+CMmrSDy4nvdw==, figureFileBig=MaQ1NlM8q7WFkbOrK8rWvw==, tableContent=null), ArticleFig(id=1154048175645123573, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, language=CN, label=图1, caption=某风场空间结构分布, figureFileSmall=T4x6dCVnU+CMmrSDy4nvdw==, figureFileBig=MaQ1NlM8q7WFkbOrK8rWvw==, tableContent=null), ArticleFig(id=1154048175699649526, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, language=EN, label=Fig. 2, caption=Process of distribution diagram to topology diagram of weights, figureFileSmall=WyqOoDrlJzvXNkWSM3XZsg==, figureFileBig=wFIeANcoTQqPHUdwCjeKRg==, tableContent=null), ArticleFig(id=1154048175754175479, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, language=CN, label=图2, caption=原风场分布至点线权值拓扑, figureFileSmall=WyqOoDrlJzvXNkWSM3XZsg==, figureFileBig=wFIeANcoTQqPHUdwCjeKRg==, tableContent=null), ArticleFig(id=1154048175804507128, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, language=EN, label=Fig. 3, caption=Aggregated spatio-temporal GCN model, figureFileSmall=GoWZcUtIKdibmWCcKm1ldA==, figureFileBig=uXz7M5N4HvkJLykQGXow7Q==, tableContent=null), ArticleFig(id=1154048175863227385, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, language=CN, label=图3, caption=聚合时空图卷积网络模型, figureFileSmall=GoWZcUtIKdibmWCcKm1ldA==, figureFileBig=uXz7M5N4HvkJLykQGXow7Q==, tableContent=null), ArticleFig(id=1154048175930336250, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, language=EN, label=Fig. 4, caption=Structure of temporal convolution, figureFileSmall=DimHgqO0i09gCoOvij+mSg==, figureFileBig=UlBmFwz2IfXUQLwVPA37nQ==, tableContent=null), ArticleFig(id=1154048175980667899, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, language=CN, label=图4, caption=时态卷积结构, figureFileSmall=DimHgqO0i09gCoOvij+mSg==, figureFileBig=UlBmFwz2IfXUQLwVPA37nQ==, tableContent=null), ArticleFig(id=1154048176026805244, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, language=EN, label=Fig. 5, caption=Process of spatio-temporal feature aggregation, figureFileSmall=n/S+wrEpkPdOw2yXv/z4eA==, figureFileBig=X0m0EIh6H9aBqFn6pDO0tQ==, tableContent=null), ArticleFig(id=1154048176072942589, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, language=CN, label=图5, caption=时空特征聚合流程, figureFileSmall=n/S+wrEpkPdOw2yXv/z4eA==, figureFileBig=X0m0EIh6H9aBqFn6pDO0tQ==, tableContent=null), ArticleFig(id=1154048176119079934, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, language=EN, label=Fig. 6, caption=Prediction error in ${0.5}- {4.0}\mathrm{\;h}$, figureFileSmall=utektXecPZkxoaFHSqJUWA==, figureFileBig=VWcEVUw+cyvBrz1K1o2Lmw==, tableContent=null), ArticleFig(id=1154048176173605887, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, language=CN, label=图6, caption=${0.5}\sim {4.0}\mathrm{\;h}$ 预测误差, figureFileSmall=utektXecPZkxoaFHSqJUWA==, figureFileBig=VWcEVUw+cyvBrz1K1o2Lmw==, tableContent=null), ArticleFig(id=1154048176257491968, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, language=EN, label=Fig. 7, caption=Training loss curves of STGCN-AGG-GLU, figureFileSmall=RSXdHhNfQh4hzzg6lZuqTg==, figureFileBig=SHO24fo5YxpRe+B3/QLKOA==, tableContent=null), ArticleFig(id=1154048176299433984, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, language=CN, label=图7, caption=STGCN-AGG-GLU 训练损失曲线, figureFileSmall=RSXdHhNfQh4hzzg6lZuqTg==, figureFileBig=SHO24fo5YxpRe+B3/QLKOA==, tableContent=null), ArticleFig(id=1154048176345571329, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, language=EN, label=Tab. 1, caption=Values of SCCF between wind speed series at reference farm and those at other farms, figureFileSmall=null, figureFileBig=null, tableContent=
参考风场 ${r}_{xy}\left({k = 0}\right)$ ${r}_{xy}\left({k = 1}\right)$ ${r}_{xy}\left({k = 2}\right)$ ${r}_{xy}\left({k = 3}\right)$
1 号风场 0.603 0.579 0.516 0.497
3 号风场 0.738 0.670 0.696 0.597
4 号风场 0.657 0.581 0.484 0.394
5 号风场 0.508 0.536 0.407 0.319
), ArticleFig(id=1154048176383320066, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, language=CN, label=表1, caption=参考风场与其他风场风速序列的 SCCF, figureFileSmall=null, figureFileBig=null, tableContent=
参考风场 ${r}_{xy}\left({k = 0}\right)$ ${r}_{xy}\left({k = 1}\right)$ ${r}_{xy}\left({k = 2}\right)$ ${r}_{xy}\left({k = 3}\right)$
1 号风场 0.603 0.579 0.516 0.497
3 号风场 0.738 0.670 0.696 0.597
4 号风场 0.657 0.581 0.484 0.394
5 号风场 0.508 0.536 0.407 0.319
), ArticleFig(id=1154048176433651715, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, language=EN, label=Tab. 22, caption=Values of SCCF between wind speed series at No. 2 farm and those at other farms, figureFileSmall=null, figureFileBig=null, tableContent=
日期 风场 1 风场 3 风场 4 风场 5
2020 年 6 月 21 日 0.603 0.738 0.657 0.508
2020 年 6 月 22 日 0.856 0.839 0.705 0.796
2020 年 6 月 23 日 0.925 0.920 0.899 0.932
2020 年 6 月 24 日 0.835 0.663 0.855 0.797
2020 年 6 月 25 日 0.849 0.748 0.651 0.873
2020 年 6 月 26 日 0.805 0.702 0.514 0.719
2020 年 6 月 27 日 0.918 0.724 0.771 0.869
), ArticleFig(id=1154048176496566276, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, language=CN, label=表22, caption=号风场与其他风场风速序列的 SCCF, figureFileSmall=null, figureFileBig=null, tableContent=
日期 风场 1 风场 3 风场 4 风场 5
2020 年 6 月 21 日 0.603 0.738 0.657 0.508
2020 年 6 月 22 日 0.856 0.839 0.705 0.796
2020 年 6 月 23 日 0.925 0.920 0.899 0.932
2020 年 6 月 24 日 0.835 0.663 0.855 0.797
2020 年 6 月 25 日 0.849 0.748 0.651 0.873
2020 年 6 月 26 日 0.805 0.702 0.514 0.719
2020 年 6 月 27 日 0.918 0.724 0.771 0.869
), ArticleFig(id=1154048176546897925, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, language=EN, label=Tab. 3, caption=Comparison among different prediction methods for $1\mathrm{\;h}$ and $3\mathrm{\;h}$, respectively (June 23), figureFileSmall=null, figureFileBig=null, tableContent=
时间 预测方法 MAE/(m$\cdot {\mathrm{s}}^{-1}$ ) RMSE/(m$\cdot {\mathrm{s}}^{-1}$ )
$1\mathrm{\;h}$ STGCN 0.864 1.145
STGCN-GLU 0.854 1.083
STGCN-AGG-GLU 0.803 1.005
$3\mathrm{\;h}$ STGCN 1.196 1.473
STGCN-GLU 1.115 1.397
STGCN-AGG-GLU 1.033 1.289
), ArticleFig(id=1154048176601423878, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, language=CN, label=表3, caption=$\;1\mathrm{\;h}$ 、 $3\mathrm{\;h}$ 预测下不同预测方法比较 ( 6 月 23 日), figureFileSmall=null, figureFileBig=null, tableContent=
时间 预测方法 MAE/(m$\cdot {\mathrm{s}}^{-1}$ ) RMSE/(m$\cdot {\mathrm{s}}^{-1}$ )
$1\mathrm{\;h}$ STGCN 0.864 1.145
STGCN-GLU 0.854 1.083
STGCN-AGG-GLU 0.803 1.005
$3\mathrm{\;h}$ STGCN 1.196 1.473
STGCN-GLU 1.115 1.397
STGCN-AGG-GLU 1.033 1.289
), ArticleFig(id=1154048176647561223, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, language=EN, label=Tab. 4, caption=Comparison among different prediction methods for $1\mathrm{\;h}$ and $3\mathrm{\;h}$, respectively (June 21), figureFileSmall=null, figureFileBig=null, tableContent=
时间 预测方法 MAE/(m$\cdot {\mathrm{s}}^{-1}$ ) RMSE/(m$\cdot {\mathrm{s}}^{-1}$ )
$1\mathrm{\;h}$ STGCN 0.928 1.258
STGCN-GLU 0.920 1.259
STGCN-AGG-GLU 0.906 1.214
$3\mathrm{\;h}$ STGCN 1.178 1.541
STGCN-GLU 1.144 1.455
STGCN-AGG-GLU 1.084 1.407
), ArticleFig(id=1154048176702087176, tenantId=1146029695717560320, journalId=1146031654075715584, articleId=1154037274888102695, language=CN, label=表4, caption=$\mathrm{\;{1h}}$ 、 $3\mathrm{\;h}$ 预测下不同预测方法比较 ( 6 月 21 日), figureFileSmall=null, figureFileBig=null, tableContent=
时间 预测方法 MAE/(m$\cdot {\mathrm{s}}^{-1}$ ) RMSE/(m$\cdot {\mathrm{s}}^{-1}$ )
$1\mathrm{\;h}$ STGCN 0.928 1.258
STGCN-GLU 0.920 1.259
STGCN-AGG-GLU 0.906 1.214
$3\mathrm{\;h}$ STGCN 1.178 1.541
STGCN-GLU 1.144 1.455
STGCN-AGG-GLU 1.084 1.407
)], 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.2095-2805.2024.4.133, detailUrlEn=https://castjournals.cast.org.cn/joweb/dyxb/EN/10.13234/j.issn.2095-2805.2024.4.133, pdfUrlCn=https://castjournals.cast.org.cn/joweb/dyxb/CN/PDF/10.13234/j.issn.2095-2805.2024.4.133, pdfUrlEn=https://castjournals.cast.org.cn/joweb/dyxb/EN/PDF/10.13234/j.issn.2095-2805.2024.4.133, aliStartDate=null, aliEndDate=null, collectionFlag=false, citedCount=null, citedUrl=null, reference=null)
收藏切换
基于聚合时空图卷积网络的多风场超短期风速预测
收藏切换
PDF下载
徐辰晓 1 , 崔承刚 1 , 郭为民 2 , 杨宁 1 , 刘备 2 , 孟青叶 2
电源学报 | 新能源系统 2024,22(4): 133-142
收起
收藏切换
电源学报 | 新能源系统 2024, 22(4): 133-142
基于聚合时空图卷积网络的多风场超短期风速预测
全屏
徐辰晓1 , 崔承刚1 , 郭为民2 , 杨宁1 , 刘备2 , 孟青叶2
作者信息
  • 1 上海电力大学 自动化工程学院 上海 200090
  • 2 润电能源科学技术有限公司 郑州 450052
  • 徐辰晓(1996-),男,硕士研究生。研究方向:人工智能在风力发电预测中的应用。E-mail: 694190047@qq.com。

    崔承刚(1981-),男,通信作者,博士,副教授。研究方向:深度学习与强化学习。E-mail: cgcui@shiep.edu.cn。

    郭为民(1972-),男,教授级高级工程师。研究方向:火电厂智能规划建设。E-mail: GUOWEIMIN8@crpower.com.cn。

    杨宁(1976-),男,博士研究生,教授。研究方向:基于强化学习的功率优化。E-mail:yangning@shiep.edu.cn。

    刘备(1993-),男,硕士研究生,工程师。研究方向:超临界火电机组主蒸汽温度控制算法。E-mail : 377179519@qq.com。

    孟青叶(1991-),女,硕士研究生,工程师。研究方向:数据驱动的变速变桨距风力发电机组功率优化控制。E-mail: M_qingye@163.com。

Ultra-short-term Wind Speed Prediction for Multiple Wind Farms Based on Aggregated Spatio-temporal Graph Convolutional Networks
Chenxiao XU1 , Chenggang CUI1 , Weimin GUO2 , Ning YANG1 , Bei LIU2 , Qingye MENG2
Affiliations
  • 1 College of Automation Engineering Shanghai University of Electrical Power Shanghai 200090 China
  • 2 Rundian Energy Science and Technology Co., Ltd Zhengzhou 450052 China
出版时间: 2024-07-30 doi: 10.13234/j.issn.2095-2805.2024.4.133
文章导航
收藏切换

在一定环境内区域风电场呈不规则分布的条件下,传统卷积神经网络预测方法无法体现出各区域风场的分布状态和影响关系,难以实现对风速的准确预测。针对此问题,采用图卷积网络进行特征建模,并根据多风场的拓扑结构和各区域风场风速的互相关系数建立连通图和权重矩阵。其次,依赖风场风速的时间动态特征,采用改进并列式卷积结构获取同一风场下多时间段的风速序列相关性。再次,利用风场风速的空间相关性和延时效应,采用二阶聚合方法将不同区域内风速的时空特征聚合。最后,经某区域风场数据验证表明,在0~4h预测尺度下该方法在多风场超短期风速预测中具有提取时空特征并提升预测性能的效果。

风速预测  /  聚合时空图卷积网络  /  时空相关性

In a certain environment where regional wind farms distribute irregularly, the traditional convolutional neural network prediction method cannot reflect the distribution states or influence relationship of regional wind farms, and it is difficult to accurately predict the wind speed. First, to solve this problem, the technology of graph convolutional networks is used for feature modeling, and the connected graph and weight matrix are established according to the topology of multiple wind farms and the cross-correlation coefficient of wind speed in each region. Second, depending on the time dynamic characteristics of wind speed at wind farms, an improved parallel convolution structure is used to obtain the correlation between wind speed series in multiple time periods at the same wind farm. Third, based on the spatial correlation and delay effect of wind speed at wind farms, the spatio-temporal characteristics of wind speed in different regions are aggregated by using a second-order aggregation method. Finally, the verification of data from one regional wind farm shows that the proposed method can extract the spatio-temporal characteristics and improve the performance of ultra short-term wind speed prediction for multiple wind farms on 0-4 h prediction scale.

Wind speed prediction  /  aggregated spatio-temporal graph convolutional networks  /  spatio-temporal correlation
徐辰晓, 崔承刚, 郭为民, 杨宁, 刘备, 孟青叶. 基于聚合时空图卷积网络的多风场超短期风速预测. 电源学报, 2024 , 22 (4) : 133 -142 . DOI: 10.13234/j.issn.2095-2805.2024.4.133
Chenxiao XU, Chenggang CUI, Weimin GUO, Ning YANG, Bei LIU, Qingye MENG. Ultra-short-term Wind Speed Prediction for Multiple Wind Farms Based on Aggregated Spatio-temporal Graph Convolutional Networks[J]. Journal of Power Supply, 2024 , 22 (4) : 133 -142 . DOI: 10.13234/j.issn.2095-2805.2024.4.133
“十四五” 期间, 为加快推动高质量发展, 贯彻新发展理念, 助力 “双碳” 目标实现, 清洁能源迅速发展[1]。近几年来,我国风电装机容量迅速增长,并网规模不断扩大[2],风电能源的集中开发, 形成了集群式的分布[3]。但由于风能的间歇性、波动性导致并网回馈的电能会对电网的安全运行产生负面影响[4],若不能减少风力发电波动带来的负面影响,风力发电的发展将难以取得较大进步[5]。 精确的短期和超短期风电功率预测有助于缓解电力系统调峰、调频的压力, 并对电网安全运行和调度经济化具有重要的意义[6]。由于我国风场规模较大且分布呈不规则状, 准确的多风场风速预测也是最大程度优化电力系统控制策略的参照[7]
风速的超短期预测是对风电场未来$0 \sim 4\mathrm{\;h}$ 风速数值的预测[8-9]。目前为止,国内外研究人员对区域多风电场功率或风速的超短期预测做了相对丰富的研究。文献[10]通过划分区域风电场, 利用分形的拉伸变换理论建立局部区域与整体区域的仿射关系, 建立了基于分形尺度因子变换的超短期多步风电功率预测模型; 文献[11]提出了多对多映射网络和堆叠去噪自编码的方法对区域内多个风场的数值天气预报进行了修正; 文献[12]通过大气运动方程和风速降尺度方程建立了风电场群下高空大气与地表风速关系的随机动态系统, 并预测各风电场未来风速的联合分布;文献[13]提出层次聚类与升尺度的区域风电功率预测方法, 即利用贪婪算法搜索出最优的子区域风场数量, 提高了预测效果; 文献[14]提出集成卷积神经网络 CNN(convolutional neural network)与长短期记忆 LSTM(long short-term memory)网络的预测时空深度网络, 利用时空相关性研究了多站点风速预测问题。
上述研究为多风场功率的准确预测提供了有效方案。相关研究[15-16] 表明,在具有不同空间位置风场的风速数据时, 可以建立空间风电场之间的联系,从而改善预测的效果[17]。这些时空特征联系包括同一空间点下风速数值的时序相关、在某个特定位置上的风速值受到处于其他不同位置的风电机组[18] 和拓扑结构及环境条件的影响[19]。人工智能技术作为时空相关特征提取的有效方法之一, 其 CNN 和循环神经网络 RNN(recurrent neural network) 得到了广泛的应用[20]。文献[21]建立 3D-CNN,实现了风速序列时空动作特征的多步分析; 文献[22]采用CNN和双向门控BGRU(bidirectional gated recurrent unit) 网络构建深层时空网络DSTN(deep spatiotemporal network) 预测模型, 从CNN中提取空间特征并利用 BGRU 捕捉同一时间断面下空间特征的时空依赖关系;文献[23]提出基于胶囊网络(capsule network)的时空风速预测模型,将 CNN 提取风场风速的空间特征封装进胶囊元中, 再通过动态路由机制捕捉时态特征。然而, 多风场分布一般呈不规则状, CNN无法准确地表达非结构化分布的风场空间特征, 且在卷积运算中, 对不同区域内风速相互影响力表征能力不足。
为解决多风场风速预测时空特性提取问题, 本文提出聚合时空图卷积网络 GCN(graph convolutional networks)对多风场风速的空间依赖和时间动态进行建模。首先, 依据各个风电场在不同时滞下的互相关系数和空间分布特征建立带权值的图卷积网络, 挖掘出风速序列的空间特征; 其次, 采用并列式卷积网络提取出不同时间段下风速的时序特征并通过门控线性单元 GLU(gated linear unit)保留数据的非线性; 最后, 通过聚合方法将空间特征与时序特征聚合, 经全连接层输出预测结果, 实现了基于时空相关性挖掘下多风场${0.5}\sim {4.0}\mathrm{\;h}$ 的风速预测。
风场内数十台风机之间的相互影响和集群分布使整个风电场内的风速特性不同于单风机的风速特性称为集群效应。其中, 集群效应包括地形地貌、尾流效应和延时效应等[24]。受集群效应的影响, 不同位置下的风电机组风速存在一定时滞和空间关联。因此, 选用某种方法在不同的位置建立合适的风速空间相关性模型, 再加上各个位置上风速的历史数据, 可以推算出相邻位置或区域内某点的未知风速, 其表达式为
${v}_{i, t}= f\left({{\mathbf{w}}_{t - 1},{\mathbf{w}}_{t - 2},\cdots,{\mathbf{w}}_{t - n}}\right)$
式中:${v}_{i, t}$ 为节点$i$$t$ 时刻下的风速序列值;${\mathbf{w}}_{t - n}$$t - n$ 时刻下历史风速向量。
本文以某区域 5 个子风电场为例, 验证风速的时空相关性。通过采样交叉相关函数[22] SCCF(sample cross-correlation function)计算 2 个风场风速的相关性。当延时函数为$k$ 时,其对应的 SCCF 相关系数可表示为${C}_{xy}\left( k\right)$表1为在不同延时时间下,以 2 号风场作为参考风场与其他风场交叉相关函数的计算值,图1为某风场空间结构分布, 对应表1多风场分布情况和$k = 0$ 时各风场空间距离下的互相关系数大小。表1图1的研究结果表明: 在一定地理范围内, 空间距离相对较近的风电场之间风速的相关性较高[15]; 随着延时函数的增加,时间相关性总体逐渐下降。由此可知: 时空相关性体现在多风场的风速序列时, 受风机尾流效应的影响, 在同一风向下, 不同位置风场风速存在一定的空间相关性, 且距离越近, 机组间影响力越大, 空间相关性越强; 受风速时滞性影响, 在同一时间下, 距离越远的风电机组影响力越小, 其所在风场相关性越弱; 在一定延时下, 距离较远风场之间风速的相关性高于无延时下风速相关性。
为建立准确的多风场连通图, 需要将各区域子风场的实际距离和风速序列的相关性考虑在内,使 GCN 能够有效处理节点信息,并将各风场以节点方式表示, 风场之间的距离以带权值邻接线表示。以表1数据和图1所展示的风场邻接结构图为例(图2(a)), 建立如图2(b)所示的多风场连通拓扑图。
为准确表征邻接风场之间的相对影响程度, 本文采用欧式距离计算邻接节点的权重。权重表达式为
${\omega }_{i, j}= \exp \left\lbrack {-\operatorname{dist}\left({{p}_{i},{p}_{j}}\right)}\right\rbrack $
式中,节点${p}_{i}= \left({{p}_{xi},{p}_{yi}}\right)$$\operatorname{dist}\left({{p}_{i},{p}_{j}}\right)$ 分别为节点${p}_{i}$ 与节点${p}_{j}$ 之间的欧式距离。
假设某时刻为$t$,则$t$ 时刻时间序列的时空图表述为
${\mathbf{G}}_{t}= \left({{V}_{t},\xi, W}\right)$
${\mathbf{V}}_{t}= \left\{{{v}_{i, t}\mid i = 1,2,3,\cdots, n}\right\}$
$\mathbf{W}= \left\{{{\omega }_{ij}\mid i, j = 1,2,3,\cdots, n}\right\}$
式中:${\mathbf{V}}_{t}= \left\{{{v}_{i, t}\mid i = 1,2,3,\cdots, n}\right\}$${\mathbf{G}}_{t}$ 中所有节点${v}_{i, t}$ 观测值的集合;${\xi }_{i, j}$ 为风场$i$ 与风场$j$ 之间的实际邻接情况;$\mathbf{W}= \left\{{{\omega }_{ij}\mid i, j = 1,2,3,\cdots, n}\right\}$${\mathbf{G}}_{t}$ 的权重连接矩阵;${\omega }_{i, j}$ 为节点$i$ 与节点$j$ 之间的权重系数。
图3为聚合时空图卷积网络模型, 其中聚合时空图卷积网络主要由空间图卷积与时态门卷积模块组成: 时态卷积是对$M$ 个时间段的观测值${V}_{t - m},\cdots$,${\mathbf{V}}_{t - 2},{\mathbf{V}}_{t - 1}$ 进行时序卷积,提取出风速序列的时间依赖性; 而空间图卷积是构建图卷积核对输入数据进行空间卷积, 提取出风速序列的空间依赖性。聚合时空图卷积网络模型的优势在于: ①图卷积网络在较好地提取空间特征的同时, 可以保留原风场分布的空间拓扑结构;②时态卷积采用并列式 CNN, 能够保留不同时间段下的风速时序特征;③采用聚合算法将时态卷积结果与图卷积结果结合, 同时根据风场分布特征设置阶数可聚合非邻接风场的时空信息。
在之前的研究中,研究者们[20] 通过假设对象为规则阵列采用二维 CNN 方法直接提取空间特征, 但由于将整个风场分割成各个部分网络, 仅捕获了局部空间相关, 忽略了风场的连通性和全局性。各区域风场一般呈非结构化的图像数据, 本文采用 GCN 提取各节点信息的空间相关性, 通过归一化拉普拉斯矩阵(Laplacian matrix)的特征向量矩阵实现频域积克服非结构化的困难, 并对图卷积公式采用切比雪夫多项式近似(Chebyshev polynomials approximation)和一阶近似(1st-order approximation) 将$K$ 阶多项式截断为一阶多项式,进一步简化谱卷积运算[25]
图卷积为外来信号$\mathbf{X}\in {\mathbf{R}}^{n}$ 与滤波器${g}_{\theta }$ 在叶域的乘积, 其表达式为
${g}_{\theta }* \mathbf{X}= \mathbf{U}{g}_{\theta }{\mathbf{U}}^{\mathrm{T}}\mathbf{X}$
${\mathbf{L}}^{\mathrm{{sym}}}= {\mathbf{D}}^{-\frac{1}{2}}\mathbf{L}{\mathbf{D}}^{-\frac{1}{2}}= {\mathbf{I}}_{N}- {\mathbf{D}}^{-\frac{1}{2}}\mathbf{W}{\mathbf{D}}^{-\frac{1}{2}}= \mathbf{U}\mathbf{\Lambda }{\mathbf{U}}^{\mathrm{T}}$
$\mathbf{L}= \mathbf{D}- \mathbf{W}$
式中:${g}_{\theta }= \operatorname{diag}\left(\widetilde{\theta }\right),\widetilde{\theta }\in {\mathbf{R}}^{n};U \in {\mathbf{R}}^{n \times n}$${\mathbf{L}}^{\text{sym }}$ 的特征向量矩阵,可由${\mathbf{L}}^{\text{sym }}$ 特征分解得到;$\mathbf{D}\in {\mathbf{R}}^{n \times n}$ 为风场网络的度矩阵;$\mathbf{L}\in {\mathbf{R}}^{n \times n}$ 为简单拉普拉斯矩阵;${\mathbf{I}}_{N}$$N$ 维的单位矩阵;$\mathbf{W}\in {\mathbf{R}}^{n \times n}$ 为风场图网络的权重邻接矩阵;$\Lambda$ 为对应的特征值矩阵。
在式(6)矩阵相乘卷积运算中,矩阵计算和特征分解复杂度过大, 忽略了空间局部性, 故本文改进滤波器权重系数, 并利用切比雪夫不等式逼近卷积核, 其表达式为
${g}_{\theta }\left(\Lambda \right)= \mathop{\sum }\limits_{{k = 0}}^{{K - 1}}{\theta }_{k}{\Lambda }^{k}$
${g}_{\theta }\left(\mathbf{\Lambda }\right)* \mathbf{X}= \left({\mathop{\sum }\limits_{{k = 0}}^{{K - 1}}{\theta }_{k}\mathbf{U}{\mathbf{\Lambda }}^{k}{\mathbf{U}}^{\mathrm{T}}}\right)\mathbf{X}= \mathop{\sum }\limits_{{k = 0}}^{{K - 1}}{\theta }_{k}{\mathbf{L}}^{k}\mathbf{X}$
为降低复杂度,将${g}_{\theta }\left( A\right)$ 参数化为切比雪夫多项式, 其表达式为
${g}_{\theta }\left(\mathbf{L}\right)= \mathop{\sum }\limits_{{k = 0}}^{{K - 1}}{\theta }_{k}{T}_{k}\left(\widetilde{\mathbf{L}}\right)$
$\widetilde{\mathbf{L}}= \frac{2}{{\lambda }_{\max }}{\mathbf{L}}^{\text{sym }}- {\mathbf{I}}_{N}$
为减少运算成本,令$K = 1,{\lambda }_{\max }= 2$,同时进一步简化归一, 最终得到图卷积表达式为
${\mathbf{H}}^{l}= \sigma \left({{\widetilde{\mathbf{D}}}^{-\frac{1}{2}}\widetilde{\mathbf{W}}{\widetilde{\mathbf{D}}}^{-\frac{1}{2}}{\mathbf{H}}^{l - 1}{\theta }^{\prime }}\right)$
式中:$\sigma$ 为激活函数; 第$l$ 层卷积的输入为${\mathbf{H}}^{l - 1}\in$ ${\mathbf{R}}^{N \times D}$;${\theta }^{\prime }\in {\mathbf{R}}^{D \times D}$ 为待训练参数。经切比雪夫多项式近似后, 保留了一阶邻域内节点特征, 提高了空间特征的提取能力。
RNN 模型在时间序列分析中得到广泛应用, 但用于风速预测的递归网络存在迭代耗时、门控机制复杂和对动态变化响应缓慢的弊端。因此, 本文选用多通道 CNN 提取不同时段下各风场风速序列的时态相关性, 采用 GLU 作为时态卷积模块的激活函数, 降低了梯度的弥散化, 保留了非线性能力, 实现了计算速度的提升。该模块时态卷积结构如图4中虚线所示。
考虑到风速的延时效应, 为充分提取风速序列的时态相关, 本文改进卷积滤波器结构, 引入自相关系数, 其表达式为
${\lambda }_{k}= \operatorname{ACF}\left( k\right)= \frac{\operatorname{Cov}\left({{\mathbf{v}}_{t},{\mathbf{v}}_{t - k}}\right)}{\operatorname{Var}\left({\mathbf{v}}_{t}\right)} $
则风速序列与滤波器$f\left(\theta \right)$ 的卷积式为
$\mathbf{X}* f\left(\theta \right)= \mathop{\sum }\limits_{{k = 1}}^{K}{\lambda }_{k}{\theta }_{k}\mathbf{X}$
式中:$\theta \in {\mathbf{R}}^{K \times C}$ 为滤波器$f\left(\theta \right)$ 的参数;$X \in {\mathbf{R}}^{C \times N \times D}$ 为输入风速序列;$K$ 为卷积宽度;$C$ 为通道数量;$N$ 为节点数量。经各通道卷积计算后由 Concat 将特征拼接入 GLU 单元中, 其表达式为
${h}_{1}\left(\widehat{\mathbf{X}}\right)= \left({\widehat{\mathbf{X}}\cdot \mathbf{M}+ b}\right)\otimes \sigma \left({\widehat{\mathbf{X}}\cdot \mathbf{V}+ c}\right)$
式中:$\widehat{x}\in {\mathbf{R}}^{N \times m}$ 为 GLU 单元的外部输入;$\mathbf{M}\in {\mathbf{R}}^{k \times m \times n}$$b \in {\mathbf{R}}^{n}$ 为外部输入$\widehat{\mathbf{X}}$ 线性投影的学习参数;$V \in {\mathbf{R}}^{k \times m \times n}\text{、}C \in {\mathbf{R}}^{n}$ 为 GLU 激活门$\sigma$ 学习的参数;$m\text{、}n$ 分别为输入、输出的特征维度;$k$ 为卷积核的大小;$\sigma$ 采用 sigmoid 函数;$\otimes$ 为矩阵之间的乘积。
在该时态卷积模块结构中: ①根据输入风速数据的序列长度按时序分配给 3 个并列结构的卷积核通道, 得到不同历史时刻下风速数据的卷积结果。 ②采用 GLU 建立门控线性卷积代替常规的循环神经网络连接, 以此减轻梯度的传播力度, 防止梯度爆炸并保留了非线性能力。
与 CNN-RNN 为代表的时空网络不同,本文所提聚合时空图卷积网络通过时态卷积与图卷积的有机组合, 使图中各节点学习到其他节点信息, 并以此聚合时空特征。时空特征聚合流程如图5所示, 其中有
${X}_{t - M}= \operatorname{concat}\left({{V}_{t - M},{V}_{t - M + 1},{V}_{t - M + 2}}\right)$
该模型采用时态卷积-图卷积-时态卷积的连接方式将时空特征聚合, 使各个节点的风速特征同时考虑到空间相关性和时序相关性。由于距离相对较远、空间相关性较弱的风场之间难以建立联系, 采用二阶聚合方式扩大时空特征感受野, 加强时空特征的聚合能力。
本文将我国中原-平原地区某风场按地形和距离分布划分为 5 个区域子风场, 每个区域子风场内由若干个风机组成,通过采用各个子风场内${100}\mathrm{\;m}$ 高测风塔 2020 年 3 月初至 2020 年 6 月底所测得的实际风速作为测试算例进行验证, 基于 Facebook 开源深度学习框架 Pytorch 搭建开发环境, 采用 NVIDIA GeForce GTX 2080Ti 实现对风场风速的预测训练。其中, 按照地理信息采取不同的测风点风速,并且每次测风间隔为${0.5}\mathrm{\;h}$,以此分别对${0.5}\sim$ ${4.0}\mathrm{\;h}$ 预测结果进行评估,并采取数据集中前${70}\%$ 作为训练集, 后 30%作为验证集。
为了对比分析得到本文所提聚合时空图卷积网络(STGCN-AGG-GLU)的有效性, 采用在风速数值预测中具有代表性的模型作为基线模型, 其中包括:多层感知器 MLP(multi-layer perceptron)、支持向量回归 SVR(support vector regression)、LSTM 等预测方法和其他时空卷积网络预测模型。其他时空卷积网络基线模型的释义如下:
(1)CNN-GRU:采用 CNN 提取空间相关性, 门控循环单元 GRU(gated recurrent unit)提取时序相关性的时空卷积网络。
(2)STGCN: 采用原始 GCN 提取空间相关性, GRU 提取时序相关性且图网络未考虑权值系数的时空图卷积网络。
(3)STGCN-GLU: 采用改进 GCN 提取空间相关性,采用 2.2 节并列式卷积结构和 GLU 构成时态卷积模块且未经二阶节点聚合的时空卷积网络。
为了消除指标之间量纲和取值范围差异对预测结果的影响, 本文采用数据归一化处理, 即
${x}^{* }= \frac{x -{x}_{\min }}{{x}_{\max }- {x}_{\min }}$
式中:$x$ 为原始数据;${x}_{\min }$ 为原始数据最小值;${x}_{\max }$ 为原始数据最大值;${x}^{* }$ 为经式 (18) 归一化后的输入数据。
为了有效地对本文提出的预测方法进行评估, 参考文献[8]中选取的评价指标选取了平均绝对误差 MAE(mean absolute error)和均方根误差 RMSE (root mean squared error)作为预测误差的评价指标。
为验证本文所提方法的有效性, 本文采用 3.2 节中基线模型与时空图卷积网络进行对比。将未来${0.5}\sim {4.0}\mathrm{\;h}$ (间隔为${0.5}\mathrm{\;h}$ )时刻的风速值作为预测目标, 5 个风场风速的预测效果以 RMSE 为误差指标, 结果如图6所示, 同时给出了 STGCN-AGG-GLU 在${0.5}\sim {1.5}\mathrm{\;h}$ 预测下训练损失,如图7所示。
通过图6的对比和图7的分析可知:
(1)时间尺度:当时间尺度较低时,与 CNN-GRU 相比, STGCN 等网络预测效果不明显。当时间尺度变大时, CNN-GRU 等预测方法的预测误差变大, 而考虑时空相关性的图卷积网络预测误差相对较小。在$2 \sim 4\mathrm{\;h}$ 预测中,相比于 CNN-GRU, STGCN 预测精度提升了 15.4%~18.4%, STGCN-AGG-GLU 预测精度提升了 18.1%~22.2%,取得了更好的预测效果。
(2)空间相关性:在$4\mathrm{\;h}$ 预测中,与 CNN-GRU 相比,引入全局带权值分布的 STGCN 预测精度提升较大,使误差减少了${0.23}\mathrm{\;m}/\mathrm{s}\left(\mathrm{{RMSE}}\right)$, STGCN-GLU 和 STGCN-AGG-GLU 相比于 CNN-GRU 预测误差分别减少了${0.28}\mathrm{\;m}/\mathrm{s}\left(\mathrm{{RMSE}}\right)$${0.31}\mathrm{\;m}/\mathrm{s}\left(\mathrm{{RMSE}}\right)$, 说明在时间尺度较高的情况, 对空间相关性有效的提取可以提高预测效果。
(3)时态相关性:由图6可知,计及时态并列卷积模型的预测误差相对较小,同时通过对 STGCN 与 STGCN-GLU 在$2 \sim 4\mathrm{\;h}$ 长时间尺度下的预测效果对比可知, 对风速序列时态特性进行改进处理后,预测精度提高了${0.03}\sim {0.05}\mathrm{\;m}/\mathrm{s}\left(\mathrm{{RMSE}}\right)$,这说明对数据时间特征的提取能小幅提升预测效果。
(4)特征聚合:2~4 h 预测下, STGCN-AGG-GLU 相比于 STGCN、STGCN-GLU,由于采用了时态卷积-空间卷积-时态卷积的聚合方式, 预测精度最高提升了${0.1}\mathrm{\;m}/\mathrm{s}\left(\mathrm{{RMSE}}\right)\text{、}{0.08}\mathrm{\;m}/\mathrm{s}\left(\mathrm{{RMSE}}\right)$
(5)训练曲线:STGCN-AGG-GLU 在迭代次数 100 以内完成收敛,${30}\mathrm{\;{min}}$ 预测下收敛于 0.050 ;${60}\mathrm{\;{min}}$ 下收敛于${0.075};{90}\mathrm{\;{min}}$ 下收敛于 0.095。这证明了该预测方法在不同时间尺度下均具有良好的预测效果。
为进一步说明基于 STGCN-GLU-AGG 提取时空相关性的有效性, 本文以 2 号风场为基准, 采用验证集$7\mathrm{\;d}$ 内 2 号风场风速与各风场空间相关最高的一天与相关性最低的一天数据进行时空有效性分析。表2$7\mathrm{\;d}$ 内 2 号与其他风场在$k = 0$ 下互相关系数值, 可见, 在 2020 年 6 月 21 日 2 号风场与其他风场之间的空间相关性最弱, 在 2020 年 6 月 23 日 2 号风场与其他风场之间的空间相关性最强。 故本文对这 2 日未来$1\mathrm{\;h}$$3\mathrm{\;h}$ 风速预测进行分析, 结果分别如表3表4所示。
通过表3表4的分析可知:
( 1 )在对未来$1\mathrm{\;h}$ (即$k = 2$ 时)风速进行预测时, 6 月 23 日的风场互相关系数值为:${r}_{12}= {0.699}\text{、}{r}_{22}=$ ${0.683}\text{、}{r}_{32}= {0.739}\text{、}{r}_{42}= {0.729}\text{、}{r}_{52}= {0.742}$。此时 2 号风场与其他各风场的互相关系数较高, 时空相关性较强。由表3可知, 基于图卷积网络的预测方法均能取得较好的预测效果。其中,相较于 STGCN 预测精度(RMSE), 本文所提方法提升了 12.21%; 相较于 STGCN-GLU 预测精度(RMSE),本文所提方法提升了7.20%。而在 6 月 21 日时,本方所提方法比 STGCN、STGCN-GLU 的预测精度(RMSE)仅提升了3.6%。
(2)在对未来$3\mathrm{\;h}$ (即$k = 6$ 时)风速进行预测时, 6 月 23 日的风场互相关系数值为:${r}_{12}= {0.414}\text{、}{r}_{22}=$ ${0.373}\text{、}{r}_{32}= {0.176}\text{、}{r}_{42}= {0.390}\text{、}{r}_{52}= {0.354}$。此时 2 号风场与其他各风场的互相关系数较低, 时空相关性也远小于低尺度。由表3可知, 预测误差均出现增长, 本文所提方法误差增长相对较小, 相较于 STGCN 预测精度(RMSE)提升了 12.49%,相较于 STGCN-GLU 预测精度(RMSE)提升了 7.73%。而在 6 月 21 日时, 本方所提方法比 STGCN 预测精度(RMSE)仅提升了 8.69%,相较于 STGCN-GLU 预测精度(RMSE)仅提升了 3.29%。
综上可见, 在延时时间较低的情况下, 风速的时间相关性占主导因素, 各方法预测误差约保持在 0.93~0.97。在较大的延时下,风速自身的时间相关性降低, 空间相关性提高, 故 CNN-GRU、STGCN 等卷积神经网络预测精度高于其他考虑空间相关性预测方法,其中 STGCN-AGG-GLU 在未来$3 \sim 4\mathrm{\;h}$ 下预测效果最好。而经表3表4分析可知, STGCN-AGG-GLU 预测效果的提升更大程度上依赖于一定时间段下各风场之间的空间相关性, 空间相关性越大则该方法效果越明显。分析结果表明: STGCN-AGG-GLU 通过采用$\mathrm{{GCN}}$ 并设置契合多风场分布的权值连接矩阵可以有效提取各个风场之间的空间相关性; 通过改进并列卷积结构和 GLU 激活函数可以提取时序信息; 采用有效时空聚合方法可以将各个风场的时空特征融合。可见, 本文所提方法通过上述 3 个方面的分析均提高了风速预测的准确度。
针对当前人工智能方法对时空相关性挖掘能力有限的问题, 本文提出了聚合时空图卷积神经网络的预测方法。经算例分析表明, 本文所提方法能够充分提取不同时滞下各风场空间性, 提高了风速的超短期预测精度, 并得出以下结论。
(1)对于多区域的风场风速预测,采用基于 GCN 的预测方法, 可以更好地还原风场分布的拓扑结构和提取空间相关性。
(2)通过 STGCN-AGG-GLU 与 STGCN-GLU 的对比可以说明: 对多风场风速时空相关性的有效聚合可以进一步提高预测精度。
在本次研究中, 所提方法仅考虑风速因素对预测产生的影响, 下一步会将温度、气压、湿度等因素考虑在内并做相关的研究和分析。
  • 国家自然科学基金资助项目(51607111)
参考文献 引证文献
排序方式:
[1]
谭显东, 刘俊, 徐志成, 等. “双碳”目标下“十四五”电力供需形势[J]. 中国电力, 2021. 54(5): 1-6.
Tan Xiandong, Liu Jun, Xu Zhicheng, et al. Power supply and demand balance during the 14th five-year plan period under the goal of carbon emission peak and carbon neutrality[J]. Electric Power, 2021. 54(5): 1-6 (in Chinese).
[2]
本刊讯. 国家电网公司在京发布《国家电网公司促进新能源发展白皮书(2016)》[J]. 电力与能源, 2016. 37(2): 171.
State grid corporation of China. released the white paper on state grid corporation's promotion of new energy development (2016) in Beijing[J]. Power & Energy, 2016. 37(2): 171. (in Chinese).
[3]
Peng Xiaosheng, Cheng Kai, Lang Jianxun, et al. Short-term wind power prediction for wind farm clusters based on SFFS feature selection and BLSTM deep learning[J]. Energies, 2021. 14(7): 1894.
[4]
崔嘉, 杨俊友, 杨理践, 等. 基于改进 CFD与小波混合神经网络组合的风电场功率预测方法[J]. 电网技术, 2017. 41(1): 79-85.
Cui Jia, Yang Junyou, Yang Lijian, et al. New method of combined wind power forecasting based on improved CFD and wavelet-HNN model[J]. Power System Technology, 2017. 41(1): 79-85 (in Chinese).
[5]
Peng Xiaosheng, Wang Hongyu, Lang Jianxun, et al. EALSTM-QR: Interval wind-power prediction model based on numerical weather prediction and deep learning[J]. Energy, 2021. 220: 119692.
[6]
徐鹏超, 李琰, 赵艳雷. 基于自适应变异蝙蝠优化 BP 神经网络的短期风电功率预测[J]. 电测与仪表, 2021. 58(4): 121-127.
Xu Pengchao, Li Yan, Zhao Yanlei. Short-term wind power prediction based on BP neural network with adaptive mutation bat optimization algorithm[J]. Electrical Measurement & Instrumentation, 2021. 58(4): 121-127 (in Chinese).
[7]
Zhu Qiaomu, Chen Jinfu, Shi Dongyuan, et al. Learning temporal and spatial correlations jointly: A unified framework for wind speed prediction[J]. IEEE Transactions on Sustainable Energy, 2020. 11(1): 509-523.
[8]
凡航, 张雪敏, 梅生伟, 等. 基于时空神经网络的风电场超短期风速预测模型[J]. 电力系统自动化, 2021. 45(1): 28-35.
Fan Hang, Zhang Xuemin, Mei Shengwei, et al. Ultra-short-term wind speed prediction model for wind farms based on spatiotemporal neural network[J]. Automation of Electric Power Systems, 2021. 45(1): 28-35 (in Chinese).
[9]
Q/GDW 10588–2015. 风电功率预测功能规范国家电网有限公司. [S]. 北京: 中国电力出版社, 2015.
Q/GDW 10588-2015. Function specification of wind power forecastingState Grid Corporation of China. [S]. Beijing: China Electric Power Press, 2015. (in Chinese).
[10]
Yang Mao, Chen Xinxin, Huang Binyang. Ultra-short-term multi-step wind power prediction based on fractal scaling factor transformation[J]. Journal of Renewable and Sustainable Energy, 2018. 10(5): 10. 1063/1.5042795
[11]
Yan Jie, Zhang Hao, Liu Yongqian, et al. Forecasting the high penetration of wind power on multiple scales using multi-to-multi mapping[J]. IEEE Transactions on Power Systems, 2018. 33(3): 3276-3284.
[12]
李湃, 管晓宏, 吴江. 基于大气动力模型的多风电场出力场景生成方法[J]. 中国电机工程学报, 2015. 35(18): 4581-4590.
Li Pai, Guan Xiaohong, Wu Jiang. Wind power scenario generation for multiple wind farms based on atmospheric dynamic model[J]. Proceedings of the CSEE, 2015. 35(18): 4581-4590 (in Chinese).
[13]
Regional wind power forecasting based on hierarchical clustering and upscaling method[C]// 2021 3rd Asia Energy and Electrical Engineering Symposium (AEEES). Chengdu, China, 2021: 713-718.
[14]
Lu Peng, Ye Lin, Zhong Wuzhi, et al. A novel spatio-temporal wind power forecasting framework based on multi-output support vector machine and optimization strategy[J]. Journal of Cleaner Production, 2020. 254: 119993.
[15]
薛禹胜, 陈宁, 王树民, 等. 关于利用空间相关性预测风速的评述[J]. 电力系统自动化, 2017. 41(10): 161-169.
Xue Yusheng, Chen Ning, Wang Shumin, et al. Review on wind speed prediction based on spatial correlation[J]. Automation of Electric Power Systems, 2017. 41(10): 161-169 (in Chinese).
[16]
叶林, 赵永宁. 基于空间相关性的风电功率预测研究综述[J]. 电力系统自动化, 2014. 38(14): 126-135.
Ye Lin, Zhao Yongning. A review on wind power prediction based on spatial correlation approach[J]. Automation of Electric Power Systems, 2014. 38(14): 126-135 (in Chinese).
[17]
Mohandes M A, Rehman S, Rahman S M. Spatial estimati-on of wind speed[J]. International Journal of Energy Research,
[18]
郁琛, 薛禹胜, 文福拴, 等. 按时序特征优化模型后在线选配的超短期风电预测[J]. 电力系统自动化, 2015. 39(8): 5-11.
Yu Chen, Xue Yusheng, Wen Fushuan, et al. An ultra-short-term wind power prediction method using "offline classification and optimization, online model matching" based on time series features[J]. Automation of Electric Power Systems, 2015. 39(8): 5-11 (in Chinese).
[19]
Adaramola M S, Krogstad P Å. Experimental investigation of wake effects on wind turbine performance[J]. Renewable Energy, 2011. 36(8): 2078-2086.
[20]
王晨, 寇鹏, 王若谷, 等. 利用多空间尺度下时空相关性的点云分布多风机风速预测[J]. 电力系统自动化, 2021. 45(22): 65-73.
Wang Chen, Kou Peng, Wang Ruogu, et al. Wind speed forecasting for multiple wind turbines with point cloud distribution using spatio-temporal correlation on multiple spatial scale[J]. Automation of Electric Power Systems, 2021. 45(22): 65-73 (in Chinese).
[21]
Liu Mingde, Ding Lin, Bai Yulong. Application of hybrid model based on empirical mode decomposition, novel recurrent neural networks and the ARIMA to wind speed prediction[J]. Energy Conversion and Management, 2021. 233: 113917.
[22]
陈金富, 朱乔木, 石东源, 等. 利用时空相关性的多位置多步风速预测模型[J]. 中国电机工程学报, 2019. 39(7): 2093-2106.
Chen Jinfu, Zhu Qiaomu, Shi Dongyuan, et al. A multi-step wind speed prediction model for multiple sites leveraging spatio-temporal correlation[J]. Proceedings of the CSEE, 2019. 39(7): 2093-2106 (in Chinese).
[23]
Zheng Ling, Zhou Bin, Or S W, et al. Spatio-temporal wind speed prediction of multiple wind farms using capsule net work[J]. Renewable Energy, 2021. 175: 718-730.
[24]
焦小敏, 耿华, 马少康, 等. 基于数据驱动的多风电机组协同控制方法[J]. 电源学报, 2020. 18(2): 24-31.
Jiao Xiaomin, Geng Hua, Ma Shaokang, et al. Data-driven cooperative control method for multiple wind turbines[J]. Journal of Power Supply, 2020. 18(2): 24-31 (in Chinese).
[25]
Kipf T N, Welling M. Semi-supervised classification with graph convolutional networks[J]. ArXiv e-Prints, 2016. arXiv: 1609.02907
2024年第22卷第4期
PDF下载
295
115
引用本文
BibTeX
文章信息
doi: 10.13234/j.issn.2095-2805.2024.4.133
  • 接收时间:2021-12-27
  • 首发时间:2025-07-21
  • 出版时间:2024-07-30
补充材料
相关文章
文章信息
作者
出版历史
  • 收稿日期:2021-12-27
  • 修回日期:2022-03-24
  • 录用日期:2022-03-25
基金
National Natural Science Foundation of China(51607111)
国家自然科学基金资助项目(51607111)
作者信息
    1 上海电力大学 自动化工程学院 上海 200090
    2 润电能源科学技术有限公司 郑州 450052
参考文献
分享链接
https://castjournals.cast.org.cn/joweb/dyxb/CN/10.13234/j.issn.2095-2805.2024.4.133
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