Article(id=1149781959334916772, tenantId=1146029695717560320, journalId=1146123166801305609, issueId=1149781952959574654, articleNumber=null, orderNo=null, doi=10.12404/j.issn.1671-1815.2403403, pmid=null, cstr=null, oa=null, hot=null, price=null, onlineType=0, articleFormat=0, articleType=null, articleTypeStr=research-article, receivedDate=1715097600000, receivedDateStr=2024-05-08, revisedDate=1735228800000, revisedDateStr=2024-12-27, acceptedDate=null, acceptedDateStr=null, onlineDate=1752058981021, onlineDateStr=2025-07-09, pubDate=1743091200000, pubDateStr=2025-03-28, doiRegisterDate=null, doiRegisterDateStr=null, onlineIssueDate=1752058981021, onlineIssueDateStr=2025-07-09, onlineJustAcceptDate=null, onlineJustAcceptDateStr=null, onlineFirstDate=null, onlineFirstDateStr=null, sourceXml=null, magXml=null, createTime=1752058981021, creator=13701087609, updateTime=1752058981021, updator=13701087609, issue=Issue{id=1149781952959574654, tenantId=1146029695717560320, journalId=1146123166801305609, year='2025', volume='25', issue='9', pageStart='3529', pageEnd='3967', issueExtLink='null', onlineDate='null', pubDate='null', beforeIssueId=null, nextIssueId=null, price=null, status=1, issueComplete=1, articleOrder=1, issueType=-1, specialIssue=0, createTime=1752058979501, creator=13701087609, updateTime=1776333392421, updator=13701087609, preIssue=null, nextIssue=null, ext={EN=IssueExt(id=1251596220226027613, tenantId=1146029695717560320, journalId=1146123166801305609, issueId=1149781952959574654, language=EN, specialIssueTitle=, coverIllustrator=, specialIssueEditor=, specialIssueAbout=), CN=IssueExt(id=1251596220226027614, tenantId=1146029695717560320, journalId=1146123166801305609, issueId=1149781952959574654, language=CN, specialIssueTitle=, coverIllustrator=, specialIssueEditor=, specialIssueAbout=)}, issueFiles=null}, startPage=3769, endPage=3777, ext={EN=ArticleExt(id=1149781959829844649, articleId=1149781959334916772, tenantId=1146029695717560320, journalId=1146123166801305609, language=EN, title=Data-driven Classification Method for Typical Load Curves in Distribution Networks, columnId=1156262729162810294, journalTitle=Science Technology and Engineering, columnName=Papers·Automation and Computational Technology, runingTitle=null, highlight=null, articleAbstract=

With the continuous promotion of the “dual carbon” strategic goals and the construction of new power systems, traditional distribution networks are gradually transforming into information-based, digital, and intelligent new distribution systems. To accurately characterize and analyze the characteristics of different types of loads in the distribution network, and support efficient operation and control of the distribution network, a data-driven classification method for typical load curves in the distribution network was proposed. Firstly, based on load data, various classification scenarios of typical loads in the distribution network were analyzed, and performance evaluation indicators for classification scenarios including error rate, accuracy, and confusion matrix were proposed. On this basis, a data-driven load classification method for distribution networks was proposed, which converts 24 dimensional daily load vectors into image data and uses convolutional neural networks to identify load curve images, achieving accurate classification of distribution network load curves. Finally, the accuracy and effectiveness of the proposed method were verified by combining actual distribution network load data, and analyzed and compared with existing methods. The results indicate that the proposed method for classifying typical load curves in power distribution networks has better classification speed and accuracy.

, correspAuthors=Dong-li JIA, 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=Dong-li JIA, Shuai WANG, Ke-yan LIU, Shuo CHEN), CN=ArticleExt(id=1149781989932364015, articleId=1149781959334916772, tenantId=1146029695717560320, journalId=1146123166801305609, language=CN, title=基于数据驱动的配电网典型负荷曲线分类方法, columnId=1156262729783567290, journalTitle=科学技术与工程, columnName=论文·自动化技术、计算机技术, runingTitle=null, highlight=null, articleAbstract=

随着“双碳”战略目标和新型电力系统建设的不断推进,传统配电网逐渐向信息化、数字化和智能化的新型配电系统转变。为准确刻画并分析配电网中不同类型负荷特性,支撑配电网高效运行管控,提出了一种基于数据驱动的配电网典型负荷曲线分类方法。首先基于负荷数据,分析了配电网典型负荷的多种分类场景,并提出了包括错误率、精度和混淆矩阵等的分类场景性能评价指标;在此基础上,提出了一种基于数据驱动的配电网负荷分类方法,将24维日负荷向量转换成图片数据,并基于卷积神经网络识别负荷曲线图片,实现对配电网负荷曲线的精准分类;最后结合实际配电网负荷数据对所提方法的准确性与有效性进行了验证,并与已有方法进行了分析与对比。结果表明所提配电网典型负荷曲线分类方法具有更好的分类速度和分类精度。

, correspAuthors=贾东梨, authorNote=null, correspAuthorsNote=null, copyrightStatement=null, copyrightOwner=null, extLink=null, articleAbsUrl=null, sourceXml=3WXch2TWtvm4/R9LfFlWPg==, magXml=jFOAAllorejhQcuYETWcxw==, pdfUrl=null, pdf=67D27Cfri85AXe5yBmw3PQ==, pdfFileSize=7226928, pdfExtLink=null, richHtmlUrl=null, mobilePdfUrl=null, reviewReport=null, pdfFirstPage=null, abstractGraph=nzk+y8Yb+9FZZPi8k7y/5w==, abstractGraphContent=null, abstractVideo=null, citation=null, cebUrl=null, magXmlContent=7jXkVTX3Sfj1Wu5G4hrY2Q==, mapNumber=null, authorCompany=null, fund=null, authors=

贾东梨(1982—),女,汉族,山东烟台人,博士,教授级高级工程师。研究方向:配电网运行分析与控制。E-mail:

, authorsList=贾东梨, 王帅, 刘科研, 陈硕)}, authors=[Author(id=1251249356070928773, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, orderNo=0, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=jiadl@epri.sgcc.com.cn, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1251249356175786382, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, authorId=1251249356070928773, language=EN, stringName=Dong-li JIA, firstName=Dong-li, middleName=null, lastName=JIA, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1 China Electric Power Research Institute, Beijing 100192, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1251249356289032599, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, authorId=1251249356070928773, language=CN, stringName=贾东梨, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1 中国电力科学研究院有限公司, 北京 100192, bio={"content":"

贾东梨(1982—),女,汉族,山东烟台人,博士,教授级高级工程师。研究方向:配电网运行分析与控制。E-mail:

"}, bioImg=null, bioContent=

贾东梨(1982—),女,汉族,山东烟台人,博士,教授级高级工程师。研究方向:配电网运行分析与控制。E-mail:

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1251249355865407858, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, xref=1, ext=[AuthorCompanyExt(id=1251249355873796468, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, companyId=1251249355865407858, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 China Electric Power Research Institute, Beijing 100192, China), AuthorCompanyExt(id=1251249355882185076, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, companyId=1251249355865407858, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 中国电力科学研究院有限公司, 北京 100192)])]), Author(id=1251249356410667428, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, 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=1251249356632965556, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, authorId=1251249356410667428, language=EN, stringName=Shuai WANG, firstName=Shuai, middleName=null, lastName=WANG, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1 China Electric Power Research Institute, Beijing 100192, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1251249356729434562, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, authorId=1251249356410667428, language=CN, stringName=王帅, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1 中国电力科学研究院有限公司, 北京 100192, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1251249355865407858, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, xref=1, ext=[AuthorCompanyExt(id=1251249355873796468, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, companyId=1251249355865407858, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 China Electric Power Research Institute, Beijing 100192, China), AuthorCompanyExt(id=1251249355882185076, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, companyId=1251249355865407858, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 中国电力科学研究院有限公司, 北京 100192)])]), Author(id=1251249356855263693, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, 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=1251249357132087771, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, authorId=1251249356855263693, language=EN, stringName=Ke-yan LIU, firstName=Ke-yan, middleName=null, lastName=LIU, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1 China Electric Power Research Institute, Beijing 100192, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1251249357270499811, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, authorId=1251249356855263693, language=CN, stringName=刘科研, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1 中国电力科学研究院有限公司, 北京 100192, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1251249355865407858, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, xref=1, ext=[AuthorCompanyExt(id=1251249355873796468, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, companyId=1251249355865407858, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 China Electric Power Research Institute, Beijing 100192, China), AuthorCompanyExt(id=1251249355882185076, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, companyId=1251249355865407858, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 中国电力科学研究院有限公司, 北京 100192)])]), Author(id=1251249357429883376, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, 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=1251249357589266946, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, authorId=1251249357429883376, language=EN, stringName=Shuo CHEN, firstName=Shuo, middleName=null, lastName=CHEN, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2 School of Computer Science, Beijing University of Posts and Telecommunications, Beijing 100876, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1251249357727678996, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, authorId=1251249357429883376, language=CN, stringName=陈硕, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2 北京邮电大学计算机学院, 北京 100876, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1251249355957682554, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, xref=2, ext=[AuthorCompanyExt(id=1251249355970265469, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, companyId=1251249355957682554, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 School of Computer Science, Beijing University of Posts and Telecommunications, Beijing 100876, China), AuthorCompanyExt(id=1251249355974459773, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, companyId=1251249355957682554, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 北京邮电大学计算机学院, 北京 100876)])])], keywords=[Keyword(id=1251249358029668913, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, language=EN, orderNo=1, keyword=data-driven), Keyword(id=1251249358142915135, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, language=EN, orderNo=2, keyword=load curve), Keyword(id=1251249358285521490, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, language=EN, orderNo=3, keyword=convolutional neural network), Keyword(id=1251249358381990495, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, language=EN, orderNo=4, keyword=supervised learning), Keyword(id=1251249358495236717, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, language=EN, orderNo=5, keyword=load classification), Keyword(id=1251249358633648770, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, language=CN, orderNo=1, keyword=数据驱动), Keyword(id=1251249360219095694, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, language=CN, orderNo=2, keyword=负荷曲线), Keyword(id=1251249360323953302, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, language=CN, orderNo=3, keyword=卷积神经网络), Keyword(id=1251249360453976740, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, language=CN, orderNo=4, keyword=监督学习), Keyword(id=1251249360604971702, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, language=CN, orderNo=5, keyword=负荷分类)], refs=[Reference(id=1251249365688467543, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, doi=null, pmid=null, pmcid=null, year=2023, volume=47, issue=5, pageStart=1741, pageEnd=1750, url=null, language=null, rfNumber=[1], rfOrder=0, authorNames=康重庆, 杜尔顺, 郭鸿业, journalName=电网技术, refType=null, unstructuredReference=康重庆, 杜尔顺, 郭鸿业, . 新型电力系统的六要素分析[J]. 电网技术, 2023, 47(5): 1741-1750., articleTitle=新型电力系统的六要素分析, refAbstract=null), Reference(id=1251249365940125797, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, doi=null, pmid=null, pmcid=null, year=2023, volume=47, issue=5, pageStart=1741, pageEnd=1750, url=null, language=null, rfNumber=[1], rfOrder=1, authorNames=Kang Chongqing, Du Ershun, Guo Hongye, journalName=Power System Technology, refType=null, unstructuredReference=Kang Chongqing, Du Ershun, Guo Hongye, et al. Primary explo-ration of six essential factors in new power system[J]. Power System Technology, 2023, 47(5): 1741-1750., articleTitle=Primary explo-ration of six essential factors in new power system, refAbstract=null), Reference(id=1251249366095315062, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, doi=null, pmid=null, pmcid=null, year=2023, volume=23, issue=4, pageStart=1549, pageEnd=1556, url=null, language=null, rfNumber=[2], rfOrder=2, authorNames=刘洋, 王剑, 唐明, journalName=科学技术与工程, refType=null, unstructuredReference=刘洋, 王剑, 唐明, . 基于Hadoop分布式计算的混合神经网络负荷分类模型[J]. 科学技术与工程, 2023, 23(4): 1549-1556., articleTitle=基于Hadoop分布式计算的混合神经网络负荷分类模型, refAbstract=null), Reference(id=1251249366233727105, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, doi=null, pmid=null, pmcid=null, year=2023, volume=23, issue=4, pageStart=1549, pageEnd=1556, url=null, language=null, rfNumber=[2], rfOrder=3, authorNames=Liu Yang, Wang Jian, Tang Ming, journalName=Science Technology and Engineering, refType=null, unstructuredReference=Liu Yang, Wang Jian, Tang Ming, et al. Hybrid neural network load classification model based on Hadoop distributed platform[J]. Science Technology and Engineering, 2023, 23(4): 1549-1556., articleTitle=Hybrid neural network load classification model based on Hadoop distributed platform, refAbstract=null), Reference(id=1251249366401499276, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, doi=null, pmid=null, pmcid=null, year=2014, volume=34, issue=22, pageStart=3576, pageEnd=3589, url=null, language=null, rfNumber=[3], rfOrder=4, authorNames=田世明, 王蓓蓓, 张晶, journalName=中国电机工程学报, refType=null, unstructuredReference=田世明, 王蓓蓓, 张晶. 智能电网条件下的需求响应关键技术[J]. 中国电机工程学报, 2014, 34(22): 3576-3589., articleTitle=智能电网条件下的需求响应关键技术, refAbstract=null), Reference(id=1251249366531522710, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, doi=null, pmid=null, pmcid=null, year=2014, volume=34, issue=22, pageStart=3576, pageEnd=3589, url=null, language=null, rfNumber=[3], rfOrder=5, authorNames=Tian Shiming, Wang Beibei, Zhang Jing, journalName=Proceedings of the Chinese Society for Electrical Engineering, refType=null, unstructuredReference=Tian Shiming, Wang Beibei, Zhang Jing. Key technologies for demand response under smart grid conditions[J]. Proceedings of the Chinese Society for Electrical Engineering, 2014, 34(22): 3576-3589., articleTitle=Key technologies for demand response under smart grid conditions, refAbstract=null), Reference(id=1251249366653157545, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, doi=null, pmid=null, pmcid=null, year=2022, volume=22, issue=14, pageStart=5555, pageEnd=5561, url=null, language=null, rfNumber=[4], rfOrder=6, authorNames=李瑞, 柳长安, 王彦平, journalName=科学技术与工程, refType=null, unstructuredReference=李瑞, 柳长安, 王彦平, . 基于多生理信息迁移学习的脑力负荷分类[J]. 科学技术与工程, 2022, 22(14): 5555-5561., articleTitle=基于多生理信息迁移学习的脑力负荷分类, refAbstract=null), Reference(id=1251249366795763894, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, doi=null, pmid=null, pmcid=null, year=2022, volume=22, issue=14, pageStart=5555, pageEnd=5561, url=null, language=null, rfNumber=[4], rfOrder=7, authorNames=Li Rui, Liu Changan, Wang Yanping, journalName=Science Technology and Engineering, refType=null, unstructuredReference=Li Rui, Liu Changan, Wang Yanping, et al. Classification of mental workload based on multi-physiological information transfer learning[J]. Science Technology and Engineering, 2022, 22(14): 5555-5561., articleTitle=Classification of mental workload based on multi-physiological information transfer learning, refAbstract=null), Reference(id=1251249366967730372, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, doi=null, pmid=null, pmcid=null, year=2023, volume=46, issue=4, pageStart=37, pageEnd=43, url=null, language=null, rfNumber=[5], rfOrder=8, authorNames=钱育树, 孔钰婷, 黄聪, journalName=四川电力技术, refType=null, unstructuredReference=钱育树, 孔钰婷, 黄聪. 电力负荷预测研究综述[J]. 四川电力技术, 2023, 46(4): 37-43., articleTitle=电力负荷预测研究综述, refAbstract=null), Reference(id=1251249367097753807, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, doi=null, pmid=null, pmcid=null, year=2023, volume=46, issue=4, pageStart=37, pageEnd=43, url=null, language=null, rfNumber=[5], rfOrder=9, authorNames=Qian Yushu, Kong Yuting, Huang Cong, journalName=Sichuan Electric Power Technology, refType=null, unstructuredReference=Qian Yushu, Kong Yuting, Huang Cong. A review of research on power load forecasting[J]. Sichuan Electric Power Technology, 2023, 46(4): 37-43., articleTitle=A review of research on power load forecasting, refAbstract=null), Reference(id=1251249367223582940, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, doi=null, pmid=null, pmcid=null, year=2021, volume=21, issue=32, pageStart=13737, pageEnd=13743, url=null, language=null, rfNumber=[6], rfOrder=10, authorNames=赵忠啟, 常喜强, 樊艳芳, journalName=科学技术与工程, refType=null, unstructuredReference=赵忠啟, 常喜强, 樊艳芳, . 基于自编码器的电力负荷聚类分析[J]. 科学技术与工程, 2021, 21(32): 13737-13743., articleTitle=基于自编码器的电力负荷聚类分析, refAbstract=null), Reference(id=1251249367357800673, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, doi=null, pmid=null, pmcid=null, year=2021, volume=21, issue=32, pageStart=13737, pageEnd=13743, url=null, language=null, rfNumber=[6], rfOrder=11, authorNames=Zhao Zhongqi, Chang Xiqiang, Fan Yanfang, journalName=Science Technology and Engineering, refType=null, unstructuredReference=Zhao Zhongqi, Chang Xiqiang, Fan Yanfang, et al. Clustering analysis of power load curve based on auto-encoder[J]. Science Technology and Engineering, 2021, 21(32): 13737-13743., articleTitle=Clustering analysis of power load curve based on auto-encoder, refAbstract=null), Reference(id=1251249367508795633, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, doi=null, pmid=null, pmcid=null, year=2023, volume=23, issue=10, pageStart=4126, pageEnd=4134, url=null, language=null, rfNumber=[7], rfOrder=12, authorNames=蔡改贫, 宋佳, 罗小燕, journalName=科学技术与工程, refType=null, unstructuredReference=蔡改贫, 宋佳, 罗小燕, . 基于相空间重构和PSO-K-means的球磨机负荷状态识别方法[J]. 科学技术与工程, 2023, 23(10): 4126-4134., articleTitle=基于相空间重构和PSO-K-means的球磨机负荷状态识别方法, refAbstract=null), Reference(id=1251249369131991294, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, doi=null, pmid=null, pmcid=null, year=2023, volume=23, issue=10, pageStart=4126, pageEnd=4134, url=null, language=null, rfNumber=[7], rfOrder=13, authorNames=Cai Gaipin, Song Jia, Luo Xiaoyan, journalName=Science Technology and Engineering, refType=null, unstructuredReference=Cai Gaipin, Song Jia, Luo Xiaoyan, et al. Load state identification method for ball mills based on phase space reconstruction and PSO-K-means[J]. Science Technology and Engineering, 2023, 23(10): 4126-4134., articleTitle=Load state identification method for ball mills based on phase space reconstruction and PSO-K-means, refAbstract=null), Reference(id=1251249369308152072, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, doi=null, pmid=null, pmcid=null, year=2023, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[8], rfOrder=14, authorNames=吴阳, journalName=基于谱聚类和随机矩阵理论的中小负荷用户用电行为特征分析, refType=null, unstructuredReference=吴阳. 基于谱聚类和随机矩阵理论的中小负荷用户用电行为特征分析[D]. 吉林: 东北电力大学, 2023., articleTitle=null, refAbstract=null), Reference(id=1251249369396232465, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, doi=null, pmid=null, pmcid=null, year=2023, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[8], rfOrder=15, authorNames=Wu Yang, journalName=Analysis of electricity consumption behavior characteristics of small and medium load users based on spectral clustering and random matrix theory, refType=null, unstructuredReference=Wu Yang. Analysis of electricity consumption behavior characteristics of small and medium load users based on spectral clustering and random matrix theory[D]. Jilin: Northeast Electric Power University, 2023., articleTitle=null, refAbstract=null), Reference(id=1251249369530450200, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, doi=null, pmid=null, pmcid=null, year=2017, volume=10, issue=2, pageStart=1599, pageEnd=1607, url=null, language=null, rfNumber=[9], rfOrder=16, authorNames=Lin S, Li F, Tian E, journalName=IEEE Transactions on Smart Grid, refType=null, unstructuredReference=Lin S, Li F, Tian E, et al. Clustering load profiles for demand response applications[J]. IEEE Transactions on Smart Grid, 2017, 10(2): 1599-1607., articleTitle=Clustering load profiles for demand response applications, refAbstract=null), Reference(id=1251249369681445155, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, doi=null, pmid=null, pmcid=null, year=2018, volume=42, issue=3, pageStart=903, pageEnd=910, url=null, language=null, rfNumber=[10], rfOrder=17, authorNames=卜凡鹏, 陈俊艺, 张琪祁, journalName=电网技术, refType=null, unstructuredReference=卜凡鹏, 陈俊艺, 张琪祁, . 一种基于双层迭代聚类分析的负荷模式可控精细化识别方法[J]. 电网技术, 2018, 42(3): 903-910., articleTitle=一种基于双层迭代聚类分析的负荷模式可控精细化识别方法, refAbstract=null), Reference(id=1251249369811468585, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, doi=null, pmid=null, pmcid=null, year=2018, volume=42, issue=3, pageStart=903, pageEnd=910, url=null, language=null, rfNumber=[10], rfOrder=18, authorNames=Bu Fanpeng, Chen Junyi, Zhang Qiqi, journalName=Power System Technology, refType=null, unstructuredReference=Bu Fanpeng, Chen Junyi, Zhang Qiqi, et al. A controllable refined recognition method of electrical load pattern based on bilayer iterative clustering analysis[J]. Power System Technology, 2018, 42(3): 903-910., articleTitle=A controllable refined recognition method of electrical load pattern based on bilayer iterative clustering analysis, refAbstract=null), Reference(id=1251249369945686322, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, doi=null, pmid=null, pmcid=null, year=2018, volume=18, issue=16, pageStart=44, pageEnd=54, url=null, language=null, rfNumber=[11], rfOrder=19, authorNames=梁青艳, 孙彦广, journalName=科学技术与工程, refType=null, unstructuredReference=梁青艳, 孙彦广. 钢铁企业电力负荷动态预测建模问题的应用研究[J]. 科学技术与工程, 2018, 18(16): 44-54., articleTitle=钢铁企业电力负荷动态预测建模问题的应用研究, refAbstract=null), Reference(id=1251249370088292671, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, doi=null, pmid=null, pmcid=null, year=2018, volume=18, issue=16, pageStart=44, pageEnd=54, url=null, language=null, rfNumber=[11], rfOrder=20, authorNames=Liang Qingyan, Sun Yanguang, journalName=Science Technology and Engineering, refType=null, unstructuredReference=Liang Qingyan, Sun Yanguang. Application research on dynamic forecasting modeling of power load in steel enterprises[J]. Science Technology and Engineering, 2018, 18(16): 44-54., articleTitle=Application research on dynamic forecasting modeling of power load in steel enterprises, refAbstract=null), Reference(id=1251249370205733189, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, doi=null, pmid=null, pmcid=null, year=2020, volume=44, issue=9, pageStart=3508, pageEnd=3515, url=null, language=null, rfNumber=[12], rfOrder=21, authorNames=林顺富, 顾乡, 汤继开, journalName=电网技术, refType=null, unstructuredReference=林顺富, 顾乡, 汤继开, . 基于稀疏自动编码器神经网络的负荷曲线分类方法[J]. 电网技术, 2020, 44(9): 3508-3515., articleTitle=基于稀疏自动编码器神经网络的负荷曲线分类方法, refAbstract=null), Reference(id=1251249370381893972, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, doi=null, pmid=null, pmcid=null, year=2020, volume=44, issue=9, pageStart=3508, pageEnd=3515, url=null, language=null, rfNumber=[12], rfOrder=22, authorNames=Lin Shunfu, Gu Xiang, Tang Jikai, journalName=Power System Technology, refType=null, unstructuredReference=Lin Shunfu, Gu Xiang, Tang Jikai, et al. Power load profile classification method based on neural network of sparse automatic encoder[J]. Power System Technology, 2020, 44(9): 3508-3515., articleTitle=Power load profile classification method based on neural network of sparse automatic encoder, refAbstract=null), Reference(id=1251249370545471841, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, doi=null, pmid=null, pmcid=null, year=2020, volume=43, issue=5, pageStart=755, pageEnd=780, url=null, language=null, rfNumber=[13], rfOrder=23, authorNames=徐冰冰, 岑科廷, 黄俊杰, journalName=计算机学报, refType=null, unstructuredReference=徐冰冰, 岑科廷, 黄俊杰, . 图卷积神经网络综述[J]. 计算机学报, 2020, 43(5): 755-780., articleTitle=图卷积神经网络综述, refAbstract=null), Reference(id=1251249370696466800, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, doi=null, pmid=null, pmcid=null, year=2020, volume=43, issue=5, pageStart=755, pageEnd=780, url=null, language=null, rfNumber=[13], rfOrder=24, authorNames=Xu Bingbing, Cen Keting, Huang Junjie, journalName=Chinese Journal of Computers, refType=null, unstructuredReference=Xu Bingbing, Cen Keting, Huang Junjie, et al. A review of graph convolutional neural networks[J]. Chinese Journal of Computers, 2020, 43(5): 755-780., articleTitle=A review of graph convolutional neural networks, refAbstract=null), Reference(id=1251249370818101628, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, doi=null, pmid=null, pmcid=null, year=2023, volume=null, issue=null, pageStart=1, pageEnd=5, url=null, language=null, rfNumber=[14], rfOrder=25, authorNames=Garimella M, Bhan M, journalName=2023 International Conference on Network, Multimedia and Information Technology (NMITCON), refType=null, unstructuredReference=Garimella M, Bhan M. Image labeling using convolutional neural network[C]// 2023 International Conference on Network, Multimedia and Information Technology (NMITCON). Bengaluru: Nitte Meenakshi Inst of Tech, 2023: 1-5., articleTitle=Image labeling using convolutional neural network, refAbstract=null), Reference(id=1251249370964902284, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, doi=null, pmid=null, pmcid=null, year=2013, volume=13, issue=34, pageStart=10177, pageEnd=10181, url=null, language=null, rfNumber=[15], rfOrder=26, authorNames=张庆新, 崔展博, 马睿, journalName=科学技术与工程, refType=null, unstructuredReference=张庆新, 崔展博, 马睿, . 基于K-means聚类与径向基神经(RBF)网络的电力系统日负荷预测[J]. 科学技术与工程, 2013, 13(34): 10177-10181., articleTitle=基于K-means聚类与径向基神经(RBF)网络的电力系统日负荷预测, refAbstract=null), Reference(id=1251249371082342808, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, doi=null, pmid=null, pmcid=null, year=2013, volume=13, issue=34, pageStart=10177, pageEnd=10181, url=null, language=null, rfNumber=[15], rfOrder=27, authorNames=Zhang Qingxin, Cui Zhanbo, Ma Rui, journalName=Science Technology and Engineering, refType=null, unstructuredReference=Zhang Qingxin, Cui Zhanbo, Ma Rui, et al. Daily load forecasting of power system based on K-means clustering and radial basis function neural network[J]. Science Technology and Engineering, 2013, 13(34): 10177-10181., articleTitle=Daily load forecasting of power system based on K-means clustering and radial basis function neural network, refAbstract=null), Reference(id=1251249371187200416, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, doi=null, pmid=null, pmcid=null, year=2019, volume=35, issue=4, pageStart=327, pageEnd=332, url=null, language=null, rfNumber=[16], rfOrder=28, authorNames=李仲恒, 刘蓉晖, journalName=上海电力大学学报, refType=null, unstructuredReference=李仲恒, 刘蓉晖. 基于ISODATA的电力负荷曲线分类[J]. 上海电力大学学报, 2019, 35(4): 327-332., articleTitle=基于ISODATA的电力负荷曲线分类, refAbstract=null), Reference(id=1251249371438858665, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, doi=null, pmid=null, pmcid=null, year=2019, volume=35, issue=4, pageStart=327, pageEnd=332, url=null, language=null, rfNumber=[16], rfOrder=29, authorNames=Li Zhongheng, Liu Ronghui, journalName=Journal of Shanghai University of Electric Power, refType=null, unstructuredReference=Li Zhongheng, Liu Ronghui. A load curve clustering algorithm based on ISODATA[J]. Journal of Shanghai University of Electric Power, 2019, 35(4): 327-332., articleTitle=A load curve clustering algorithm based on ISODATA, refAbstract=null), Reference(id=1251249371568882101, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, doi=null, pmid=null, pmcid=null, year=2020, volume=44, issue=9, pageStart=3497, pageEnd=3507, url=null, language=null, rfNumber=[17], rfOrder=30, authorNames=朱天怡, 艾芊, 贺兴, journalName=电网技术, refType=null, unstructuredReference=朱天怡, 艾芊, 贺兴, . 基于数据驱动的用电行为分析方法及应用综述[J]. 电网技术, 2020, 44(9): 3497-3507., articleTitle=基于数据驱动的用电行为分析方法及应用综述, refAbstract=null), Reference(id=1251249371711488448, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, doi=null, pmid=null, pmcid=null, year=2020, volume=44, issue=9, pageStart=3497, pageEnd=3507, url=null, language=null, rfNumber=[17], rfOrder=31, authorNames=Zhu Tianyi, Ai Qian, He Xing, journalName=Power System Technology, refType=null, unstructuredReference=Zhu Tianyi, Ai Qian, He Xing, et al. An overview of data-driven electricity consumption behavior analysis method and application[J]. Power System Technology, 2020, 44(9): 3497-3507., articleTitle=An overview of data-driven electricity consumption behavior analysis method and application, refAbstract=null), Reference(id=1251249371870872012, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, doi=null, pmid=null, pmcid=null, year=2023, volume=47, issue=10, pageStart=4291, pageEnd=4301, url=null, language=null, rfNumber=[18], rfOrder=32, authorNames=董雷, 陈振平, 韩富佳, journalName=电网技术, refType=null, unstructuredReference=董雷, 陈振平, 韩富佳, . 基于图卷积神经网络与K-means聚类的居民用户集群短期负荷预测[J]. 电网技术, 2023, 47(10): 4291-4301., articleTitle=基于图卷积神经网络与K-means聚类的居民用户集群短期负荷预测, refAbstract=null), Reference(id=1251249372038644182, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, doi=null, pmid=null, pmcid=null, year=2023, volume=47, issue=10, pageStart=4291, pageEnd=4301, url=null, language=null, rfNumber=[18], rfOrder=33, authorNames=Dong Lei, Chen Zhenping, Han Fujia, journalName=Power System Technology, refType=null, unstructuredReference=Dong Lei, Chen Zhenping, Han Fujia, et al. Short-term load forecasting of residential user groups based on graph convolutional neural network and K-means clustering[J]. Power System Technology, 2023, 47(10): 4291-4301., articleTitle=Short-term load forecasting of residential user groups based on graph convolutional neural network and K-means clustering, refAbstract=null), Reference(id=1251249373213049310, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, doi=null, pmid=null, pmcid=null, year=2020, volume=41, issue=8, pageStart=17, pageEnd=24, url=null, language=null, rfNumber=[19], rfOrder=34, authorNames=唐子卓, 刘洋, 许立雄, journalName=电力建设, refType=null, unstructuredReference=唐子卓, 刘洋, 许立雄, . 基于负荷数据频域特征和LSTM网络的类别不平衡负荷典型用电模式提取方法[J]. 电力建设, 2020, 41(8): 17-24., articleTitle=基于负荷数据频域特征和LSTM网络的类别不平衡负荷典型用电模式提取方法, refAbstract=null), Reference(id=1251249373418570222, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, doi=null, pmid=null, pmcid=null, year=2020, volume=41, issue=8, pageStart=17, pageEnd=24, url=null, language=null, rfNumber=[19], rfOrder=35, authorNames=Tang Zizhuo, Liu Yang, Xu Lixiong, journalName=Electric Power Construction, refType=null, unstructuredReference=Tang Zizhuo, Liu Yang, Xu Lixiong, et al. Imbalanced-load pattern extraction method based on frequency domain characteristics of load data and LSTM network[J]. Electric Power Construction, 2020, 41(8): 17-24., articleTitle=Imbalanced-load pattern extraction method based on frequency domain characteristics of load data and LSTM network, refAbstract=null), Reference(id=1251249373527622138, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, doi=null, pmid=null, pmcid=null, year=2022, volume=43, issue=2, pageStart=89, pageEnd=97, url=null, language=null, rfNumber=[20], rfOrder=36, authorNames=张杰, 刘洋, 李文峰, journalName=电力建设, refType=null, unstructuredReference=张杰, 刘洋, 李文峰, . 基于两阶段数据增强和双向深度残差TCN的用户负荷曲线分类方法[J]. 电力建设, 2022, 43(2): 89-97., articleTitle=基于两阶段数据增强和双向深度残差TCN的用户负荷曲线分类方法, refAbstract=null), Reference(id=1251249373687005695, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, doi=null, pmid=null, pmcid=null, year=2022, volume=43, issue=2, pageStart=89, pageEnd=97, url=null, language=null, rfNumber=[20], rfOrder=37, authorNames=Zhang Jie, Liu Yang, Li Wenfeng, journalName=Electric Power Construction, refType=null, unstructuredReference=Zhang Jie, Liu Yang, Li Wenfeng, et al. Power load curve identification method based on two-phase data enhancement and bi-directional deep residual TCN[J]. Electric Power Construction, 2022, 43(2): 89-97., articleTitle=Power load curve identification method based on two-phase data enhancement and bi-directional deep residual TCN, refAbstract=null), Reference(id=1251249373800251914, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, doi=null, pmid=null, pmcid=null, year=2022, volume=null, issue=null, pageStart=104, pageEnd=109, url=null, language=null, rfNumber=[21], rfOrder=38, authorNames=Xiong J, Hong H, Xu C, journalName=null, refType=null, unstructuredReference=Xiong J, Hong H, Xu C, et al. Graph convolutional network based electricity demand forecasting in power distribution networks[C]//2022 International Conference on Artificial Intelligence and Autonomous Robot Systems (AIARS). Bristol: IEEE, 2022: 104-109., articleTitle=Graph convolutional network based electricity demand forecasting in power distribution networks, refAbstract=null), Reference(id=1251249373930275345, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, doi=null, pmid=null, pmcid=null, year=2021, volume=227, issue=null, pageStart=120480, pageEnd=null, url=null, language=null, rfNumber=[22], rfOrder=39, authorNames=Imani M, journalName=Energy, refType=null, unstructuredReference=Imani M. Electrical load-temperature CNN for residential load fore-casting[J]. Energy, 2021, 227: 120480., articleTitle=Electrical load-temperature CNN for residential load fore-casting, refAbstract=null), Reference(id=1251249374047715864, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, doi=null, pmid=null, pmcid=null, year=2017, volume=null, issue=null, pageStart=11, pageEnd=21, url=null, language=null, rfNumber=[23], rfOrder=40, authorNames=de Paiva P D, Castro A R G, journalName=3rd International Conference on Artificial Intelligence and Applications, refType=null, unstructuredReference=de Paiva P D, Castro A R G. Convolutional neural network applied to the identification of residential equipment in non-intrusive load monitoring systems[C]// 3rd International Conference on Artificial Intelligence and Applications. Tokyo: AIAT, 2017: 11-21., articleTitle=Convolutional neural network applied to the identification of residential equipment in non-intrusive load monitoring systems, refAbstract=null), Reference(id=1251249374127407643, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, doi=null, pmid=null, pmcid=null, year=2016, volume=40, issue=3, pageStart=804, pageEnd=811, url=null, language=null, rfNumber=[24], rfOrder=41, authorNames=张铁峰, 顾明迪, journalName=电网技术, refType=null, unstructuredReference=张铁峰, 顾明迪. 电力用户负荷模式提取技术及应用综述[J]. 电网技术, 2016, 40(3): 804-811., articleTitle=电力用户负荷模式提取技术及应用综述, refAbstract=null), Reference(id=1251249374253236773, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, doi=null, pmid=null, pmcid=null, year=2016, volume=40, issue=3, pageStart=804, pageEnd=811, url=null, language=null, rfNumber=[24], rfOrder=42, authorNames=Zhang Tiefeng, Gu Mingdi, journalName=Power System Technology, refType=null, unstructuredReference=Zhang Tiefeng, Gu Mingdi. Overview of electricity customer load pattern extraction technology and its application[J]. Power System Technology, 2016, 40(3): 804-811., articleTitle=Overview of electricity customer load pattern extraction technology and its application, refAbstract=null), Reference(id=1251249374383260208, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, doi=null, pmid=null, pmcid=null, year=2018, volume=42, issue=10, pageStart=3393, pageEnd=3400, url=null, language=null, rfNumber=[25], rfOrder=43, authorNames=王德文, 周昉昉, journalName=电网技术, refType=null, unstructuredReference=王德文, 周昉昉. 基于无监督极限学习机的用电负荷模式提取[J]. 电网技术, 2018, 42(10): 3393-3400., articleTitle=基于无监督极限学习机的用电负荷模式提取, refAbstract=null), Reference(id=1251249374525866555, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, doi=null, pmid=null, pmcid=null, year=2018, volume=42, issue=10, pageStart=3393, pageEnd=3400, url=null, language=null, rfNumber=[25], rfOrder=44, authorNames=Wang Dewen, Zhou Fangfang, journalName=Power System Technology, refType=null, unstructuredReference=Wang Dewen, Zhou Fangfang. Extraction of electricity consumption load pattern based on unsupervised extreme learning machine[J]. Power System Technology, 2018, 42(10): 3393-3400., articleTitle=Extraction of electricity consumption load pattern based on unsupervised extreme learning machine, refAbstract=null)], funds=[Fund(id=1251249365327757364, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, awardId=5400-202255154A-1-1-ZN, language=CN, fundingSource=国家电网有限公司总部科技项目(5400-202255154A-1-1-ZN), fundOrder=null, country=null)], companyList=[AuthorCompany(id=1251249355865407858, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, xref=1, ext=[AuthorCompanyExt(id=1251249355873796468, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, companyId=1251249355865407858, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 China Electric Power Research Institute, Beijing 100192, China), AuthorCompanyExt(id=1251249355882185076, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, companyId=1251249355865407858, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 中国电力科学研究院有限公司, 北京 100192)]), AuthorCompany(id=1251249355957682554, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, xref=2, ext=[AuthorCompanyExt(id=1251249355970265469, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, companyId=1251249355957682554, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 School of Computer Science, Beijing University of Posts and Telecommunications, Beijing 100876, China), AuthorCompanyExt(id=1251249355974459773, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, companyId=1251249355957682554, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 北京邮电大学计算机学院, 北京 100876)])], figs=[ArticleFig(id=1251249360885990108, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, language=EN, label=Fig.1, caption=Model diagram, figureFileSmall=0eEqwg1KXVJ9HEnc5uZK+g==, figureFileBig=tc69NEiAP6++ua1esBLejg==, tableContent=null), ArticleFig(id=1251249360990847726, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, language=CN, label=图1, caption=模型架构图, figureFileSmall=0eEqwg1KXVJ9HEnc5uZK+g==, figureFileBig=tc69NEiAP6++ua1esBLejg==, tableContent=null), ArticleFig(id=1251249361133454083, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, language=EN, label=Fig.2, caption=Load image convolution process, figureFileSmall=2k49AWj2g8OiVa7iIRha+w==, figureFileBig=RaiD3C+OHJ7BWhq5hB0GtQ==, tableContent=null), ArticleFig(id=1251249361229923092, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, language=CN, label=图2, caption=负荷图片卷积流程, figureFileSmall=2k49AWj2g8OiVa7iIRha+w==, figureFileBig=RaiD3C+OHJ7BWhq5hB0GtQ==, tableContent=null), ArticleFig(id=1251249361368335141, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, language=EN, label=Fig.3, caption=ResNet34 network diagram, figureFileSmall=s43u25norFrs28OLfAeSvg==, figureFileBig=eH/Vql67on0OVo892ih21w==, tableContent=null), ArticleFig(id=1251249361527718717, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, language=CN, label=图3, caption=ResNet34网络图, figureFileSmall=s43u25norFrs28OLfAeSvg==, figureFileBig=eH/Vql67on0OVo892ih21w==, tableContent=null), ArticleFig(id=1251249361695490895, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, language=EN, label=Fig.4, caption=Different electric load line charts, figureFileSmall=fyPgKEKTWr7TeiFCLdmg+Q==, figureFileBig=AE8Xs2RgNTSgtXQVsxeuag==, tableContent=null), ArticleFig(id=1251249361846485862, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, language=CN, label=图4, caption=不同的用电负荷折线图, figureFileSmall=fyPgKEKTWr7TeiFCLdmg+Q==, figureFileBig=AE8Xs2RgNTSgtXQVsxeuag==, tableContent=null), ArticleFig(id=1251249361934566254, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, language=EN, label=Fig.5, caption=Variation of training accuracy of different networks, figureFileSmall=qswi5XG2EVdmyKvxrVYC4w==, figureFileBig=sPp7Y9h1HsKgzCpIICkRsA==, tableContent=null), ArticleFig(id=1251249362026840956, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, language=CN, label=图5, caption=不同网络训练精度变化, figureFileSmall=qswi5XG2EVdmyKvxrVYC4w==, figureFileBig=sPp7Y9h1HsKgzCpIICkRsA==, tableContent=null), ArticleFig(id=1251249362152670095, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, language=EN, label=Fig.6, caption=Variation of training accuracy of residual NN, figureFileSmall=rfHn6cWH4B+AY3pU2u6Mlw==, figureFileBig=fGwaM9ImogXrEZ7kJ5Z3WQ==, tableContent=null), ArticleFig(id=1251249362240750489, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, language=CN, label=图6, caption=残差神经网络训练精度变化, figureFileSmall=rfHn6cWH4B+AY3pU2u6Mlw==, figureFileBig=fGwaM9ImogXrEZ7kJ5Z3WQ==, tableContent=null), ArticleFig(id=1251249362450465703, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, language=EN, label=Fig.7, caption=The results of load forecasting, figureFileSmall=uQ9y+wvnCVMiR6iPYw630g==, figureFileBig=ooaVJqobcObVU+sjw/LT4A==, tableContent=null), ArticleFig(id=1251249362584683445, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, language=CN, label=图7, caption=负荷预测结果, figureFileSmall=uQ9y+wvnCVMiR6iPYw630g==, figureFileBig=ooaVJqobcObVU+sjw/LT4A==, tableContent=null), ArticleFig(id=1251249362668569535, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, language=EN, label=Table 1, caption=

Each network structure

, figureFileSmall=null, figureFileBig=null, tableContent=
视觉几何组16 残差网络-18 残差网络-34 残差网络-50
224×224三通道图片
卷积核3, 64
卷积核3, 64
卷积核7, 64,步长2
最大池化,步长2 最大池化,步长2
卷积核3,128
卷积核3,128
2$\left[\begin{array}{l}\mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{3,64}\\ \mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{3,64}\end{array}\right]$ 3$\left[\begin{array}{l}\mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{3,64}\\ \mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{3,64}\end{array}\right]$ 3$\left[\begin{array}{l}\mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{1,64}\\ \mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{3,64}\\ \mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{1,256}\end{array}\right]$
最大池化
步长2
1$\left[\begin{array}{l}\mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{3,128}\\ \mathrm{步}\mathrm{长}2\\ \mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{3,128}\end{array}\right]$ 1$\left[\begin{array}{l}\mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{3,128}\\ \mathrm{步}\mathrm{长}2\\ \mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{3,128}\end{array}\right]$ 1$\left[\begin{array}{l}\begin{array}{l}\mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{1,128}\\ \mathrm{步}\mathrm{长}2\end{array}\\ \begin{array}{l}\mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{3,128}\\ \mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{1,512}\end{array}\end{array}\right]$
卷积核3,256
卷积核3,256
卷积核1,256
1$\left[\begin{array}{l}\mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{3,128}\\ \mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{3,128}\end{array}\right]$ 3$\left[\begin{array}{l}\mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{3,128}\\ \mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{3,128}\end{array}\right]$ 3$\left[\begin{array}{l}\mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{1,128}\\ \mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{3,128}\\ \mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{1,512}\end{array}\right]$
最大池化
步长2
1$\left[\begin{array}{l}\mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{3,256}\\ \mathrm{步}\mathrm{长}2\\ \mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{3,256}\end{array}\right]$ 1$\left[\begin{array}{l}\mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{3,256}\\ \mathrm{步}\mathrm{长}2\\ \mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{3,256}\end{array}\right]$ 1$\left[\begin{array}{l}\begin{array}{l}\mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{1,256}\\ \mathrm{步}\mathrm{长}2\end{array}\\ \begin{array}{l}\mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{3,256}\\ \mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{1,1024}\end{array}\end{array}\right]$
卷积核3,512
卷积核3,512
卷积核1,512
1$\left[\begin{array}{l}\mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{3,256}\\ \mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{3,256}\end{array}\right]$ 5$\left[\begin{array}{l}\mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{3,256}\\ \mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{3,256}\end{array}\right]$ 5$\left[\begin{array}{l}\mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{1,256}\\ \mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{3,256}\\ \mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{1,1}\mathrm{ }024\end{array}\right]$
最大池化
步长2
1$\left[\begin{array}{l}\mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{3,512}\\ \mathrm{步}\mathrm{长}2\\ \mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{3,512}\end{array}\right]$ 1$\left[\begin{array}{l}\mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{3,512}\\ \mathrm{步}\mathrm{长}2\\ \mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{3,512}\end{array}\right]$ 1$\left[\begin{array}{l}\begin{array}{l}\mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{1,512}\\ \mathrm{步}\mathrm{长}2\end{array}\\ \begin{array}{l}\mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{3,512}\\ \mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{1,2048}\end{array}\end{array}\right]$
卷积核3,512
卷积核3,512
卷积核1,512
1$\left[\begin{array}{l}\mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{3,512}\\ \mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{3,512}\end{array}\right]$ 2$\left[\begin{array}{l}\mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{3,512}\\ \mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{3,512}\end{array}\right]$ 2$\left[\begin{array}{l}\mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{1,512}\\ \mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{3,512}\\ \mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{1,2}\mathrm{ }048\end{array}\right]$
全局平均池化
全连接层1 024神经元
归一化指数函数输出维度16
), ArticleFig(id=1251249362790204363, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, language=CN, label=表1, caption=

各网络结构

, figureFileSmall=null, figureFileBig=null, tableContent=
视觉几何组16 残差网络-18 残差网络-34 残差网络-50
224×224三通道图片
卷积核3, 64
卷积核3, 64
卷积核7, 64,步长2
最大池化,步长2 最大池化,步长2
卷积核3,128
卷积核3,128
2$\left[\begin{array}{l}\mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{3,64}\\ \mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{3,64}\end{array}\right]$ 3$\left[\begin{array}{l}\mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{3,64}\\ \mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{3,64}\end{array}\right]$ 3$\left[\begin{array}{l}\mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{1,64}\\ \mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{3,64}\\ \mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{1,256}\end{array}\right]$
最大池化
步长2
1$\left[\begin{array}{l}\mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{3,128}\\ \mathrm{步}\mathrm{长}2\\ \mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{3,128}\end{array}\right]$ 1$\left[\begin{array}{l}\mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{3,128}\\ \mathrm{步}\mathrm{长}2\\ \mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{3,128}\end{array}\right]$ 1$\left[\begin{array}{l}\begin{array}{l}\mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{1,128}\\ \mathrm{步}\mathrm{长}2\end{array}\\ \begin{array}{l}\mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{3,128}\\ \mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{1,512}\end{array}\end{array}\right]$
卷积核3,256
卷积核3,256
卷积核1,256
1$\left[\begin{array}{l}\mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{3,128}\\ \mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{3,128}\end{array}\right]$ 3$\left[\begin{array}{l}\mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{3,128}\\ \mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{3,128}\end{array}\right]$ 3$\left[\begin{array}{l}\mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{1,128}\\ \mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{3,128}\\ \mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{1,512}\end{array}\right]$
最大池化
步长2
1$\left[\begin{array}{l}\mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{3,256}\\ \mathrm{步}\mathrm{长}2\\ \mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{3,256}\end{array}\right]$ 1$\left[\begin{array}{l}\mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{3,256}\\ \mathrm{步}\mathrm{长}2\\ \mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{3,256}\end{array}\right]$ 1$\left[\begin{array}{l}\begin{array}{l}\mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{1,256}\\ \mathrm{步}\mathrm{长}2\end{array}\\ \begin{array}{l}\mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{3,256}\\ \mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{1,1024}\end{array}\end{array}\right]$
卷积核3,512
卷积核3,512
卷积核1,512
1$\left[\begin{array}{l}\mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{3,256}\\ \mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{3,256}\end{array}\right]$ 5$\left[\begin{array}{l}\mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{3,256}\\ \mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{3,256}\end{array}\right]$ 5$\left[\begin{array}{l}\mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{1,256}\\ \mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{3,256}\\ \mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{1,1}\mathrm{ }024\end{array}\right]$
最大池化
步长2
1$\left[\begin{array}{l}\mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{3,512}\\ \mathrm{步}\mathrm{长}2\\ \mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{3,512}\end{array}\right]$ 1$\left[\begin{array}{l}\mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{3,512}\\ \mathrm{步}\mathrm{长}2\\ \mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{3,512}\end{array}\right]$ 1$\left[\begin{array}{l}\begin{array}{l}\mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{1,512}\\ \mathrm{步}\mathrm{长}2\end{array}\\ \begin{array}{l}\mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{3,512}\\ \mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{1,2048}\end{array}\end{array}\right]$
卷积核3,512
卷积核3,512
卷积核1,512
1$\left[\begin{array}{l}\mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{3,512}\\ \mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{3,512}\end{array}\right]$ 2$\left[\begin{array}{l}\mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{3,512}\\ \mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{3,512}\end{array}\right]$ 2$\left[\begin{array}{l}\mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{1,512}\\ \mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{3,512}\\ \mathrm{卷}\mathrm{积}\mathrm{核}\mathrm{1,2}\mathrm{ }048\end{array}\right]$
全局平均池化
全连接层1 024神经元
归一化指数函数输出维度16
), ArticleFig(id=1251249362903450588, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, language=EN, label=Table 2, caption=

Different network accuracy

, figureFileSmall=null, figureFileBig=null, tableContent=
模型名称 测试集准确度/% 模型名称 测试集准确度/%
VGG-16 88.21 ResNet-34 90.34
ResNet-18 90.38 ResNet-50 86.11
), ArticleFig(id=1251249363012502505, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, language=CN, label=表2, caption=

不同网络准确度

, figureFileSmall=null, figureFileBig=null, tableContent=
模型名称 测试集准确度/% 模型名称 测试集准确度/%
VGG-16 88.21 ResNet-34 90.34
ResNet-18 90.38 ResNet-50 86.11
), ArticleFig(id=1251249363117360115, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, language=EN, label=Table 3, caption=

Accuracy comparison between random forest and residual neural network

, figureFileSmall=null, figureFileBig=null, tableContent=
数据集 模型 精度
测试集 随机森林 0.883
残差神经网络 0.903
一周数据投票 随机森林 0.934
残差神经网络 0.941
), ArticleFig(id=1251249364686028801, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, language=CN, label=表3, caption=

随机森林与残差神经网络精度对比

, figureFileSmall=null, figureFileBig=null, tableContent=
数据集 模型 精度
测试集 随机森林 0.883
残差神经网络 0.903
一周数据投票 随机森林 0.934
残差神经网络 0.941
), ArticleFig(id=1251249364866383886, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, language=EN, label=Table 4, caption=

MAPE for classification prediction and non-classification prediction

, figureFileSmall=null, figureFileBig=null, tableContent=
预测类别 MAPE/%
分类预测 不分类预测
全服务餐厅 10.1 32.8
医院 10.5 33.6
大型旅馆 10.3 35.8
大型办公室 10.0 36.1
中型办公室 9.3 36.6
多层公寓 9.3 38.7
门诊 9.4 40.1
小学 9.5 39.1
快餐厅 9.2 36.9
中学 9.9 35.5
小型旅馆 10.4 34.0
小型办公室 10.6 32.7
零售店 10.6 31.7
购物中心 10.4 30.6
超市 10.2 29.2
仓库 9.0 28.6
), ArticleFig(id=1251249365050933277, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781959334916772, language=CN, label=表4, caption=

分类预测和不分类预测的MAPE

, figureFileSmall=null, figureFileBig=null, tableContent=
预测类别 MAPE/%
分类预测 不分类预测
全服务餐厅 10.1 32.8
医院 10.5 33.6
大型旅馆 10.3 35.8
大型办公室 10.0 36.1
中型办公室 9.3 36.6
多层公寓 9.3 38.7
门诊 9.4 40.1
小学 9.5 39.1
快餐厅 9.2 36.9
中学 9.9 35.5
小型旅馆 10.4 34.0
小型办公室 10.6 32.7
零售店 10.6 31.7
购物中心 10.4 30.6
超市 10.2 29.2
仓库 9.0 28.6
)], attaches=null, journal=Journal(id=1146119176004939786, delFlag=0, nameCn=科学技术与工程, nameEn=Science Technology and Engineering, nameHistory1=null, nameHistory2=null, issn=1671-1815, eissn=, cn=11-4688/T, coden=null, periodic=4, language=CN, oaType=是, ccby=null, superviseOffice=null, ownerOffice=null, pubOffice=null, editorOffice=null, officeType=null, aims=null, clcCode=null, officeProv=null, officeCity=null, officeAddr=null, officeZip=null, officeEmail=null, officePhone=null, editDirector=null, officeDirector=null, officeDirectorPhone=null, officeStaffNum=null, officeEmpNum=null, coverPicUrl=UKU/O7GSka5polgCTkbIIw==, journalPrice=null, startedYear=null, abbrevIsoEn=Sci Technol Eng, journalRemark=null, publicationField=null, createdTime=null, updatedTime=1754445529766, createdBy=null, updatedBy=13701087609, firstLetterCn=S, firstLetterEn=S, subjectCode=Natural Sciences, subjectName=自然科学, subjectCodeEn=Natural Sciences, subjectNameEn=null, picCn=UKU/O7GSka5polgCTkbIIw==, picEn=5hwlULoNwcbj3xUmVi9MAQ==, jcr=null, cjcr=null, exts=[JournalExt(id=1159791870395564357, language=CN, name=科学技术与工程, nameHistory1=null, nameHistory2=null, managedBy=, sponsoredBy=, publishedBy=, editorOffice=, officeProv=null, officeCity=null, officeAddr=, officeZip=, editDirector=null, officeDirector=null, officePhone=null, coverPicUrl=null, journalRemark=, submitArticleUrl=null, websiteUrl=http://www.stae.com.cn/jsygc/home, createdTime=1754445529793, updatedTime=1754445529793, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=http://www.stae.com.cn/jsygc/site/menus/20090429150146001, submissionAuthorUrl=http://www.stae.com.cn/jsygc/author/login, submissionEditorUrl=http://www.stae.com.cn/jsygc/editor/login, submissionReviewUrl=http://www.stae.com.cn/jsygc/reviewer/login, submissionCeEditorUrl=, submissionAeEditorUrl=, option={"copyright":""}), JournalExt(id=1159791870441701702, language=EN, name=Science Technology and Engineering, nameHistory1=null, nameHistory2=null, managedBy=, sponsoredBy=, publishedBy=, editorOffice=, officeProv=null, officeCity=null, officeAddr=, officeZip=, editDirector=null, officeDirector=null, officePhone=null, coverPicUrl=null, journalRemark=, submitArticleUrl=null, websiteUrl=http://www.stae.com.cn/jsygc/home, createdTime=1754445529804, updatedTime=1754445529804, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=, submissionAuthorUrl=http://www.stae.com.cn/jsygc/author/login, submissionEditorUrl=http://www.stae.com.cn/jsygc/editor/login, submissionReviewUrl=http://www.stae.com.cn/jsygc/reviewer/login, submissionCeEditorUrl=, submissionAeEditorUrl=, option={"copyright":""})], databaseList=null, tenantJournalId=1146123166801305609, websiteList=[Website(id=1148243202391400884, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1146123166801305609, journalNameCn=null, journalNameEn=null, grayFlag=null, tenantId=1146029695717560320, platformId=null, journalGroupId=null, journalGroupNameCn=null, journalGroupNameEn=null, type=1, domain=https://castjournals.cast.org.cn/joweb/kxjsygc/CN, language=CN, createTime=1751692112777, createBy=18614031015, updateTime=1753520965431, updateBy=18614031015, name=科学技术与工程-中文站点, tplId=1146099689490845704, title=科学技术与工程, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1148622798802673703, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202391400884, code=articleTextType, value=kx, createTime=1751782615614, updateTime=1751782615614, creator=18614031015, updator=18614031015), WebsiteProps(id=1148622798781702180, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202391400884, code=banner, value=null, createTime=1751782615609, updateTime=1751782615609, creator=18614031015, updator=18614031015), WebsiteProps(id=1148622798769119267, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202391400884, code=logo, value=https://castjournals.cast.org.cn/joweb/kjdb/CN/file/pic?fileId=j86gbwi+p0Idkyl5SzIlmQ==, createTime=1751782615606, updateTime=1751782615606, creator=18614031015, updator=18614031015), WebsiteProps(id=1148622798794285094, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202391400884, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/kjdb/CN/file/pic, createTime=1751782615612, updateTime=1751782615612, creator=18614031015, updator=18614031015), WebsiteProps(id=1148622798790090789, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202391400884, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_cn_619/, createTime=1751782615611, updateTime=1751782615611, creator=18614031015, updator=18614031015)]), Website(id=1155914124811976731, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1146123166801305609, journalNameCn=null, journalNameEn=null, grayFlag=null, tenantId=1146029695717560320, platformId=null, journalGroupId=null, journalGroupNameCn=null, journalGroupNameEn=null, type=1, domain=https://castjournals.cast.org.cn/joweb/kxjsygc/EN, language=EN, createTime=1753521003206, createBy=18614031015, updateTime=1753521003206, updateBy=18614031015, name=科学技术与工程-英文站点, tplId=1146101810881728533, title=Science Technology and Engineering, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1155914371227308235, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155914124811976731, code=articleTextType, value=kx, createTime=1753521061952, updateTime=1753521061952, creator=18614031015, updator=18614031015), WebsiteProps(id=1155914371210531016, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155914124811976731, code=banner, value=null, createTime=1753521061947, updateTime=1753521061947, creator=18614031015, updator=18614031015), WebsiteProps(id=1155914371202142407, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155914124811976731, code=logo, value=https://castjournals.cast.org.cn/joweb/kjdb/CN/file/pic?fileId=j86gbwi+p0Idkyl5SzIlmQ==, createTime=1753521061945, updateTime=1753521061945, creator=18614031015, updator=18614031015), WebsiteProps(id=1155914371223113930, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155914124811976731, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/kjdb/CN/file/pic, createTime=1753521061950, updateTime=1753521061950, creator=18614031015, updator=18614031015), WebsiteProps(id=1155914371218919625, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155914124811976731, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_cn_619/, createTime=1753521061949, updateTime=1753521061949, creator=18614031015, updator=18614031015)])], journalTitle=科学技术与工程, weixinUrl=null, journalUrl=null, iacademicId=null, status=0, seqNo=null, journalTitleEn=Science Technology and Engineering, journalPhotoCn=UKU/O7GSka5polgCTkbIIw==, journalPhotoEn=5hwlULoNwcbj3xUmVi9MAQ==, journalFirstLetter=S, journalRecommend=null, journalNew=null, journalCollection=null, jcrJf=null, cjcrJf=null, jcrJfStr=null, cjcrJfStr=null, submissionFirstDecision=null, sciSubjectClassification=null, casSubjectClassification=null, citeScore=null, totalCitationFrequency=null, icpCode=null, psCode=null, advertisingLicenseCode=null, copyrightInformation=null, country=null, option=null, provinceCode=null, provinceName=null, collectFlag=false), detailUrlCn=https://castjournals.cast.org.cn/joweb/kxjsygc/CN/10.12404/j.issn.1671-1815.2403403, detailUrlEn=https://castjournals.cast.org.cn/joweb/kxjsygc/EN/10.12404/j.issn.1671-1815.2403403, pdfUrlCn=https://castjournals.cast.org.cn/joweb/kxjsygc/CN/PDF/10.12404/j.issn.1671-1815.2403403, pdfUrlEn=https://castjournals.cast.org.cn/joweb/kxjsygc/EN/PDF/10.12404/j.issn.1671-1815.2403403, aliStartDate=null, aliEndDate=null, collectionFlag=false, citedCount=null, citedUrl=null, reference=null)
收藏切换
基于数据驱动的配电网典型负荷曲线分类方法
收藏切换
PDF下载
贾东梨 1 , 王帅 1 , 刘科研 1 , 陈硕 2
科学技术与工程 | 论文·自动化技术、计算机技术 2025,25(9): 3769-3777
收起
收藏切换
科学技术与工程 | 论文·自动化技术、计算机技术 2025, 25(9): 3769-3777
基于数据驱动的配电网典型负荷曲线分类方法
全屏
贾东梨1 , 王帅1, 刘科研1, 陈硕2
作者信息
  • 1 中国电力科学研究院有限公司, 北京 100192
  • 2 北京邮电大学计算机学院, 北京 100876
  • 贾东梨(1982—),女,汉族,山东烟台人,博士,教授级高级工程师。研究方向:配电网运行分析与控制。E-mail:

Data-driven Classification Method for Typical Load Curves in Distribution Networks
Dong-li JIA1 , Shuai WANG1, Ke-yan LIU1, Shuo CHEN2
Affiliations
  • 1 China Electric Power Research Institute, Beijing 100192, China
  • 2 School of Computer Science, Beijing University of Posts and Telecommunications, Beijing 100876, China
出版时间: 2025-03-28 doi: 10.12404/j.issn.1671-1815.2403403
文章导航
收藏切换

随着“双碳”战略目标和新型电力系统建设的不断推进,传统配电网逐渐向信息化、数字化和智能化的新型配电系统转变。为准确刻画并分析配电网中不同类型负荷特性,支撑配电网高效运行管控,提出了一种基于数据驱动的配电网典型负荷曲线分类方法。首先基于负荷数据,分析了配电网典型负荷的多种分类场景,并提出了包括错误率、精度和混淆矩阵等的分类场景性能评价指标;在此基础上,提出了一种基于数据驱动的配电网负荷分类方法,将24维日负荷向量转换成图片数据,并基于卷积神经网络识别负荷曲线图片,实现对配电网负荷曲线的精准分类;最后结合实际配电网负荷数据对所提方法的准确性与有效性进行了验证,并与已有方法进行了分析与对比。结果表明所提配电网典型负荷曲线分类方法具有更好的分类速度和分类精度。

数据驱动  /  负荷曲线  /  卷积神经网络  /  监督学习  /  负荷分类

With the continuous promotion of the “dual carbon” strategic goals and the construction of new power systems, traditional distribution networks are gradually transforming into information-based, digital, and intelligent new distribution systems. To accurately characterize and analyze the characteristics of different types of loads in the distribution network, and support efficient operation and control of the distribution network, a data-driven classification method for typical load curves in the distribution network was proposed. Firstly, based on load data, various classification scenarios of typical loads in the distribution network were analyzed, and performance evaluation indicators for classification scenarios including error rate, accuracy, and confusion matrix were proposed. On this basis, a data-driven load classification method for distribution networks was proposed, which converts 24 dimensional daily load vectors into image data and uses convolutional neural networks to identify load curve images, achieving accurate classification of distribution network load curves. Finally, the accuracy and effectiveness of the proposed method were verified by combining actual distribution network load data, and analyzed and compared with existing methods. The results indicate that the proposed method for classifying typical load curves in power distribution networks has better classification speed and accuracy.

data-driven  /  load curve  /  convolutional neural network  /  supervised learning  /  load classification
贾东梨, 王帅, 刘科研, 陈硕. 基于数据驱动的配电网典型负荷曲线分类方法. 科学技术与工程, 2025 , 25 (9) : 3769 -3777 . DOI: 10.12404/j.issn.1671-1815.2403403
Dong-li JIA, Shuai WANG, Ke-yan LIU, Shuo CHEN. Data-driven Classification Method for Typical Load Curves in Distribution Networks[J]. Science Technology and Engineering, 2025 , 25 (9) : 3769 -3777 . DOI: 10.12404/j.issn.1671-1815.2403403
随着“双碳”战略目标和新型电力系统建设的不断推进,传统配电网逐渐向信息化、数字化和智能化的新型配电系统转变。面向碳中和目标,能源电力系统的安全高效、绿色低碳转型及数字化智能化技术创新已成为全球发展趋势及我国能源科技创新的关键抓手[1]。配电网中用户负荷曲线也趋于多样化和复杂化,对其进行分类是需求侧管理、负荷建模等工作的重要基础,对配电系统数字化分析、运行与控制等具有重要意义[2-4]
现有配电网负荷曲线分类方法主要有无监督与有监督等方法[5]。其中无监督分类多采用K均值算法(K-means)、谱聚类等聚类方法[6-8]。Lin等[9]采用皮尔逊系数代替欧氏距离实现负荷的聚类分类,卜凡鹏等[10]提出了一种基于双层聚类距离判断方法,尽可能有效识别负荷类型。数字化和智能化的新型配电系统中,负荷种类众多,负荷曲线数量随之众多,且形态差异显著,通过调整单一的传统距离度量方式对聚类质量进行优化,未能考虑负荷曲线形态特征上的相似性,聚类效果并不理想。
有监督分类方法主要有随机森林、支持向量机以及人工神经网络等方法。梁青艳等[11]运用静态预测和动态预测相结合的方法对钢铁企业电力负荷进行建模研究。林顺富等[12]基于稀疏自编码器神经网络对负荷曲线进行分类,旨在通过自编码器挖掘负荷曲线的隐藏特性。上述神经网络方法虽然尽可能地反映出负荷数据中的特征,但是未把负荷曲线本身作为一个最大的特征进行分类。
目前直接面向用户用电负荷曲线的分类研究大多数还是以无监督聚类的方法为主。张庆新等[15] 提出了借助卡尔曼滤波对“失真数据”进行实时估计,弱化陈旧数据权值的K-means聚类和径向基网络的组合预测模型。李仲恒等[16]引入了迭代自组织数据分析方法(iterative selforganizing data analysis techniques algorithm, ISODATA),并基于统计模式识别对负荷曲线进行非监督动态聚类。朱天怡等[17]提出了一种典型的数据驱动研究架构,并总结分析了用电行为关键聚类技术。然而,上述工作涉及的方法对负荷曲线分类需要定义理想的距离或相似性函数,找到理想的相似度判别方法难度较大。
近年来一些研究将神经网络引入了电网负荷曲线分类中。董雷等[18]提出了基于自适应时空同步图卷积网络(adaptive spatiotemporal synchronized graph convolutional networks, ASTSGCN)与K-means聚类的居民用户集群短期负荷预测方法,考虑到了用户用电行为之间存在的潜在关联关系。唐子卓等[19]引入深度长短期记忆神经网络(long short-term memory, LSTM)对负荷进行分类,旨在消除类别不平衡的问题。张杰等[20]提出了一种基于数据增强和双向深度残差时间卷积网络(temporal convolutional networks, TCN)的分类方法来解决数据缺失的问题。
卷积神经网络(convolutional neural networks,CNN)是一种有强大图像处理能力的神经网络,近年来也被应用于电网各种特征的预测与分类等领域中来[13-14,21]。Imani[22]将用电负载温度输入卷积神经网络来对居民用电进行预测。de Paiva等[23]通过卷积神经网络应用于非侵入式用电监控系统来对用户用电进行识别。张铁峰等[24]总结了负荷模式提取的流程及方法,并分析了不同的数据预处理方法、聚类方法以及评价指标对聚类结果的影响。王德文等[25]提出了一种基于数据降维的负荷聚类方法,并给出了具体流程,可以用于负荷提取。
现提出一种基于卷积神经网络的有监督学习方法,用于配电网典型负荷曲线分类。首先基于负荷数据分析配电网典型负荷的多种分类场景,并提出分类场景的性能评价指标;在此基础上,提出一种基于数据驱动的配电网负荷分类方法,将24维日负荷向量转换成图片数据,并基于卷积神经网络识别负荷曲线图片,实现对配电网负荷曲线的精准分类;最后结合实际配电网负荷数据对所提方法的准确性与有效性进行验证,并与现有方法进行对比与分析,验证所提方法的优越性。
面向典型负荷场景分类应用的数据输入主要为单用户每日负荷数据,采集频率设定为每小时一次,每天采样24个点,故定义矩阵X

X=(x1,x2,…,xn)T

式(1)中:X为365×24矩阵,表示某用户全年的用电负荷数据,行向量维数为365,表示365 d,列向量维数为24,表示全天24个时间采样点的电力负荷数据,每个采样数据的单位为kW·h。
同时,定义数据集合对应的分类结果为

Y=(y1,y2,…,yn)T

式(2)中:Yn×m矩阵,表示n个由m维向量组成的结果样本集合,m维表示分类场景的数量。
采用One-Hot向量表示分类结果,若该向量分类为第n类,则Y向量的第n维为1,其他维为0,即
Vector[i]=$\left\{\begin{array}{ll}1,& \mathrm{c}\mathrm{l}\mathrm{a}\mathrm{s}\mathrm{s}=i\\ 0,& \mathrm{其}\mathrm{他}\end{array}\right.$
此时,电力负荷数据X和分类结果Y之间存在潜在规律为

Y=f(X)

通过一定方法刻画表征该潜在规律,即负荷曲线分类方法,可实现配电网典型负荷曲线的多场景分类。
引入相关指标度量负荷曲线分类方法的正确性与有效性,包括错误率、精度以及混淆矩阵等,具体如下。
(1)错误率。错误率表示采用所提负荷曲线分类方法,分类错误的样本数占样本总数的比例,其表达式为
E(f;D)=$\frac{1}{m}\stackrel{m}{\sum _{i=1}}$IA[f(xi)≠yi]
式(5)中:f为所提出的负荷曲线分类方法f(X)=Y;D为样本集;m为样本集的个数;IA(·)为指示函数,在·为真和假的时候分别取值为1和0。
(2)精度。精度表示采用所提出的负荷曲线分类方法,分类正确的样本数占样本总数的比例,其表达式为
acc(f;D)=$\frac{1}{m}\stackrel{m}{\sum _{i=1}}$IA[f(xi)≠yi]=1-E(f;D)
式(6)中:f为所提出的负荷曲线分类方法f(X)=Y;D为样本集;m为样本集的个数;IA(·)为指示函数,在·为真和假的时候分别取值为1和0。
(3)混淆矩阵。所述分类方法的准确率定义为
accuracy=$\frac{\stackrel{n}{\sum _{i=1}}cfm\left[i\right]\left[j\right]}{\stackrel{n}{\sum _{i=1}}\stackrel{n}{\sum _{j=1}}cfm\left[i\right]\left[j\right]}$
式(7)中:cfm为混淆矩阵;cfm[i][j]为真实类别为i且预测类别为j的样本数量。
通过上述3个指标,能够有效评价所提负荷曲线分类方法的优劣性能。
配电网负荷数据涵盖各个用户一年中每天负荷信息,其均符合同一分类结果的特性。为提升分类方法效果,设计基于不同天数负荷数据的多类型分类器,比较并选择其中较好的分类方法。
数据驱动的典型负荷曲线分类方法整理流程如图1所示。首先对配电网负荷数据进行图片化,然后使用卷积神经网络对图片进行特征提取,最后根据负荷数据图的不同特征进行分类。
配电网数据属于一种时间序列数据,针对时间序列分类问题往往有两种主要的方法,基于距离分类与基于特征分类。由于配电网日负荷数据仅包含24维,大多数特征法无法发挥作用,因此传统方法多为基于原始数据的距离分类方法。
将24维的日负荷向量转换成图片数据,然后基于卷积神经网络识别负荷曲线图片的模式,完成对配电网负荷曲线进行分类。
针对任意用户,其所有的日数据都反映同一个Y,因此卷积神经网络的输入应只包括一天的数据,定义为
Xm=(${d}_{1}^{m}$,${d}_{2}^{m}$,…,${d}_{n}^{m}$)
式(8)中:Xm为第m天的所有电力负荷数据;${d}_{n}^{m}$为一个24维的向量,表示第n名用户在第m天的电力负荷数据。
定义卷积神经网络模型为

f[I(Xm)]=Y

式(9)中:I(·)为绘图函数,针对每一个24维的数据都可以绘制成一个固定长宽的折线图;f为卷积神经网络分类器,其输出为分类结果。
负荷图片可以看作一个三维的向量数据,即宽度、高度和通道数,通道数一般为3个。卷积操作是卷积神经网络的核心。通过在负荷图片上滑动卷积核(也称为过滤器),提取局部特征。卷积操作有助于捕捉图像中的空间关系,减少参数数量。卷积操作会生成一个特征图(feature map)或者激活图,其中每个元素代表原始图像中的某种特征。特征图能够放大负荷的特征,同时会显示出哪些地方最有可能有曲线。从而保证后续负荷分类的正确性。
通过卷积运算提取负荷图像特征。其具体过程如下:首先负荷图片通过卷积层,实现对负荷图片特征的抽取,然后通过激活层增加网络的非线性特征,最后利用池化层降低特征图的空间尺寸,减少计算量,同时保持重要特征。
在卷积层中,执行的基本操作是卷积运算,其数学表达式为
F(i,j)=$\stackrel{a}{\sum _{u=-a}}\stackrel{b}{\sum _{v=-b}}$I(i+u,j+v)K(u,v)
式(10)中:F(i,j)为卷积后的输出特征图在位置(i,j)的值;I(i,j)为输入图像;K(u,v)为卷积核的权重矩阵;ab为卷积核的维度。
在激活层中,采用了Relu(rectified linear unit)实现激活操作,其具体的数学表达式为

Relu(x)=max(0,x)

通过激活层的处理,大大增加了卷积层所抽取的负荷图片特征的非线性特性。而接着使用池化层实现对特征图的大小调整以适应后面的网络输入,其数学表达式为

P(i,j)=maxu,vWF(i+u,j+v)

式(12)中:P(i,j)为池化后的输出,它在窗口W中遍历,选择最大值。
总的来说,通过采用卷积神经网络对负荷图片进行特征抽取,使得后续网络更适用于处理负荷图像等网格状数据,同时降低了参数的数量,提高了学习效率和泛化能力。卷积后得到的特征图能够让模型在更高的层次观察负荷曲线图,对于复杂的负荷曲线,可以使用多层卷积来识别高层次的负荷图片特征。同时多层卷积可以扩大模型对负荷图片的感知视野。然后使用池化层对负荷图片的特征图进行降维,减小计算量,并提取主要特征。将卷积层的输出拉平成一维向量,并连接到一个全连接层。全连接层用于整合前面层次提取的特征,进行分类或回归任务。
在卷积网络的特征抽取之后,一般可以直接通过一个全连接层进行分类,从而实现图片的分类。但是经过课题组的实验,发现直接采用全连接层无法实现较好的图片分类,说明单一的CNN特征抽取难以应对负荷图片复杂的特征情况,为此拟引入具有堆叠和残差结构的特征抽取结构。
基于图片的卷积网络所具有的堆叠层和残差连接的结构可以有效捕获图像数据中的特征。利用卷积网络的特征提取能力,能够更好地捕获不同用户间负荷曲线图像的相似性和模式。通过深层结构,网络能够学习到负荷曲线图像中的抽象特征,这些特征可能与不同用户之间的使用习惯、负荷类型或特定负荷变化相关。残差连接允许信息更快速地传递和学习,易于捕获负荷曲线中的长期依赖性和变化模式。
采用的34层的残差神经网络如图3所示。残差网络ResNet的经典网络结构有:ResNet-18、ResNet-34、ResNet-50、ResNet-101、ResNet-152。这些结构的选择是经过实验和调整的结果,目的是在深度学习任务中找到一种平衡,既能够提供足够的深度以捕捉复杂的特征,又能够在训练中保持相对较高的性能。
负荷预测是负荷数据分析中常用的分析方法。探究本文模型对于负荷预测方面的帮助。对于相同的负荷类别,它们的特征往往非常相似。因此,提出了在负荷分类基础上的负荷预测。
首先使用本文模型对负荷数据进行分类。然后对每一类负荷单独训练预测模型(如2层的LSTM)。之后,对于输入的负荷数据使用提出的分类方法得到准确的类别,最后使用对应的预测模型进行预测,能够有效提高预测效果。
仿真数据基于浙江省2020年的配电网负荷数据,共包含14 976个用户,共包含16个分类,分别为全服务餐厅(full service restaurant),医院(hospital),新建大型旅馆(large hotel new),大型办公室(large office),中型办公室(medium office),多层公寓(midrise apartment),门诊(out patient),小学(primary school),快餐厅(quick service restaurant),中学(secondary school),小型旅馆(small hotel),小型办公室(small office),零售店(stand-alone retail),购物中心(strip mall),超市(super market),仓库(warehouse)。每个分类含936个用户,经过清洗抛弃掉缺失值后可用用户为14 958个,随机选择第120天至第127天共计104 706条日负荷数据用于仿真分析。
为了保证图片的趋势不丢失,在绘制折线图时应确保每个折线图按照最大值和最小值来绘制。
不同用户负荷曲线分类图如图4所示,可知,不同分类的折线图有着相似的模式,例如中型办公室和全服务餐厅等均呈现中午和下午高峰,晚上低谷形态,然而不同的用户的使用阈值有着很大的差异,且曲线图的趋势却相近,传统基于距离的分类方法可能难以学习该规律。
为提升模型的泛化性能,采取一段时间的数据进行训练,然后基于其他天数的数据确定预测结果。仿真中采用前120 d的数据进行训练,取其中80%作为训练集,20%作为测试集。然后使用第120天至第126天的数据作为输入,面向第127天的数据进行预测。
分别对比不同的神经网络参数对配电网负荷图片识别结果的差异,以及对比支持向量机(support vector machine,SVM)、随机森林(random forest,RF)、一维卷积网络和所提出的卷积神经网络等方法的分类效果,验证所提方法的有效性和优越性。
不同的卷积网络间往往有很大的区别,为了能够得到更好的结果,对多种网络进行了调优,图5反映了多种网络的训练结果,都采用学习率为0.1的动量法进行训练,训练机器都为11 GB的Nvidia Geforce 1080 Ti,以及TensorFlow进行,具体的网络结构如表1所示。
首先对比了VGG-16、ResNet-18、ResNet-34和ResNet-50等不同的卷积神经网络对负荷图片识别的影响,如图5所示,根据结果选取效果最好的网络,以供后续分类使用。
图5展示了不同网络的精度变化,可以看出ResNet-18和ResNet-34收敛最快,并在后续的迭代中保持最高的训练精度。表2是不同模型收敛后在测试集上的准确度,可以看到残差网络有较好的效果,同时ResNet-34层网络效果最好。
34层的残差神经网络在图像分类问题中有着很好的表现,实验通过该网络与其他的分类方法进行横向对比,所用网络主要参考了VGG和ResNet的设计。
首先针对残差神经网络进行训练,采用学习率为0.1的动量法去训练,在第5轮就达到了收敛,其训练轮数与准确度的关系如图6所示,其中With Voting表示数据,训练采用一块11 GB的Nvidia Geforce 1080 Ti进行,使用TensorFlow进行,训练的batch size 为32。最终得到的单一模型在测试集上准确度为90.3%,使用一周的数据进行投票的结果为94.1%。高于其他传统模型。
表3是不同的分类器的准确度,测试数据表示测试集上的准确度,周数据投票表示用一周的数据进行投票后的结果,可以看到卷积神经网络在测试集和投票中的准确率都高于传统的随机森林方法。
随着分类模型的发展,可以将图片特征提取模块作为编码器连接到分类器的前面,实现更加优秀的分类效果。
探究提出的模型对于负荷预测的提升效果。使用上述数据集,将实验分为两组。一组将数据分为16类,对于每一类分别训练预测模型,使用提出的分类模型对输入的负荷进行分类,然后选择对应的预测模型进行预测。二是使用数据集整体训练预测模型,直接进行预测。
预测模型使用LSTM,单元数为256,层数为2 层。训练采用一块11GB的Nvidia Geforce 1080 Ti进行,使用TensorFlow进行训练,batch size 为32。部分预测结果如图7所示。
评价指标采用平均绝对百分比误差(mean absolute percentage error,MAPE),预测效果评价如表4所示。
MAPE=$\frac{1}{n}\stackrel{n}{\sum _{i=1}}\left|\frac{y-\stackrel{\wedge }{y}}{y}\right|$×100%
式(10)中:n为预测的数据个数;y为实际负荷值;$\stackrel{\wedge }{y}$为预测负荷值。
34层的神经网络在测试集以及投票数据集上都有优于传统机器学习方法的效果,可见将神经网络应用于配电网负荷曲线的折线图上有着不错的效果。同时神经网络可以很方便地进行迁移学习,通过现有模型可以对不同地区的配电网负荷曲线数据进行迁移学习,也可以迁移到不同聚类问题上,较传统方法有更大的扩展性。
在实际的数字孪生应用中,本文算法将作为先导算法用于对不同负荷类型的判断,负荷预测算法将构建在当前算法的分类结果之上,针对不同类别的负荷特点,微调预测算法的参数,从而实现更好的数字孪生映射。
电力负荷数据数字孪生是针对现实符合数据的数字化映射,所提出的负荷分类算法可为为数字孪生中的数据精准预测提供支撑,主要体现为通过精确划分不同负荷类别,从而使得负荷预测可根据不同类别进行精准的预测。
通过对配电网负荷曲线数据进行分析,提出了利用卷积神经网络对配电网负荷曲线进行分类的方法。通过对比实验,得到了相对传统的随机森林法更好的效果。可见卷积神经网络除了对于图像数据外,对于时间序列数据也有一定的作用,甚至可能高于传统的机器学习方法。卷积神经网络目前主要的影响领域还是图片数据,而对于大量的时间序列数据往往还是依赖传统的机器学习方法去处理,但实验表明卷积神经网络在处理这些时间序列的时候也有着不错的表现,因此卷积网络在时间序列数据上仍有着很大的探索空间。而且所提出的模型还可以用于提升负荷预测的效果。
  • 国家电网有限公司总部科技项目(5400-202255154A-1-1-ZN)
参考文献 引证文献
排序方式:
[1]
康重庆, 杜尔顺, 郭鸿业, . 新型电力系统的六要素分析[J]. 电网技术, 2023, 47(5): 1741-1750.
Kang Chongqing, Du Ershun, Guo Hongye, et al. Primary explo-ration of six essential factors in new power system[J]. Power System Technology, 2023, 47(5): 1741-1750.
[2]
刘洋, 王剑, 唐明, . 基于Hadoop分布式计算的混合神经网络负荷分类模型[J]. 科学技术与工程, 2023, 23(4): 1549-1556.
Liu Yang, Wang Jian, Tang Ming, et al. Hybrid neural network load classification model based on Hadoop distributed platform[J]. Science Technology and Engineering, 2023, 23(4): 1549-1556.
[3]
田世明, 王蓓蓓, 张晶. 智能电网条件下的需求响应关键技术[J]. 中国电机工程学报, 2014, 34(22): 3576-3589.
Tian Shiming, Wang Beibei, Zhang Jing. Key technologies for demand response under smart grid conditions[J]. Proceedings of the Chinese Society for Electrical Engineering, 2014, 34(22): 3576-3589.
[4]
李瑞, 柳长安, 王彦平, . 基于多生理信息迁移学习的脑力负荷分类[J]. 科学技术与工程, 2022, 22(14): 5555-5561.
Li Rui, Liu Changan, Wang Yanping, et al. Classification of mental workload based on multi-physiological information transfer learning[J]. Science Technology and Engineering, 2022, 22(14): 5555-5561.
[5]
钱育树, 孔钰婷, 黄聪. 电力负荷预测研究综述[J]. 四川电力技术, 2023, 46(4): 37-43.
Qian Yushu, Kong Yuting, Huang Cong. A review of research on power load forecasting[J]. Sichuan Electric Power Technology, 2023, 46(4): 37-43.
[6]
赵忠啟, 常喜强, 樊艳芳, . 基于自编码器的电力负荷聚类分析[J]. 科学技术与工程, 2021, 21(32): 13737-13743.
Zhao Zhongqi, Chang Xiqiang, Fan Yanfang, et al. Clustering analysis of power load curve based on auto-encoder[J]. Science Technology and Engineering, 2021, 21(32): 13737-13743.
[7]
蔡改贫, 宋佳, 罗小燕, . 基于相空间重构和PSO-K-means的球磨机负荷状态识别方法[J]. 科学技术与工程, 2023, 23(10): 4126-4134.
Cai Gaipin, Song Jia, Luo Xiaoyan, et al. Load state identification method for ball mills based on phase space reconstruction and PSO-K-means[J]. Science Technology and Engineering, 2023, 23(10): 4126-4134.
[8]
吴阳. 基于谱聚类和随机矩阵理论的中小负荷用户用电行为特征分析[D]. 吉林: 东北电力大学, 2023.
Wu Yang. Analysis of electricity consumption behavior characteristics of small and medium load users based on spectral clustering and random matrix theory[D]. Jilin: Northeast Electric Power University, 2023.
[9]
Lin S, Li F, Tian E, et al. Clustering load profiles for demand response applications[J]. IEEE Transactions on Smart Grid, 2017, 10(2): 1599-1607.
[10]
卜凡鹏, 陈俊艺, 张琪祁, . 一种基于双层迭代聚类分析的负荷模式可控精细化识别方法[J]. 电网技术, 2018, 42(3): 903-910.
Bu Fanpeng, Chen Junyi, Zhang Qiqi, et al. A controllable refined recognition method of electrical load pattern based on bilayer iterative clustering analysis[J]. Power System Technology, 2018, 42(3): 903-910.
[11]
梁青艳, 孙彦广. 钢铁企业电力负荷动态预测建模问题的应用研究[J]. 科学技术与工程, 2018, 18(16): 44-54.
Liang Qingyan, Sun Yanguang. Application research on dynamic forecasting modeling of power load in steel enterprises[J]. Science Technology and Engineering, 2018, 18(16): 44-54.
[12]
林顺富, 顾乡, 汤继开, . 基于稀疏自动编码器神经网络的负荷曲线分类方法[J]. 电网技术, 2020, 44(9): 3508-3515.
Lin Shunfu, Gu Xiang, Tang Jikai, et al. Power load profile classification method based on neural network of sparse automatic encoder[J]. Power System Technology, 2020, 44(9): 3508-3515.
[13]
徐冰冰, 岑科廷, 黄俊杰, . 图卷积神经网络综述[J]. 计算机学报, 2020, 43(5): 755-780.
Xu Bingbing, Cen Keting, Huang Junjie, et al. A review of graph convolutional neural networks[J]. Chinese Journal of Computers, 2020, 43(5): 755-780.
[14]
Garimella M, Bhan M. Image labeling using convolutional neural network[C]// 2023 International Conference on Network, Multimedia and Information Technology (NMITCON). Bengaluru: Nitte Meenakshi Inst of Tech, 2023: 1-5.
[15]
张庆新, 崔展博, 马睿, . 基于K-means聚类与径向基神经(RBF)网络的电力系统日负荷预测[J]. 科学技术与工程, 2013, 13(34): 10177-10181.
Zhang Qingxin, Cui Zhanbo, Ma Rui, et al. Daily load forecasting of power system based on K-means clustering and radial basis function neural network[J]. Science Technology and Engineering, 2013, 13(34): 10177-10181.
[16]
李仲恒, 刘蓉晖. 基于ISODATA的电力负荷曲线分类[J]. 上海电力大学学报, 2019, 35(4): 327-332.
Li Zhongheng, Liu Ronghui. A load curve clustering algorithm based on ISODATA[J]. Journal of Shanghai University of Electric Power, 2019, 35(4): 327-332.
[17]
朱天怡, 艾芊, 贺兴, . 基于数据驱动的用电行为分析方法及应用综述[J]. 电网技术, 2020, 44(9): 3497-3507.
Zhu Tianyi, Ai Qian, He Xing, et al. An overview of data-driven electricity consumption behavior analysis method and application[J]. Power System Technology, 2020, 44(9): 3497-3507.
[18]
董雷, 陈振平, 韩富佳, . 基于图卷积神经网络与K-means聚类的居民用户集群短期负荷预测[J]. 电网技术, 2023, 47(10): 4291-4301.
Dong Lei, Chen Zhenping, Han Fujia, et al. Short-term load forecasting of residential user groups based on graph convolutional neural network and K-means clustering[J]. Power System Technology, 2023, 47(10): 4291-4301.
[19]
唐子卓, 刘洋, 许立雄, . 基于负荷数据频域特征和LSTM网络的类别不平衡负荷典型用电模式提取方法[J]. 电力建设, 2020, 41(8): 17-24.
Tang Zizhuo, Liu Yang, Xu Lixiong, et al. Imbalanced-load pattern extraction method based on frequency domain characteristics of load data and LSTM network[J]. Electric Power Construction, 2020, 41(8): 17-24.
[20]
张杰, 刘洋, 李文峰, . 基于两阶段数据增强和双向深度残差TCN的用户负荷曲线分类方法[J]. 电力建设, 2022, 43(2): 89-97.
Zhang Jie, Liu Yang, Li Wenfeng, et al. Power load curve identification method based on two-phase data enhancement and bi-directional deep residual TCN[J]. Electric Power Construction, 2022, 43(2): 89-97.
[21]
Xiong J, Hong H, Xu C, et al. Graph convolutional network based electricity demand forecasting in power distribution networks[C]//2022 International Conference on Artificial Intelligence and Autonomous Robot Systems (AIARS). Bristol: IEEE, 2022: 104-109.
[22]
Imani M. Electrical load-temperature CNN for residential load fore-casting[J]. Energy, 2021, 227: 120480.
[23]
de Paiva P D, Castro A R G. Convolutional neural network applied to the identification of residential equipment in non-intrusive load monitoring systems[C]// 3rd International Conference on Artificial Intelligence and Applications. Tokyo: AIAT, 2017: 11-21.
[24]
张铁峰, 顾明迪. 电力用户负荷模式提取技术及应用综述[J]. 电网技术, 2016, 40(3): 804-811.
Zhang Tiefeng, Gu Mingdi. Overview of electricity customer load pattern extraction technology and its application[J]. Power System Technology, 2016, 40(3): 804-811.
[25]
王德文, 周昉昉. 基于无监督极限学习机的用电负荷模式提取[J]. 电网技术, 2018, 42(10): 3393-3400.
Wang Dewen, Zhou Fangfang. Extraction of electricity consumption load pattern based on unsupervised extreme learning machine[J]. Power System Technology, 2018, 42(10): 3393-3400.
2025年第25卷第9期
PDF下载
404
164
引用本文
BibTeX
文章信息
doi: 10.12404/j.issn.1671-1815.2403403
  • 接收时间:2024-05-08
  • 首发时间:2025-07-09
  • 出版时间:2025-03-28
补充材料
相关文章
文章信息
作者
出版历史
  • 收稿日期:2024-05-08
  • 修回日期:2024-12-27
基金
国家电网有限公司总部科技项目(5400-202255154A-1-1-ZN)
作者信息
    1 中国电力科学研究院有限公司, 北京 100192
    2 北京邮电大学计算机学院, 北京 100876
参考文献
分享链接
https://castjournals.cast.org.cn/joweb/kxjsygc/CN/10.12404/j.issn.1671-1815.2403403
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