Article(id=1190337959950909999, tenantId=1146029695717560320, journalId=1189987059142926344, issueId=1190337956201202212, articleNumber=null, orderNo=null, doi=10.19457/j.1001-2095.dqcd25612, pmid=null, cstr=null, oa=null, hot=null, price=null, onlineType=0, articleFormat=0, articleType=null, articleTypeStr=research-article, receivedDate=1705593600000, receivedDateStr=2024-01-19, revisedDate=1712505600000, revisedDateStr=2024-04-08, acceptedDate=null, acceptedDateStr=null, onlineDate=1761728285061, onlineDateStr=2025-10-29, pubDate=1755619200000, pubDateStr=2025-08-20, doiRegisterDate=null, doiRegisterDateStr=null, onlineIssueDate=1761728285061, onlineIssueDateStr=2025-10-29, onlineJustAcceptDate=null, onlineJustAcceptDateStr=null, onlineFirstDate=null, onlineFirstDateStr=null, sourceXml=null, magXml=null, createTime=1761728285061, creator=13701087609, updateTime=1761728285061, updator=13701087609, issue=Issue{id=1190337956201202212, tenantId=1146029695717560320, journalId=1189987059142926344, year='2025', volume='55', issue='8', pageStart='3', pageEnd='96', issueExtLink='null', onlineDate='null', pubDate='null', beforeIssueId=null, nextIssueId=null, price=null, status=1, issueComplete=1, articleOrder=1, issueType=1, specialIssue=0, createTime=1761728284168, creator=13701087609, updateTime=1761728464442, updator=13701087609, preIssue=null, nextIssue=null, ext={EN=IssueExt(id=1190338712388079738, tenantId=1146029695717560320, journalId=1189987059142926344, issueId=1190337956201202212, language=EN, specialIssueTitle=, coverIllustrator=null, specialIssueEditor=, specialIssueAbout=), CN=IssueExt(id=1190338712388079739, tenantId=1146029695717560320, journalId=1189987059142926344, issueId=1190337956201202212, language=CN, specialIssueTitle=, coverIllustrator=null, specialIssueEditor=, specialIssueAbout=)}, issueFiles=null}, startPage=89, endPage=96, ext={EN=ArticleExt(id=1190337960173208112, articleId=1190337959950909999, tenantId=1146029695717560320, journalId=1189987059142926344, language=EN, title=Short-term Electrical Load Forecasting Method Based on AT-LSTM and Stacking Ensemble Learning, columnId=null, journalTitle=Electric Drive, columnName=null, runingTitle=null, highlight=null, articleAbstract=

Accurate short-term electrical load forecasting is of great significance for the design and optimization scheduling of integrated energy systems(IES). However,the load data in real integrated energy systems are low-quality and fluctuating,so the forecasting accuracy of existing prediction models is low. A short-term electrical load forecasting method based on attention-based long short-term memory(AT-LSTM)and Stacking learning was proposed. Under the framework of Stacking ensemble learning,AT-LSTM,random forest and decision tree were ensembled to forecast short-term electrical load which can make up for the low prediction accuracy of a single model. Based on the exploratory analysis results of data,the data feature engineering model was constructed to input features,and this prediction method was used for short-term electricity load prediction. The experimental results of the integrated energy system in Beijing show that compared to other algorithms,the proposed method has a maximum prediction error reduction of 24.8%.

, 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=Baorui ZHANG, Jingtao WANG, Yao TAN, Zhiguang WANG, Tang LI, Yang SHI), CN=ArticleExt(id=1190338317922173908, articleId=1190337959950909999, tenantId=1146029695717560320, journalId=1189987059142926344, language=CN, title=基于AT-LSTM及集成学习的短期电负荷预测方法, columnId=1190325328129131131, journalTitle=电气传动, columnName=综合能源与现代电网, runingTitle=null, highlight=null, articleAbstract=

精确的短期电负荷预测对综合能源系统的规划和优化调度具有重要意义,然而真实综合能源系统中的负荷数据由于数据质量低、波动趋势多变等情况,预测精度较低。为此,提出一种带注意力机制的长短时记忆(AT-LSTM)及集成学习的短期电负荷预测方法,在Stacking集成学习的框架下,对AT-LSTM、随机森林、决策树进行集成并预测短期电负荷,弥补单一模型预测精度较低的不足。具体地,基于对真实综合能源系统中的数据探索性分析结果,通过数据综合特征工程构造模型输入特征后,采用该预测方法进行短期电负荷预测。以北京某综合能源系统为例的实验结果表明,相比长短期记忆网络、决策树及随机森林等其他算法,所提方法预测误差最高降低了24.8%。

, correspAuthors=null, authorNote=null, correspAuthorsNote=null, copyrightStatement=null, copyrightOwner=null, extLink=null, articleAbsUrl=null, sourceXml=rIjoislVrggyxzg7dNFB9w==, magXml=gUkUGMDxyGZLjzlCfNpU0g==, pdfUrl=null, pdf=6r9QnPcHB8v6eRLMlqHNng==, pdfFileSize=5131461, pdfExtLink=null, richHtmlUrl=null, mobilePdfUrl=null, reviewReport=null, pdfFirstPage=null, abstractGraph=RHODV+djTKm9M6qPH4cotA==, abstractGraphContent=null, abstractVideo=null, citation=null, cebUrl=null, magXmlContent=DDq5jPzx7rRIISvNhpF2DA==, mapNumber=null, authorCompany=null, fund=null, authors=

张保瑞(1976—),男,硕士,高级工程师,主要研究方向为低碳能源、建筑节能、供冷供热,Email:

, authorsList=张保瑞, 王靖韬, 谭瑶, 王志光, 李唐, 史洋)}, authors=[Author(id=1190338318320632797, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, orderNo=0, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=sjzlirui@126.com, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1190338318383547360, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, authorId=1190338318320632797, language=EN, stringName=Baorui ZHANG, firstName=Baorui, middleName=null, lastName=ZHANG, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1. State Grid Xiong'an Comprehensive Energy Services Co.,Ltd.,Baoding 071700,Hebei,China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1190338318459044833, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, authorId=1190338318320632797, language=CN, stringName=张保瑞, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1.国网雄安综合能源服务有限公司,河北 保定 071700, bio={"content":"

张保瑞(1976—),男,硕士,高级工程师,主要研究方向为低碳能源、建筑节能、供冷供热,Email:

"}, bioImg=null, bioContent=

张保瑞(1976—),男,硕士,高级工程师,主要研究方向为低碳能源、建筑节能、供冷供热,Email:

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1190338318161249238, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, xref=null, ext=[AuthorCompanyExt(id=1190338318169637847, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, companyId=1190338318161249238, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1. State Grid Xiong'an Comprehensive Energy Services Co.,Ltd.,Baoding 071700,Hebei,China), AuthorCompanyExt(id=1190338318178026456, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, companyId=1190338318161249238, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1.国网雄安综合能源服务有限公司,河北 保定 071700)])]), Author(id=1190338318526153699, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, 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=1190338318593262565, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, authorId=1190338318526153699, language=EN, stringName=Jingtao WANG, firstName=Jingtao, middleName=null, lastName=WANG, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2. Beijing NARI Digital Technology Co.,Ltd.,Beijing 100193,China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1190338318668760038, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, authorId=1190338318526153699, language=CN, stringName=王靖韬, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2.北京南瑞数字技术有限公司,北京 100193, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1190338318245135321, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, xref=null, ext=[AuthorCompanyExt(id=1190338318253523930, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, companyId=1190338318245135321, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2. Beijing NARI Digital Technology Co.,Ltd.,Beijing 100193,China), AuthorCompanyExt(id=1190338318261912539, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, companyId=1190338318245135321, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2.北京南瑞数字技术有限公司,北京 100193)])]), Author(id=1190338318740063208, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, 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=1190338318798783466, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, authorId=1190338318740063208, language=EN, stringName=Yao TAN, firstName=Yao, middleName=null, lastName=TAN, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2. Beijing NARI Digital Technology Co.,Ltd.,Beijing 100193,China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1190338318857503723, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, authorId=1190338318740063208, language=CN, stringName=谭瑶, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2.北京南瑞数字技术有限公司,北京 100193, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1190338318245135321, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, xref=null, ext=[AuthorCompanyExt(id=1190338318253523930, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, companyId=1190338318245135321, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2. Beijing NARI Digital Technology Co.,Ltd.,Beijing 100193,China), AuthorCompanyExt(id=1190338318261912539, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, companyId=1190338318245135321, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2.北京南瑞数字技术有限公司,北京 100193)])]), Author(id=1190338318912029677, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, 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=1190338318987527151, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, authorId=1190338318912029677, language=EN, stringName=Zhiguang WANG, firstName=Zhiguang, middleName=null, lastName=WANG, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2. Beijing NARI Digital Technology Co.,Ltd.,Beijing 100193,China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1190338319058830320, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, authorId=1190338318912029677, language=CN, stringName=王志光, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2.北京南瑞数字技术有限公司,北京 100193, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1190338318245135321, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, xref=null, ext=[AuthorCompanyExt(id=1190338318253523930, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, companyId=1190338318245135321, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2. Beijing NARI Digital Technology Co.,Ltd.,Beijing 100193,China), AuthorCompanyExt(id=1190338318261912539, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, companyId=1190338318245135321, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2.北京南瑞数字技术有限公司,北京 100193)])]), Author(id=1190338319121744882, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, orderNo=4, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1190338319193048052, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, authorId=1190338319121744882, language=EN, stringName=Tang LI, firstName=Tang, middleName=null, lastName=LI, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1. State Grid Xiong'an Comprehensive Energy Services Co.,Ltd.,Baoding 071700,Hebei,China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1190338319251768309, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, authorId=1190338319121744882, language=CN, stringName=李唐, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1.国网雄安综合能源服务有限公司,河北 保定 071700, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1190338318161249238, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, xref=null, ext=[AuthorCompanyExt(id=1190338318169637847, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, companyId=1190338318161249238, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1. State Grid Xiong'an Comprehensive Energy Services Co.,Ltd.,Baoding 071700,Hebei,China), AuthorCompanyExt(id=1190338318178026456, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, companyId=1190338318161249238, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1.国网雄安综合能源服务有限公司,河北 保定 071700)])]), Author(id=1190338319318877175, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, orderNo=5, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1190338319373403129, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, authorId=1190338319318877175, language=EN, stringName=Yang SHI, firstName=Yang, middleName=null, lastName=SHI, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1. State Grid Xiong'an Comprehensive Energy Services Co.,Ltd.,Baoding 071700,Hebei,China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1190338319432123386, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, authorId=1190338319318877175, language=CN, stringName=史洋, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1.国网雄安综合能源服务有限公司,河北 保定 071700, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1190338318161249238, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, xref=null, ext=[AuthorCompanyExt(id=1190338318169637847, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, companyId=1190338318161249238, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1. State Grid Xiong'an Comprehensive Energy Services Co.,Ltd.,Baoding 071700,Hebei,China), AuthorCompanyExt(id=1190338318178026456, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, companyId=1190338318161249238, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1.国网雄安综合能源服务有限公司,河北 保定 071700)])])], keywords=[Keyword(id=1190338319528592379, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, language=EN, orderNo=1, keyword=short-term electrical load forecasting), Keyword(id=1190338319578924028, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, language=EN, orderNo=2, keyword=attention-based long short-term memory(AT-LSTM)), Keyword(id=1190338319641838589, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, language=EN, orderNo=3, keyword=Stacking ensemble learning), Keyword(id=1190338319734113278, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, language=EN, orderNo=4, keyword=integrated energy system(IES)), Keyword(id=1190338319805416447, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, language=EN, orderNo=5, keyword=feature engineering), Keyword(id=1190338319864136704, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, language=CN, orderNo=1, keyword=短期电负荷预测), Keyword(id=1190338319927050240, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, language=CN, orderNo=2, keyword=基于注意力机制的长短期记忆), Keyword(id=1190338319998353409, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, language=CN, orderNo=3, keyword=集成学习), Keyword(id=1190338320090628098, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, language=CN, orderNo=4, keyword=综合能源系统), Keyword(id=1190338320153542659, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, language=CN, orderNo=5, keyword=特征工程)], refs=[Reference(id=1190338321931927579, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, doi=null, pmid=null, pmcid=null, year=2017, volume=37, issue=14, pageStart=4060, pageEnd=4069, url=null, language=null, rfNumber=[1], rfOrder=0, authorNames=程耀华, 张宁, 康重庆, journalName=中国电机工程学报, refType=null, unstructuredReference=程耀华, 张宁, 康重庆, 等. 低碳多能源系统的研究框架及展望[J]. 中国电机工程学报, 2017, 37(14):4060-4069., articleTitle=低碳多能源系统的研究框架及展望, refAbstract=null), Reference(id=1190338322020007964, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, doi=null, pmid=null, pmcid=null, year=2017, volume=37, issue=14, pageStart=4060, pageEnd=4069, url=null, language=null, rfNumber=[1], rfOrder=1, authorNames=CHENG Yaohua, ZHANG Ning, KANG Chongqing, journalName=Proceedings of the CSEE, refType=null, unstructuredReference=CHENG Yaohua, ZHANG Ning, KANG Chongqing, et al. Research framework and prospects of low-carbon multiple energy systems[J]. Proceedings of the CSEE, 2017, 37(14):4060-4069., articleTitle=Research framework and prospects of low-carbon multiple energy systems, refAbstract=null), Reference(id=1190338322087116829, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, doi=null, pmid=null, pmcid=null, year=2009, volume=33, issue=2, pageStart=1, pageEnd=7, url=null, language=null, rfNumber=[2], rfOrder=2, authorNames=康重庆, 陈启鑫, 夏清, journalName=电网技术, refType=null, unstructuredReference=康重庆, 陈启鑫, 夏清. 低碳电力技术的研究展望[J]. 电网技术, 2009, 33(2):1-7., articleTitle=低碳电力技术的研究展望, refAbstract=null), Reference(id=1190338322145837086, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, doi=null, pmid=null, pmcid=null, year=2009, volume=33, issue=2, pageStart=1, pageEnd=7, url=null, language=null, rfNumber=[2], rfOrder=3, authorNames=KANG Chongqing, CHEN Qixin, XIA Qing, journalName=Power System Technology, refType=null, unstructuredReference=KANG Chongqing, CHEN Qixin, XIA Qing. Prospects of low-carbon electricity[J]. Power System Technology, 2009, 33(2):1-7., articleTitle=Prospects of low-carbon electricity, refAbstract=null), Reference(id=1190338322212945951, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, doi=null, pmid=null, pmcid=null, year=2021, volume=47, issue=1, pageStart=23, pageEnd=32, url=null, language=null, rfNumber=[3], rfOrder=4, authorNames=罗凤章, 张旭, 杨欣, journalName=高电压技术, refType=null, unstructuredReference=罗凤章, 张旭, 杨欣, 等. 基于深度学习的综合能源配电系统负荷分析预测[J]. 高电压技术, 2021, 47(1):23-32., articleTitle=基于深度学习的综合能源配电系统负荷分析预测, refAbstract=null), Reference(id=1190338322271666208, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, doi=null, pmid=null, pmcid=null, year=2021, volume=47, issue=1, pageStart=23, pageEnd=32, url=null, language=null, rfNumber=[3], rfOrder=5, authorNames=LUO Fengzhang, ZHANG Xu, YANG Xin, journalName=High Voltage Engineering, refType=null, unstructuredReference=LUO Fengzhang, ZHANG Xu, YANG Xin, et al. Load analysis and prediction of integrated energy distribution system based on deep learning[J]. High Voltage Engineering, 2021, 47(1):23-32., articleTitle=Load analysis and prediction of integrated energy distribution system based on deep learning, refAbstract=null), Reference(id=1190338322330386465, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, doi=null, pmid=null, pmcid=null, year=2021, volume=45, issue=5, pageStart=63, pageEnd=70, url=null, language=null, rfNumber=[4], rfOrder=6, authorNames=孙庆凯, 王小君, 张义志, journalName=电力系统自动化, refType=null, unstructuredReference=孙庆凯, 王小君, 张义志, 等. 基于LSTM和多任务学习的综合能源系统多元负荷预测[J]. 电力系统自动化, 2021, 45(5):63-70., articleTitle=基于LSTM和多任务学习的综合能源系统多元负荷预测, refAbstract=null), Reference(id=1190338322401689634, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, doi=null, pmid=null, pmcid=null, year=2021, volume=45, issue=5, pageStart=63, pageEnd=70, url=null, language=null, rfNumber=[4], rfOrder=7, authorNames=SUN Qingkai, WANG Xiaojun, ZHANG Yizhi, journalName=Automation of Electric Power Systems, refType=null, unstructuredReference=SUN Qingkai, WANG Xiaojun, ZHANG Yizhi, et al. Multiple load prediction of integrated energy system based on long short term memory and multi-task learning[J]. Automation of Electric Power Systems, 2021, 45(5):63-70., articleTitle=Multiple load prediction of integrated energy system based on long short term memory and multi-task learning, refAbstract=null), Reference(id=1190338322489770019, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, doi=null, pmid=null, pmcid=null, year=2019, volume=43, issue=7, pageStart=2, pageEnd=13, url=null, language=null, rfNumber=[5], rfOrder=8, authorNames=程浩忠, 胡枭, 王莉, journalName=电力系统自动化, refType=null, unstructuredReference=程浩忠, 胡枭, 王莉, 等. 区域综合能源系统规划研究综述[J]. 电力系统自动化, 2019, 43(7):2-13., articleTitle=区域综合能源系统规划研究综述, refAbstract=null), Reference(id=1190338322565267492, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, doi=null, pmid=null, pmcid=null, year=2019, volume=43, issue=7, pageStart=2, pageEnd=13, url=null, language=null, rfNumber=[5], rfOrder=9, authorNames=CHENG Haozhong, HU Xiao, WANG Li, journalName=Automation of Electric Power Systems, refType=null, unstructuredReference=CHENG Haozhong, HU Xiao, WANG Li, et al. Review on research of regional integrated energy system planning[J]. Automation of Electric Power Systems, 2019, 43(7):2-13., articleTitle=Review on research of regional integrated energy system planning, refAbstract=null), Reference(id=1190338322628182053, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, doi=null, pmid=null, pmcid=null, year=2018, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[6], rfOrder=10, authorNames=KHAN M, JAVAID N, IQBAL M N, journalName=Proceedings of the 12th International Conference on Complex,Intelligent,and Software Intensive Systems,Kunibiki Messe,Matsue,Japan, refType=null, unstructuredReference=KHAN M, JAVAID N, IQBAL M N, et al. Load prediction based on multivariate time series forecasting for energy consumption and behavioral analytics[C]// Proceedings of the 12th International Conference on Complex,Intelligent,and Software Intensive Systems,Kunibiki Messe,Matsue,Japan, 2018., articleTitle=Load prediction based on multivariate time series forecasting for energy consumption and behavioral analytics, refAbstract=null), Reference(id=1190338322695290918, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, doi=null, pmid=null, pmcid=null, year=2019, volume=10, issue=1, pageStart=841, pageEnd=851, url=null, language=null, rfNumber=[7], rfOrder=11, authorNames=KONG W, DONG Z Y, JIA Y W, journalName=IEEE Transactions on Smart Grid, refType=null, unstructuredReference=KONG W, DONG Z Y, JIA Y W, et al. Short-term residential load forecasting based on LSTM recurrent neural network[J]. IEEE Transactions on Smart Grid, 2019, 10(1):841-851., articleTitle=Short-term residential load forecasting based on LSTM recurrent neural network, refAbstract=null), Reference(id=1190338322762399783, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, doi=null, pmid=null, pmcid=null, year=2021, volume=9, issue=null, pageStart=32436, pageEnd=32448, url=null, language=null, rfNumber=[8], rfOrder=12, authorNames=RAFI S H, NAHID-AL-MASOOD, DEEBA S R, journalName=IEEE Access, refType=null, unstructuredReference=RAFI S H, NAHID-AL-MASOOD, DEEBA S R, et al. A short-term load forecasting method using integrated CNN and LSTM network[J]. IEEE Access, 2021,9:32436-32448., articleTitle=A short-term load forecasting method using integrated CNN and LSTM network, refAbstract=null), Reference(id=1190338322816925736, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, doi=null, pmid=null, pmcid=null, year=2020, volume=35, issue=null, pageStart=2937, pageEnd=2948, url=null, language=null, rfNumber=[9], rfOrder=13, authorNames=TAN M, YUAN S, LI S, journalName=IEEE Transactions on Power Systems, refType=null, unstructuredReference=TAN M, YUAN S, LI S, et al. Ultra-short-term industrial power demand forecasting using LSTM based hybrid ensemble learning[J]. IEEE Transactions on Power Systems, 2020,35:2937-2948., articleTitle=Ultra-short-term industrial power demand forecasting using LSTM based hybrid ensemble learning, refAbstract=null), Reference(id=1190338322879840297, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, doi=null, pmid=null, pmcid=null, year=2022, volume=52, issue=21, pageStart=74, pageEnd=80, url=null, language=null, rfNumber=[10], rfOrder=14, authorNames=卢武, 郑人杰, 赵文彬, journalName=电气传动, refType=null, unstructuredReference=卢武, 郑人杰, 赵文彬, 等. 基于EMD-GRU的短期电力负荷预测方法[J]. 电气传动, 2022, 52(21):74-80., articleTitle=基于EMD-GRU的短期电力负荷预测方法, refAbstract=null), Reference(id=1190338322951143466, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, doi=null, pmid=null, pmcid=null, year=2022, volume=52, issue=21, pageStart=74, pageEnd=80, url=null, language=null, rfNumber=[10], rfOrder=15, authorNames=LU Wu, ZHENG Renjie, ZHAO Wenbin, journalName=Electric Drive, refType=null, unstructuredReference=LU Wu, ZHENG Renjie, ZHAO Wenbin, et al. Short term power load forecasting method based on EMD-GRU[J]. Electric Drive, 2022, 52(21):74-80., articleTitle=Short term power load forecasting method based on EMD-GRU, refAbstract=null), Reference(id=1190338323014058027, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, doi=null, pmid=null, pmcid=null, year=2022, volume=43, issue=9, pageStart=499, pageEnd=507, url=null, language=null, rfNumber=[11], rfOrder=16, authorNames=欧阳静, 杨吕, 尹康, journalName=太阳能学报, refType=null, unstructuredReference=欧阳静, 杨吕, 尹康, 等. 基于ALIF-LSTM多任务学习的综合能源系统短期负荷预测[J]. 太阳能学报, 2022, 43(9):499-507., articleTitle=基于ALIF-LSTM多任务学习的综合能源系统短期负荷预测, refAbstract=null), Reference(id=1190338323081166892, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, doi=null, pmid=null, pmcid=null, year=2022, volume=43, issue=9, pageStart=499, pageEnd=507, url=null, language=null, rfNumber=[11], rfOrder=17, authorNames=OUYANG Jing, YANG LÜ, YIN Kang, journalName=Acta Energiae Solaris Sinica, refType=null, unstructuredReference=OUYANG Jing, YANG , YIN Kang, et al. Short-term load forecasting of integrated energy systems based on ALIF-LSTM multi task learning[J]. Acta Energiae Solaris Sinica, 2022, 43(9):499-507., articleTitle=Short-term load forecasting of integrated energy systems based on ALIF-LSTM multi task learning, refAbstract=null), Reference(id=1190338323139887149, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, doi=null, pmid=null, pmcid=null, year=2006, volume=null, issue=null, pageStart=3069, pageEnd=3072, url=null, language=null, rfNumber=[12], rfOrder=18, authorNames=SUN W, ZHANG Y, LI F, journalName=International Conference on Machine Learning and Cybernetics, refType=null, unstructuredReference=SUN W, ZHANG Y, LI F. The neural network model based on PSO for short-term load forecasting[C]// International Conference on Machine Learning and Cybernetics, Dalian: IEEE, 2006:3069-3072., articleTitle=The neural network model based on PSO for short-term load forecasting, refAbstract=null), Reference(id=1190338323198607406, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, doi=null, pmid=null, pmcid=null, year=2008, volume=null, issue=null, pageStart=82, pageEnd=85, url=null, language=null, rfNumber=[13], rfOrder=19, authorNames=ZHANG C, MING L, TANG Mingyang, journalName=2008 International Conference on Information Management,Innovation Management and Industrial Engineering, refType=null, unstructuredReference=ZHANG C, MING L, TANG Mingyang. BP neural network optimized with PSO algorithm for daily load forecasting[C]// 2008 International Conference on Information Management,Innovation Management and Industrial Engineering, Taipei: IEEE, 2008:82-85., articleTitle=BP neural network optimized with PSO algorithm for daily load forecasting, refAbstract=null), Reference(id=1190338323274104879, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, doi=null, pmid=null, pmcid=null, year=2019, volume=43, issue=8, pageStart=131, pageEnd=137, url=null, language=null, rfNumber=[14], rfOrder=20, authorNames=陆继翔, 张琪培, 杨志宏, journalName=电力系统自动化, refType=null, unstructuredReference=陆继翔, 张琪培, 杨志宏, 等. 基于CNN-LSTM混合神经网络模型的短期负荷预测方法[J]. 电力系统自动化, 2019, 43(8):131-137., articleTitle=基于CNN-LSTM混合神经网络模型的短期负荷预测方法, refAbstract=null), Reference(id=1190338323337019440, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, doi=null, pmid=null, pmcid=null, year=2019, volume=43, issue=8, pageStart=131, pageEnd=137, url=null, language=null, rfNumber=[14], rfOrder=21, authorNames=LU J X, ZHANG Q P, YANG Z H, journalName=Automation of Electric Power Systems, refType=null, unstructuredReference=LU J X, ZHANG Q P, YANG Z H, et al. Short-term load forecasting method based on CNN-LSTM hybrid neural network model[J]. Automation of Electric Power Systems, 2019, 43(8):131-137., articleTitle=Short-term load forecasting method based on CNN-LSTM hybrid neural network model, refAbstract=null), Reference(id=1190338323391545393, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, doi=null, pmid=null, pmcid=null, year=2023, volume=53, issue=5, pageStart=63, pageEnd=70, url=null, language=null, rfNumber=[15], rfOrder=22, authorNames=顾吉鹏, 邵亮, 陆垂基, journalName=电气传动, refType=null, unstructuredReference=顾吉鹏, 邵亮, 陆垂基, 等. 基于LSTNet模型的配电台区短期电力负荷预测研究[J]. 电气传动, 2023, 53(5):63-70., articleTitle=基于LSTNet模型的配电台区短期电力负荷预测研究, refAbstract=null), Reference(id=1190338323462848562, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, doi=null, pmid=null, pmcid=null, year=2023, volume=53, issue=5, pageStart=63, pageEnd=70, url=null, language=null, rfNumber=[15], rfOrder=23, authorNames=GU Jipeng, SHAO Liang, LU Chuiji, journalName=Electric Drive, refType=null, unstructuredReference=GU Jipeng, SHAO Liang, LU Chuiji, et al. Research on short term power load prediction in distribution stations based on LSTNet model[J]. Electric Drive, 2023, 53(5):63-70., articleTitle=Research on short term power load prediction in distribution stations based on LSTNet model, refAbstract=null), Reference(id=1190338323538346035, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, doi=null, pmid=null, pmcid=null, year=2022, volume=37, issue=5, pageStart=1242, pageEnd=1251, url=null, language=null, rfNumber=[16], rfOrder=24, authorNames=赵洋, 王瀚墨, 康丽, journalName=电工技术学报, refType=null, unstructuredReference=赵洋, 王瀚墨, 康丽, 等. 基于时间卷积网络的短期电力负荷预测[J]. 电工技术学报, 2022, 37(5):1242-1251., articleTitle=基于时间卷积网络的短期电力负荷预测, refAbstract=null), Reference(id=1190338323634815028, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, doi=null, pmid=null, pmcid=null, year=2022, volume=37, issue=5, pageStart=1242, pageEnd=1251, url=null, language=null, rfNumber=[16], rfOrder=25, authorNames=ZHAO Yang, WANG Hanmo, KANG Li, journalName=Transactions of China Electrotechnical Society, refType=null, unstructuredReference=ZHAO Yang, WANG Hanmo, KANG Li, et al. Temporal convolution network-based short-term electrical load forecasting[J]. Transactions of China Electrotechnical Society, 2022, 37(5):1242-1251., articleTitle=Temporal convolution network-based short-term electrical load forecasting, refAbstract=null), Reference(id=1190338323718701109, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, doi=null, pmid=null, pmcid=null, year=2020, volume=7, issue=1, pageStart=1, pageEnd=41, url=null, language=null, rfNumber=[17], rfOrder=26, authorNames=HANCOCK John, KHOSHGOFTAAR Taghi, journalName=Journal of Big Data, refType=null, unstructuredReference=HANCOCK John, KHOSHGOFTAAR Taghi. Survey on categorical data for neural networks[J]. Journal of Big Data, 2020, 7(1):1-41., articleTitle=Survey on categorical data for neural networks, refAbstract=null), Reference(id=1190338323798392886, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, doi=null, pmid=null, pmcid=null, year=2016, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[18], rfOrder=27, authorNames=MIWA M, BANSAL M, journalName=Annual meeting of the Association for Computational Linguistics, refType=null, unstructuredReference=MIWA M, BANSAL M. End-to-end relation extraction using LSTMs on sequences and tree structures[C]// Annual meeting of the Association for Computational Linguistics, 2016., articleTitle=End-to-end relation extraction using LSTMs on sequences and tree structures, refAbstract=null)], funds=[Fund(id=1190338321839652890, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, awardId=2022XAGG0116, language=CN, fundingSource=雄安新区科技创新专项《面向雄安新区现代商业园区碳中和路径规划与智慧降碳关键技术研究与示范》(2022XAGG0116), fundOrder=null, country=null)], companyList=[AuthorCompany(id=1190338318161249238, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, xref=null, ext=[AuthorCompanyExt(id=1190338318169637847, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, companyId=1190338318161249238, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1. State Grid Xiong'an Comprehensive Energy Services Co.,Ltd.,Baoding 071700,Hebei,China), AuthorCompanyExt(id=1190338318178026456, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, companyId=1190338318161249238, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1.国网雄安综合能源服务有限公司,河北 保定 071700)]), AuthorCompany(id=1190338318245135321, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, xref=null, ext=[AuthorCompanyExt(id=1190338318253523930, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, companyId=1190338318245135321, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2. Beijing NARI Digital Technology Co.,Ltd.,Beijing 100193,China), AuthorCompanyExt(id=1190338318261912539, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, companyId=1190338318245135321, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2.北京南瑞数字技术有限公司,北京 100193)])], figs=[ArticleFig(id=1190338320275177476, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, language=EN, label=Fig.1, caption=The figure of fluctuation trend and cyclical trend of electrical load, figureFileSmall=OdhMqpOUwsXRDS1ZiIQAQQ==, figureFileBig=RHODV+djTKm9M6qPH4cotA==, tableContent=null), ArticleFig(id=1190338320342286341, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, language=CN, label=图1, caption=电负荷的波动趋势和周期走势分析图, figureFileSmall=OdhMqpOUwsXRDS1ZiIQAQQ==, figureFileBig=RHODV+djTKm9M6qPH4cotA==, tableContent=null), ArticleFig(id=1190338320510058502, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, language=EN, label=Fig.2, caption=Comparison of the fluctuation trend of electrical load before and after using the minimum and maximum value method, figureFileSmall=y1TpGD0Ai8l+r4IVKu5k1Q==, figureFileBig=hNn2r401A9mcfmWKctcjnQ==, tableContent=null), ArticleFig(id=1190338320568778759, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, language=CN, label=图2, caption=最小最大值法处理前后的电负荷波动趋势对比分析图, figureFileSmall=y1TpGD0Ai8l+r4IVKu5k1Q==, figureFileBig=hNn2r401A9mcfmWKctcjnQ==, tableContent=null), ArticleFig(id=1190338320631693320, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, language=EN, label=Fig.3, caption=The histogram of 24 h of electricity load within one day, figureFileSmall=TjjGfSPw2ZBeBoXRkDI5nA==, figureFileBig=Ho1Rcg7wlk7t6PreoGSbaw==, tableContent=null), ArticleFig(id=1190338320694607881, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, language=CN, label=图3, caption=一日内24 h电负荷的数据分布图, figureFileSmall=TjjGfSPw2ZBeBoXRkDI5nA==, figureFileBig=Ho1Rcg7wlk7t6PreoGSbaw==, tableContent=null), ArticleFig(id=1190338320753328138, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, language=EN, label=Fig.4, caption=The structural diagram of box-plot, figureFileSmall=bHw8yqWPbJFJ+ul8VDXaoQ==, figureFileBig=yPzv47jfbR1lhcUSFBWJ1A==, tableContent=null), ArticleFig(id=1190338320824631307, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, language=CN, label=图4, caption=箱型图的结构图, figureFileSmall=bHw8yqWPbJFJ+ul8VDXaoQ==, figureFileBig=yPzv47jfbR1lhcUSFBWJ1A==, tableContent=null), ArticleFig(id=1190338320887545868, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, language=EN, label=Fig.5, caption=The box-plot of electricity load, figureFileSmall=jeNV2F/2aUHGWu0IpzYpYw==, figureFileBig=TOFZLKq+FRU7o5YnGvd6BA==, tableContent=null), ArticleFig(id=1190338320958849037, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, language=CN, label=图5, caption=电负荷数据箱型图, figureFileSmall=jeNV2F/2aUHGWu0IpzYpYw==, figureFileBig=TOFZLKq+FRU7o5YnGvd6BA==, tableContent=null), ArticleFig(id=1190338321021763598, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, language=EN, label=Fig.6, caption=The heat map of input characteristic for electrical load, figureFileSmall=l6E4EOmcTHO49XKlbp68NA==, figureFileBig=7AuhrGcmsUK+61mreWEsIA==, tableContent=null), ArticleFig(id=1190338321088872463, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, language=CN, label=图6, caption=电负荷的输入特征热度图, figureFileSmall=l6E4EOmcTHO49XKlbp68NA==, figureFileBig=7AuhrGcmsUK+61mreWEsIA==, tableContent=null), ArticleFig(id=1190338321160175632, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, language=EN, label=Fig.7, caption=The network structure diagram of AT-LSTM model, figureFileSmall=rjB2qcC9wPNh4qr/tZiojA==, figureFileBig=dOyWMyD03ArM8ApYQJMPJw==, tableContent=null), ArticleFig(id=1190338321239867409, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, language=CN, label=图7, caption=AT-LSTM模型的网络结构图, figureFileSmall=rjB2qcC9wPNh4qr/tZiojA==, figureFileBig=dOyWMyD03ArM8ApYQJMPJw==, tableContent=null), ArticleFig(id=1190338321294393362, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, language=EN, label=Fig.8, caption=The structure diagram of the short-term electricity load prediction model proposed in this paper, figureFileSmall=C1IZtcr+mNGUuu2t0o8ksg==, figureFileBig=CPLKk5CdvQqOg3rol+7G2g==, tableContent=null), ArticleFig(id=1190338321365696531, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, language=CN, label=图8, caption=本文提出的短期电负荷预测模型结构图, figureFileSmall=C1IZtcr+mNGUuu2t0o8ksg==, figureFileBig=CPLKk5CdvQqOg3rol+7G2g==, tableContent=null), ArticleFig(id=1190338321424416788, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, language=EN, label=Fig.9, caption=The comparison of the prediction results of 5 models in August 2023, figureFileSmall=o53qLjJ0fMmA94uc7/EIAA==, figureFileBig=mn82rrjOK2/PfNkd7bqKng==, tableContent=null), ArticleFig(id=1190338321483137045, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, language=CN, label=图9, caption=5种模型在2023年8月的预测结果对比分析图, figureFileSmall=o53qLjJ0fMmA94uc7/EIAA==, figureFileBig=mn82rrjOK2/PfNkd7bqKng==, tableContent=null), ArticleFig(id=1190338321546051606, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, language=EN, label=Fig.10, caption=The comparison of the prediction results of five models in December 2023, figureFileSmall=fGwzGxbeulqOg/bgoTCdqg==, figureFileBig=ycWi3zNU0lH7451gzPCrgQ==, tableContent=null), ArticleFig(id=1190338321604771863, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, language=CN, label=图10, caption=5种模型在2023年12月的预测结果对比分析图, figureFileSmall=fGwzGxbeulqOg/bgoTCdqg==, figureFileBig=ycWi3zNU0lH7451gzPCrgQ==, tableContent=null), ArticleFig(id=1190338321667686424, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, language=EN, label=Tab.1, caption=

The predictive performance analysis results of this article's model and other algorithms

, figureFileSmall=null, figureFileBig=null, tableContent=
模型 RMSE MAE CC
LSTM 45.096 35.323 0.857
AT-LSTM 39.537 29.503 0.799
决策树 47.232 35.568 0.813
随机森林 38.808 29.062 0.866
本文模型 35.586 26.735 0.901
), ArticleFig(id=1190338321726406681, tenantId=1146029695717560320, journalId=1189987059142926344, articleId=1190337959950909999, language=CN, label=表1, caption=

本文模型与其他算法的预测性能分析结果

, figureFileSmall=null, figureFileBig=null, tableContent=
模型 RMSE MAE CC
LSTM 45.096 35.323 0.857
AT-LSTM 39.537 29.503 0.799
决策树 47.232 35.568 0.813
随机森林 38.808 29.062 0.866
本文模型 35.586 26.735 0.901
)], attaches=null, journal=Journal(id=1149695973913305096, delFlag=0, nameCn=电气传动, nameEn=Electric Drive, nameHistory1=null, nameHistory2=null, issn=1001-2095, eissn=null, cn=12-1067/TP, coden=null, periodic=0, language=CN, oaType=null, ccby=null, superviseOffice=null, ownerOffice=null, pubOffice=null, editorOffice=null, officeType=null, aims=null, clcCode=null, officeProv=null, officeCity=null, officeAddr=null, officeZip=null, officeEmail=null, officePhone=null, editDirector=null, officeDirector=null, officeDirectorPhone=null, officeStaffNum=null, officeEmpNum=null, coverPicUrl=MX9kQyxauhRIcWSHz1lsfg==, journalPrice=null, startedYear=null, abbrevIsoEn=null, journalRemark=null, publicationField=null, createdTime=1752038480501, updatedTime=1761735785885, createdBy=18614031015, updatedBy=13701087609, firstLetterCn=E, firstLetterEn=E, subjectCode=Engineering, subjectName=Engineering, subjectCodeEn=Engineering, subjectNameEn=null, picCn=MX9kQyxauhRIcWSHz1lsfg==, picEn=R8RQJTk995s1dk9HSLANAQ==, jcr=null, cjcr=null, exts=[JournalExt(id=1190369420775096487, language=CN, name=电气传动, nameHistory1=null, nameHistory2=null, managedBy=, sponsoredBy=, publishedBy=, editorOffice=, officeProv=null, officeCity=null, officeAddr=, officeZip=, editDirector=, officeDirector=null, officePhone=null, coverPicUrl=null, journalRemark=, submitArticleUrl=null, websiteUrl=, createdTime=1761735785907, updatedTime=1761735785907, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=, submissionAuthorUrl=https://tg.au365.cn/Login/index.html, submissionEditorUrl=https://tg.au365.cn/Login/index/cate/3.html, submissionReviewUrl=https://tg.au365.cn/Login/index/cate/2.html, submissionCeEditorUrl=, submissionAeEditorUrl=, option={"copyright":""}), JournalExt(id=1190369420829622440, language=EN, name=Electric Drive, nameHistory1=null, nameHistory2=null, managedBy=, sponsoredBy=, publishedBy=, editorOffice=, officeProv=null, officeCity=null, officeAddr=, officeZip=, editDirector=, officeDirector=null, officePhone=null, coverPicUrl=null, journalRemark=, submitArticleUrl=null, websiteUrl=, createdTime=1761735785920, updatedTime=1761735785920, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=, submissionAuthorUrl=https://tg.au365.cn/Login/index.html, submissionEditorUrl=https://tg.au365.cn/Login/index/cate/3.html, submissionReviewUrl=https://tg.au365.cn/Login/index/cate/2.html, submissionCeEditorUrl=, submissionAeEditorUrl=, option={"copyright":""})], databaseList=null, tenantJournalId=1189987059142926344, websiteList=[Website(id=1189988338468713189, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1189987059142926344, journalNameCn=null, journalNameEn=null, grayFlag=null, tenantId=1146029695717560320, platformId=null, journalGroupId=null, journalGroupNameCn=null, journalGroupNameEn=null, type=1, domain=https://castjournals.cast.org.cn/joweb/dqcd/CN, language=CN, createTime=1761644928805, createBy=18614031015, updateTime=1761644944247, updateBy=18614031015, name=电气传动-中文, tplId=1146099689490845704, title=电气传动, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1190236467705254391, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189988338468713189, code=articleTextType, value=kx, createTime=1761704087425, updateTime=1761704087425, creator=18614031015, updator=18614031015), WebsiteProps(id=1190236467671699956, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189988338468713189, code=banner, value=null, createTime=1761704087417, updateTime=1761704087417, creator=18614031015, updator=18614031015), WebsiteProps(id=1190236467763974650, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189988338468713189, code=grayFlag, value=0, createTime=1761704087439, updateTime=1761704087439, creator=18614031015, updator=18614031015), WebsiteProps(id=1190236467663311347, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189988338468713189, code=logo, value=https://castjournals.cast.org.cn/joweb/dqcd/CN/file/pic?fileId=C+51VGe7It8gHMdS0MAlXg==, createTime=1761704087415, updateTime=1761704087415, creator=18614031015, updator=18614031015), WebsiteProps(id=1190236467784946172, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189988338468713189, code=minRunFlag, value=0, createTime=1761704087444, updateTime=1761704087444, creator=18614031015, updator=18614031015), WebsiteProps(id=1190236467692671478, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189988338468713189, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/dqcd/CN/file/pic, createTime=1761704087422, updateTime=1761704087422, creator=18614031015, updator=18614031015), WebsiteProps(id=1190236467776557563, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189988338468713189, code=silenceFlag, value=0, createTime=1761704087442, updateTime=1761704087442, creator=18614031015, updator=18614031015), WebsiteProps(id=1190236467680088565, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189988338468713189, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_cn_619/, createTime=1761704087419, updateTime=1761704087419, creator=18614031015, updator=18614031015), WebsiteProps(id=1190236467713643000, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189988338468713189, code=themeColor, value=null, createTime=1761704087427, updateTime=1761704087427, creator=18614031015, updator=18614031015), WebsiteProps(id=1190236467726225913, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189988338468713189, code=themeStyle, value=null, createTime=1761704087430, updateTime=1761704087430, creator=18614031015, updator=18614031015)]), Website(id=1189988338527433447, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1189987059142926344, journalNameCn=null, journalNameEn=null, grayFlag=null, tenantId=1146029695717560320, platformId=null, journalGroupId=null, journalGroupNameCn=null, journalGroupNameEn=null, type=1, domain=https://castjournals.cast.org.cn/joweb/dqcd/EN, language=EN, createTime=1761644928819, createBy=18614031015, updateTime=1761644963810, updateBy=18614031015, name=电气传动-英文, tplId=1146101810881728533, title=Electric Drive, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1190236496637563393, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189988338527433447, code=articleTextType, value=kx, createTime=1761704094323, updateTime=1761704094323, creator=18614031015, updator=18614031015), WebsiteProps(id=1190236496612397566, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189988338527433447, code=banner, value=null, createTime=1761704094317, updateTime=1761704094317, creator=18614031015, updator=18614031015), WebsiteProps(id=1190236496666923524, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189988338527433447, code=grayFlag, value=0, createTime=1761704094330, updateTime=1761704094330, creator=18614031015, updator=18614031015), WebsiteProps(id=1190236496604008957, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189988338527433447, code=logo, value=https://castjournals.cast.org.cn/joweb/dqcd/EN/file/pic?fileId=C+51VGe7It8gHMdS0MAlXg==, createTime=1761704094315, updateTime=1761704094315, creator=18614031015, updator=18614031015), WebsiteProps(id=1190236496683700742, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189988338527433447, code=minRunFlag, value=0, createTime=1761704094334, updateTime=1761704094334, creator=18614031015, updator=18614031015), WebsiteProps(id=1190236496629174784, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189988338527433447, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/dqcd/EN/file/pic, createTime=1761704094321, updateTime=1761704094321, creator=18614031015, updator=18614031015), WebsiteProps(id=1190236496675312133, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189988338527433447, code=silenceFlag, value=0, createTime=1761704094332, updateTime=1761704094332, creator=18614031015, updator=18614031015), WebsiteProps(id=1190236496620786175, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189988338527433447, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_en_623/, createTime=1761704094319, updateTime=1761704094319, creator=18614031015, updator=18614031015), WebsiteProps(id=1190236496650146306, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189988338527433447, code=themeColor, value=null, createTime=1761704094326, updateTime=1761704094326, creator=18614031015, updator=18614031015), WebsiteProps(id=1190236496658534915, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1189988338527433447, code=themeStyle, value=null, createTime=1761704094328, updateTime=1761704094328, creator=18614031015, updator=18614031015)])], journalTitle=电气传动, weixinUrl=null, journalUrl=https://www.au365.cn/, iacademicId=null, status=1, seqNo=null, journalTitleEn=Electric Drive, journalPhotoCn=MX9kQyxauhRIcWSHz1lsfg==, journalPhotoEn=R8RQJTk995s1dk9HSLANAQ==, journalFirstLetter=E, journalRecommend=null, journalNew=null, journalCollection=null, jcrJf=null, cjcrJf=null, jcrJfStr=null, cjcrJfStr=null, submissionFirstDecision=null, sciSubjectClassification=null, casSubjectClassification=null, citeScore=null, totalCitationFrequency=null, icpCode=null, psCode=null, advertisingLicenseCode=null, copyrightInformation=null, country=null, option=, provinceCode=null, provinceName=null, collectFlag=false), detailUrlCn=https://castjournals.cast.org.cn/joweb/dqcd/CN/10.19457/j.1001-2095.dqcd25612, detailUrlEn=https://castjournals.cast.org.cn/joweb/dqcd/EN/10.19457/j.1001-2095.dqcd25612, pdfUrlCn=https://castjournals.cast.org.cn/joweb/dqcd/CN/PDF/10.19457/j.1001-2095.dqcd25612, pdfUrlEn=https://castjournals.cast.org.cn/joweb/dqcd/EN/PDF/10.19457/j.1001-2095.dqcd25612, aliStartDate=null, aliEndDate=null, collectionFlag=false, citedCount=null, citedUrl=null, reference=null)
收藏切换
基于AT-LSTM及集成学习的短期电负荷预测方法
收藏切换
PDF下载
张保瑞 1 , 王靖韬 2 , 谭瑶 2 , 王志光 2 , 李唐 1 , 史洋 1
电气传动 | 综合能源与现代电网 2025,55(8): 89-96
收起
收藏切换
电气传动 | 综合能源与现代电网 2025, 55(8): 89-96
基于AT-LSTM及集成学习的短期电负荷预测方法
全屏
张保瑞1 , 王靖韬2, 谭瑶2, 王志光2, 李唐1, 史洋1
作者信息
  • 1.国网雄安综合能源服务有限公司,河北 保定 071700
  • 2.北京南瑞数字技术有限公司,北京 100193
  • 张保瑞(1976—),男,硕士,高级工程师,主要研究方向为低碳能源、建筑节能、供冷供热,Email:

Short-term Electrical Load Forecasting Method Based on AT-LSTM and Stacking Ensemble Learning
Baorui ZHANG1 , Jingtao WANG2, Yao TAN2, Zhiguang WANG2, Tang LI1, Yang SHI1
Affiliations
  • 1. State Grid Xiong'an Comprehensive Energy Services Co.,Ltd.,Baoding 071700,Hebei,China
  • 2. Beijing NARI Digital Technology Co.,Ltd.,Beijing 100193,China
出版时间: 2025-08-20 doi: 10.19457/j.1001-2095.dqcd25612
文章导航
收藏切换

精确的短期电负荷预测对综合能源系统的规划和优化调度具有重要意义,然而真实综合能源系统中的负荷数据由于数据质量低、波动趋势多变等情况,预测精度较低。为此,提出一种带注意力机制的长短时记忆(AT-LSTM)及集成学习的短期电负荷预测方法,在Stacking集成学习的框架下,对AT-LSTM、随机森林、决策树进行集成并预测短期电负荷,弥补单一模型预测精度较低的不足。具体地,基于对真实综合能源系统中的数据探索性分析结果,通过数据综合特征工程构造模型输入特征后,采用该预测方法进行短期电负荷预测。以北京某综合能源系统为例的实验结果表明,相比长短期记忆网络、决策树及随机森林等其他算法,所提方法预测误差最高降低了24.8%。

短期电负荷预测  /  基于注意力机制的长短期记忆  /  集成学习  /  综合能源系统  /  特征工程

Accurate short-term electrical load forecasting is of great significance for the design and optimization scheduling of integrated energy systems(IES). However,the load data in real integrated energy systems are low-quality and fluctuating,so the forecasting accuracy of existing prediction models is low. A short-term electrical load forecasting method based on attention-based long short-term memory(AT-LSTM)and Stacking learning was proposed. Under the framework of Stacking ensemble learning,AT-LSTM,random forest and decision tree were ensembled to forecast short-term electrical load which can make up for the low prediction accuracy of a single model. Based on the exploratory analysis results of data,the data feature engineering model was constructed to input features,and this prediction method was used for short-term electricity load prediction. The experimental results of the integrated energy system in Beijing show that compared to other algorithms,the proposed method has a maximum prediction error reduction of 24.8%.

short-term electrical load forecasting  /  attention-based long short-term memory(AT-LSTM)  /  Stacking ensemble learning  /  integrated energy system(IES)  /  feature engineering
张保瑞, 王靖韬, 谭瑶, 王志光, 李唐, 史洋. 基于AT-LSTM及集成学习的短期电负荷预测方法. 电气传动, 2025 , 55 (8) : 89 -96 . DOI: 10.19457/j.1001-2095.dqcd25612
Baorui ZHANG, Jingtao WANG, Yao TAN, Zhiguang WANG, Tang LI, Yang SHI. Short-term Electrical Load Forecasting Method Based on AT-LSTM and Stacking Ensemble Learning[J]. Electric Drive, 2025 , 55 (8) : 89 -96 . DOI: 10.19457/j.1001-2095.dqcd25612
随着社会的高速发展,能源可持续供应及环境污染问题日益严重,如何促进新能源消纳,推动环保进程成为世界各国关注的重点[1-2]。综合能源系统(IES)集成多种形式的能量供应、转换和储存设备,实现不同类型能源在源、网、荷、储等环节的耦合,是新一代能源系统的重要体现形式[3]。相比于单一能源系统,综合能源系统由电、冷、热、气等多种能源网络耦合形成,能够实现多种异质能源子系统之间的协调规划、优化运行、协同管理、交互响应和互补互济。在满足系统内多元化用能需求的同时,要有效地提升能源利用效率,促进能源可持续发展。
短期负荷预测是综合能源系统设计和运行的重要环节,为综合能源系统运行优化提供数据基础,因此如何提高短期负荷预测的精确度是国内外学者的研究热点[4-5]
国内外利用人工智能方法在电负荷预测方面展开了大量研究,文献[6]将电负荷预测建模为非线性模型,利用支持向量回归(support vector regression,SVR)算法进行建筑居民用电预测;文献[7-10]采用长短期记忆(LSTM)神经网络或门控循环单元进行负荷预测,可以对时间序列间的连续性进行建模,通过端到端的方式在气象、历史数据等相关因素中挖掘数据规律,实现负荷预测;文献[11]提出基于自适应局部迭代滤波的历史负荷数据分解法,并结合长短期记忆网络实现多任务学习的综合能源系统短期负荷预测模型;文献[12-13]提出一种粒子群算法用于优化神经网络的超参数,以提高神经网络方法在负荷预测模型的精度。在深度学习模型快速发展的同时,其在电力负荷预测领域也得到了较好的应用,文献[14-15]结合卷积神经网络(convolutional neural networks,CNN)和长短期记忆网络共同用于电力负荷预测,文献[14]通过构建CNN-LSTM的组合模型进行负荷预测,文献[15]提出基于LSTNet模型的配电台区短期电力负荷预测研究,用卷积神经网络提取负荷数据间的局部依赖关系,用LSTM神经网络提取负荷数据长期变化趋势;文献[16]提出一种基于时间卷积网络的短期电力负荷预测方法,对具有强非线性特征的短期电力负荷预测提升效果明显。
然而,将现有的短期负荷预测方法在应用到真实综合能源系统时,存在精度不高、不适配等问题,无法适应真实综合能源系统中面临的复杂多变的负荷波动情况。另一方面,真实综合能源系统产生的负荷数据存在较多的噪声、缺失、异常点等问题,数据的低质量会对负荷预测的效果产生较大的影响。因此本文提出了一种基于带注意力机制的长短期记忆(AT-LSTM)及集成学习的综合能源系统短期电负荷预测方法,并在构建预测模型前进行了充分的负荷特性分析,为实现负荷预测模型提供可靠的特征数据。最后,本文在真实综合能源系统的数据集上进行了实验分析,实验表明本文提出的方法提升了预测精度,降低了预测误差。
本文以北京某综合能源系统2021年11月至2023年5月的数据为基础展开探索性分析,总结其复杂多变的数据特点,为后续预测模型的训练提供数据参考。该综合能源系统涉及的多能负荷包括电负荷和空调负荷,相关供能设备包括光伏、充电桩、直燃机、冷水机组等设备。通过对园区内能源设备供能特点、用能特点、空间位置等能源特点进行聚合归类,形成的能源子系统包括光伏系统、电动汽车可控负荷系统、直燃机系统、冷水机组系统和柔性空调末端系统。
结合本综合能源系统的实际运行情况及电负荷的关联因素分析,该系统电负荷的影响因素可总结为4类:历史负荷波动规律、气象变化、用电习惯以及供能设备用电功率。不同类型的负荷影响因素时间尺度不同,其中综合能源系统中历史负荷及设备用电功率数据的采集时间尺度为15 min/次,气象数据的时间尺度为1 h/次。
本文首先基于折线图对负荷数据的波动趋势和周期走势进行分析,图1分别展示了电负荷的波动趋势和一日内的周期走势。由图1可知:1)电负荷历史数据中存在较大或较小的异常点,无法看到真实波动趋势;2)电负荷历史数据中存在一些空值的情况。
为了明确负荷的真实波动及周期趋势,本文首先通过最小最大值法去除负荷的异常点,极大异常值由规定的最大值替代,极小异常值由规定的最小值替代。图2展示了最小最大值法处理前后的电负荷波动趋势对比结果,应用最小最大值法后能够还原电负荷原本的波动趋势。如图2所示,电负荷存在日均负荷较小的两段时期,结合疫情等现实因素,推测该段时期的数据不具备参考性,应删除日均负荷较小的两段数据。另外,图2中电负荷仍然存在部分噪声点,后续考虑在特征工程的数据预处理环节中处理。
其次,本文还通过电负荷的数据分布判断数据的异常情况,结合直方图分析了一日内24 h电负荷的数据分布情况,由图3可知,历史电负荷在每个小时的数据分布呈现中间密集,两边稀疏的特征,服从正态分布。因此处理后的电负荷数据符合正常数据规律,可用于短期电负荷预测。
结合上述负荷数据探索性分析后,为适用真实综合能源系统的复杂数据场景,本文在负荷数据、设备出力数据、天气气象以及用电习惯等多类影响因素组成的输入数据上应用特征工程,实现数据预处理、特征设计、特征选择等过程,构造最终的数据集。
本文的数据预处理步骤中将实现数据清洗、异常检测等过程,基于上述数据探索性分析结果,本文将先后采用最小最大值法及箱型图法,实现对输入数据的数据预处理。其中,本文在利用箱型图方法检测异常点时,采用周期中位数替换异常点。箱型图是1977年由美国统计学家John Tukey发明的,它由5个数值点组成:最小值(Min)、下四分位数(Q1)、中位数(median)、上四分位数(Q3)、最大值(Max)。其中下四分位数、上四分位数组成一个盒子,如图4所示。
箱型图中不在[MinMax]范围内的数据点为异常点,本文中采用的MinMax取值分别为
M i n = Q 1 - 1.5 I Q R
M a x = Q 3 + 1.5 I Q R
其中
I Q R = Q 3 - Q 1
式中:IQR为四分位距。
一日内01:00—24:00电负荷数据的箱型图如图5所示,本文综合分析一日内24 h电负荷数据的箱型图,分别对各个小时去除不在[MinMax]范围内的异常点,而非整体统一去除异常点,尽可能保留电负荷的日内波动趋势,避免对预测结果的准确性造成影响。
由于电负荷数据本身具有周期性、季节性等特点,因此在原本的4类负荷影响因素之外,本文将设计季节特征和月、日、时刻特征作为时间特征。其中季节特征包括夏季、冬季、过渡季3种类型,月特征的范围为1月—12月,日特征的范围为1日—31日,时刻特征范围为01:00—24:00。
由于气象特征中的气象摘要、风向,时间特征中的季节类型等类别特征无法直接用于预测模型的输入,本文利用独热编码(one-hot enco-ding)方法[17]实现对离散的类别特征连续。独热编码是采用位状态寄存器对变量的每个状态进行编码,每个状态均有独立的寄存器位,并且在任意时候只有一位有效。例如:季节类型中夏季、冬季、过渡季由3维向量表征,夏季为[1,0,0],冬季为[0,1,0],过渡季为[0,0,1]。
特征选择是去除冗余或不相关特征,筛选重要特征的过程,其目的是降低特征维度,降低模型复杂度。本文基于Pearson相关系数法计算输入特征与电负荷之间的相关系数,衡量输入变量与电负荷的相关密切程度,相关系数的绝对值越大,相关性越强,相关系数越接近于1或-1,相关度越强,相关系数越接近于0,相关度越弱。
Pearson相关系数是一种线性相关系数,它度量的是两个变量之间的线性关系,该相关系数的取值范围为[-1,1],负数表示负相关,正数表示正相关,绝对值越大,相关性越强。
相关系数定义如下:
ρ X , Y = ( X - X - ) ( Y - Y - ) ( X - X - ) 2 ( Y - Y - ) 2
式中: X -为变量X的平均值; Y -为变量Y的平均值。
通过对电负荷与相关的特征计算相关系数后,对应的热度图如图6所示,因此为保留重要特征,本文选择相关系数绝对值大于阈值0.3的特征作为最终的特征。因此,通过数据预处理、特征设计、特征选择等环节后,确定最终的预测模型输入包括历史负荷、设备出力、气象天气等16维特征。
LSTM网络是具有长期记忆能力的一种时间递归神经网络(RNN),它在1997年被提出用于解决传统RNN随时间反向传播中权重消失的问题[18]。LSTM网络由相同结构的LSTM单元组成,主要包括Input Gate,Output Gate和Forget Gate。Input Gate决定了当前时刻的输入保存在单元状态的内容,Forget Gate决定了上一个时刻的单元输出保留在当前时刻的内容,Output Gate决定了单元状态输出的内容。LSTM网络能够对时间进行建模,对于时间序列中数据规律的挖掘有天然的优势,能够用于时间序列预测中。
另外,人类的视觉处理系统往往会选择性地聚焦在图像的某些部分上,而忽略其他不相关的信息,从而有助于我们感知,这便是注意力在人类的生物系统上的体现。注意力机制原理的本质是从关注全部内容到关注重点内容,通过让模型仅动态地关注有助于有效执行手头任务的部分输入,注意力模型引入了这种概念。注意力模型可以被分解为3个步骤:1)query和key进行相似度计算,得到权值;2)将权值进行归一化,得到直接可用的权重;3)将权重和value进行加权求和。因此,注意力模型可以应用在复杂网络机构中,以改善原有网络模型的效果。
本文将采用AT-LSTM,其在原有的LSTM结构中增加注意力层,AT-LSTM模型的网络结构如图7所示。AT-LSTM模型中包括输入层、隐层、注意力层、全连接层、输出层。假设输入层的输入为 X = [ x 1 , x 2 , x 3 , , x T ],经过隐藏层LSTM单元的处理后,得到隐层的输出 H = [ h 1 , h 2 , h 3 , , h T ],在注意力机制后,首先根据打分函数Score计算查询向量q与每个输入 h t之间的相关性,得到一个分数,利用softmax函数对打分结果进行归一化后得到注意力分布 α = [ α 1 , α 2 , α 3 , , α T ]。以 α t为例,其计算公式如下:
                    α t = s o f t m a x [ S c o r e ( h t , q ) ] = e x p [ S c o r e ( h t , q ) ] t T { [ S c o r e ( h t , q ) ] }
最终的注意力层输出结果为 C = t = 1 T α t h t,与隐层的输出结果 h T进行特征融合后,送入全连接层后,计算得到输出层的结果y。其中,本文采用的打分函数为
S c o r e ( h t , q ) = h t · q T
其次,为了弥补单一方法对短期负荷预测精度较低的不足,本文在Stacking集成学习框架下构建集成的综合能源系统短期负荷预测模型。Stacking集成学习是指基于基模型进行集成训练,第j个基模型对第i个训练样本的预测值将作为新的训练集中第i个样本的第j个特征值,最后基于新的训练集进行元模型(meta-model)训练。本文将采用随机森林(random forest)、决策树(decision tree)以及AT-LSTM 3种模型作为集成学习的基模型,线性回归模型作为元模型,具体结构图如图8所示。其次,为降低Stacking集成学习的过拟合风险,针对线性回归的元模型,本文采用带L2正则项的线性回归模型,即岭回归模型。
本文在训练阶段,首先利用训练集分别训练随机森林、决策树以及AT-LSTM 3种模型后,将3种模型对训练集的预测值进行特征融合后与原始label共同构成新的训练集,基于新的训练集进行线性回归模型的训练,输入为训练集 x i , y i i = 1 n,输出为最终模型 h l r x,具体算法如下所示:
1)用训练集训练AT-LSTM基模型 F A T - L S T M x
2)计算预测结果 T r a _ P r e A T - L S T M
3)用训练集训练随机森林基模型 F R F x
4)计算预测结果 T r a _ P r e R F
5)用训练集训练决策树基模型 F D T x
6)计算预测结果 T r a _ P r e D T
7)拼接基模型的预测结果,构造新训练集N-data。 { ( [ T r a _ P r e A T - L S T M , T r a _ P r e R F , T r a _ P r e D T ] , y i ) } i = 1 n用新训练集N-data训练元学习器 h l r x
在测试阶段,测试数据通过3种基模型的预测后输出预测值,对预测值进行特征融合后利用线性回归模型得到最终的预测值。
结合上述数据分析及本文提出的综合能源系统长短期电负荷预测方法,本文在北京某综合能源系统上进行了模型验证。由于该综合能源系统需要提前2日预测全天01:00—24:00的电负荷值,以便提前计算并调整日前调度优化计划。因此,本文将数据集按照时间顺序划分,以2021年10月至2023年9月的数据为训练集进行了模型训练,以2023年8月及12月的数据为测试集进行了验证,训练集包括14 592个样本,测试集包括1 246个样本,以第t日的数据预测第t+2日的电负荷。本文首先通过5折交叉验证以及网格搜索法,确定模型最优的参数,具体地,设置AT-LSTM的学习率为0.005,并采用Adam优化器进行优化训练;随机森林及决策树模型中的参数树最大深度max_depth设置为10。在确定最优超参数后,本文模型在训练过程中随着迭代次数的增加,训练集的损失逐渐减小。
本文在实验阶段采用了3种评价指标,包括平均绝对误差(mean absolute error,MAE)、均方根误差(root mean squared error,RMSE)以及相关系数(correlation coefficient,CC)。RMSEMAE测量预测与真实值之间的误差,相关系数可以测量预测与真实值之间的相关性,具体定义如下:
R M S E = i = 1 i = N ( y i - y i ̑ ) 2 N
M A E = i = 1 i = N | y i - y i ̑ | N
C C = i = 1 i = N ( y i - y - ) ( y i ̑ - y ^ ¯ ) i = 1 i = N ( y i - y - ) 2 i = 1 i = N ( y i - y ^ ¯ ) 2
为了客观地体现本文方法的有效性,本文与LSTM,AT-LSTM、随机森林、决策树等常用算法进行预测性能比较,图9展示了5种模型在测试集上的预测值与真实电负荷值的对比结果,表1展示了本文模型与其他算法的预测性能分析结果。其中,本文模型在测试集上预测的均方根误差为35.586,平均绝对误差为26.735,相关系数为0.901,预测性能均优于单独的AT-LSTM、随机森林、决策树模型等方法。具体地,相比于仅采用AT-LSTM模型的预测结果,RMSEMAE降低了9.9%,相关系数CC提升了12%;相比于仅采用决策树模型的预测结果,RMSEMAE降低了24.8%,相关系数CC提升了10.8%;相比于仅采用随机森林模型的预测结果,RMSEMAE降低了8.3%,相关系数CC提升了4%。实验结果验证了本文所提出方法中Stacking集成学习框架的有效性,能够弥补单一模型预测精度较低的不足。
另外,AT-LSTM模型相比于普通的LSTM模型的预测结果,其RMSEMAE降低了16.5%,实验结果表明相比于普通的LSTM模型,AT-LSTM模型的预测性能更优、收敛速度更快。引入了注意力机制的AT-LSTM模型,更能够挖掘数据间的知识规律,提高短期电负荷的预测精度。
针对目前短期电负荷预测模型对真实综合能源系统预测不够精确的现状,本文提出了一种新的基于AT-LSTM及集成学习的综合能源系统短期电负荷预测方法,在Stacking集成学习的框架下,集成了AT-LSTM、决策树、随机森林等模型。本文实现了对真实综合能源系统中的数据探索性分析,并结合数据分析结果,通过切实有用的特征工程构造模型输入特征,实现短期电负荷预测,提升了预测精度。为综合能源日前优化计划的计算提供可靠的数据基础。
本文的贡献及创新点在于首次将基于AT-LSTM及集成学习的方法应用到短期电负荷预测中,提高了预测精度,能够应对真实的综合能源系统短期电负荷预测难度较大的情况,实现高精度的短期电负荷预测。本文下一步将从短期电负荷波动和平稳特性角度对短期电负荷预测进行深入研究,针对不同波动特点的电负荷分别进行预测,以应对综合能源系统中复杂多变的真实负荷。
  • 雄安新区科技创新专项《面向雄安新区现代商业园区碳中和路径规划与智慧降碳关键技术研究与示范》(2022XAGG0116)
参考文献 引证文献
排序方式:
[1]
程耀华, 张宁, 康重庆, 等. 低碳多能源系统的研究框架及展望[J]. 中国电机工程学报, 2017, 37(14):4060-4069.
CHENG Yaohua, ZHANG Ning, KANG Chongqing, et al. Research framework and prospects of low-carbon multiple energy systems[J]. Proceedings of the CSEE, 2017, 37(14):4060-4069.
[2]
康重庆, 陈启鑫, 夏清. 低碳电力技术的研究展望[J]. 电网技术, 2009, 33(2):1-7.
KANG Chongqing, CHEN Qixin, XIA Qing. Prospects of low-carbon electricity[J]. Power System Technology, 2009, 33(2):1-7.
[3]
罗凤章, 张旭, 杨欣, 等. 基于深度学习的综合能源配电系统负荷分析预测[J]. 高电压技术, 2021, 47(1):23-32.
LUO Fengzhang, ZHANG Xu, YANG Xin, et al. Load analysis and prediction of integrated energy distribution system based on deep learning[J]. High Voltage Engineering, 2021, 47(1):23-32.
[4]
孙庆凯, 王小君, 张义志, 等. 基于LSTM和多任务学习的综合能源系统多元负荷预测[J]. 电力系统自动化, 2021, 45(5):63-70.
SUN Qingkai, WANG Xiaojun, ZHANG Yizhi, et al. Multiple load prediction of integrated energy system based on long short term memory and multi-task learning[J]. Automation of Electric Power Systems, 2021, 45(5):63-70.
[5]
程浩忠, 胡枭, 王莉, 等. 区域综合能源系统规划研究综述[J]. 电力系统自动化, 2019, 43(7):2-13.
CHENG Haozhong, HU Xiao, WANG Li, et al. Review on research of regional integrated energy system planning[J]. Automation of Electric Power Systems, 2019, 43(7):2-13.
[6]
KHAN M, JAVAID N, IQBAL M N, et al. Load prediction based on multivariate time series forecasting for energy consumption and behavioral analytics[C]// Proceedings of the 12th International Conference on Complex,Intelligent,and Software Intensive Systems,Kunibiki Messe,Matsue,Japan, 2018.
[7]
KONG W, DONG Z Y, JIA Y W, et al. Short-term residential load forecasting based on LSTM recurrent neural network[J]. IEEE Transactions on Smart Grid, 2019, 10(1):841-851.
[8]
RAFI S H, NAHID-AL-MASOOD, DEEBA S R, et al. A short-term load forecasting method using integrated CNN and LSTM network[J]. IEEE Access, 2021,9:32436-32448.
[9]
TAN M, YUAN S, LI S, et al. Ultra-short-term industrial power demand forecasting using LSTM based hybrid ensemble learning[J]. IEEE Transactions on Power Systems, 2020,35:2937-2948.
[10]
卢武, 郑人杰, 赵文彬, 等. 基于EMD-GRU的短期电力负荷预测方法[J]. 电气传动, 2022, 52(21):74-80.
LU Wu, ZHENG Renjie, ZHAO Wenbin, et al. Short term power load forecasting method based on EMD-GRU[J]. Electric Drive, 2022, 52(21):74-80.
[11]
欧阳静, 杨吕, 尹康, 等. 基于ALIF-LSTM多任务学习的综合能源系统短期负荷预测[J]. 太阳能学报, 2022, 43(9):499-507.
OUYANG Jing, YANG , YIN Kang, et al. Short-term load forecasting of integrated energy systems based on ALIF-LSTM multi task learning[J]. Acta Energiae Solaris Sinica, 2022, 43(9):499-507.
[12]
SUN W, ZHANG Y, LI F. The neural network model based on PSO for short-term load forecasting[C]// International Conference on Machine Learning and Cybernetics, Dalian: IEEE, 2006:3069-3072.
[13]
ZHANG C, MING L, TANG Mingyang. BP neural network optimized with PSO algorithm for daily load forecasting[C]// 2008 International Conference on Information Management,Innovation Management and Industrial Engineering, Taipei: IEEE, 2008:82-85.
[14]
陆继翔, 张琪培, 杨志宏, 等. 基于CNN-LSTM混合神经网络模型的短期负荷预测方法[J]. 电力系统自动化, 2019, 43(8):131-137.
LU J X, ZHANG Q P, YANG Z H, et al. Short-term load forecasting method based on CNN-LSTM hybrid neural network model[J]. Automation of Electric Power Systems, 2019, 43(8):131-137.
[15]
顾吉鹏, 邵亮, 陆垂基, 等. 基于LSTNet模型的配电台区短期电力负荷预测研究[J]. 电气传动, 2023, 53(5):63-70.
GU Jipeng, SHAO Liang, LU Chuiji, et al. Research on short term power load prediction in distribution stations based on LSTNet model[J]. Electric Drive, 2023, 53(5):63-70.
[16]
赵洋, 王瀚墨, 康丽, 等. 基于时间卷积网络的短期电力负荷预测[J]. 电工技术学报, 2022, 37(5):1242-1251.
ZHAO Yang, WANG Hanmo, KANG Li, et al. Temporal convolution network-based short-term electrical load forecasting[J]. Transactions of China Electrotechnical Society, 2022, 37(5):1242-1251.
[17]
HANCOCK John, KHOSHGOFTAAR Taghi. Survey on categorical data for neural networks[J]. Journal of Big Data, 2020, 7(1):1-41.
[18]
MIWA M, BANSAL M. End-to-end relation extraction using LSTMs on sequences and tree structures[C]// Annual meeting of the Association for Computational Linguistics, 2016.
2025年第55卷第8期
PDF下载
207
99
引用本文
BibTeX
文章信息
doi: 10.19457/j.1001-2095.dqcd25612
  • 接收时间:2024-01-19
  • 首发时间:2025-10-29
  • 出版时间:2025-08-20
补充材料
相关文章
文章信息
作者
出版历史
  • 收稿日期:2024-01-19
  • 修回日期:2024-04-08
基金
雄安新区科技创新专项《面向雄安新区现代商业园区碳中和路径规划与智慧降碳关键技术研究与示范》(2022XAGG0116)
作者信息
    1.国网雄安综合能源服务有限公司,河北 保定 071700
    2.北京南瑞数字技术有限公司,北京 100193
参考文献
分享链接
https://castjournals.cast.org.cn/joweb/dqcd/CN/10.19457/j.1001-2095.dqcd25612
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