Article(id=1149776904888217655, tenantId=1146029695717560320, journalId=1146123166801305609, issueId=1149776900194791454, articleNumber=null, orderNo=null, doi=10.12404/j.issn.1671-1815.2404503, pmid=null, cstr=null, oa=null, hot=null, price=null, onlineType=0, articleFormat=0, articleType=null, articleTypeStr=research-article, receivedDate=1718553600000, receivedDateStr=2024-06-17, revisedDate=1730131200000, revisedDateStr=2024-10-29, acceptedDate=null, acceptedDateStr=null, onlineDate=1752057775946, onlineDateStr=2025-07-09, pubDate=1744905600000, pubDateStr=2025-04-18, doiRegisterDate=null, doiRegisterDateStr=null, onlineIssueDate=1752057775946, onlineIssueDateStr=2025-07-09, onlineJustAcceptDate=null, onlineJustAcceptDateStr=null, onlineFirstDate=null, onlineFirstDateStr=null, sourceXml=null, magXml=null, createTime=1752057775946, creator=13701087609, updateTime=1752057775946, updator=13701087609, issue=Issue{id=1149776900194791454, tenantId=1146029695717560320, journalId=1146123166801305609, year='2025', volume='25', issue='11', pageStart='4397', pageEnd='4826', issueExtLink='null', onlineDate='null', pubDate='null', beforeIssueId=null, nextIssueId=null, price=null, status=1, issueComplete=1, articleOrder=1, issueType=-1, specialIssue=0, createTime=1752057774827, creator=13701087609, updateTime=1768456666677, updator=13701087609, preIssue=null, nextIssue=null, ext={EN=IssueExt(id=1218558837930512931, tenantId=1146029695717560320, journalId=1146123166801305609, issueId=1149776900194791454, language=EN, specialIssueTitle=, coverIllustrator=, specialIssueEditor=, specialIssueAbout=), CN=IssueExt(id=1218558837930512932, tenantId=1146029695717560320, journalId=1146123166801305609, issueId=1149776900194791454, language=CN, specialIssueTitle=, coverIllustrator=, specialIssueEditor=, specialIssueAbout=)}, issueFiles=null}, startPage=4583, endPage=4597, ext={EN=ArticleExt(id=1149776905223761977, articleId=1149776904888217655, tenantId=1146029695717560320, journalId=1146123166801305609, language=EN, title=NGO-CNN-LSTM Power Load Short-term Combination Forecasting Model Based on ALIF-VMD Quadratic Decomposition, columnId=1156262733675876713, journalTitle=Science Technology and Engineering, columnName=Papers·Electrical Technology, runingTitle=null, highlight=null, articleAbstract=

Aiming at obvious load fluctuation trend, strong randomness and low accuracy caused by unreasonable parameter values of the prediction model involved into the power load forecasting process, a combined prediction model composing of ALIF (adaptive local iterative filtering), VMD (variational mode decomposition), NGO (northern goshawk optimization) and CNN-LSTM (convolutional neural networks - long short-term memory) was established. Firstly, CCM (convergent cross-mapping) method was used to identify the key factors affecting the power load. Secondly, an innovative combination of ALIF, NGO-based VMD and FE (fuzzy entropy) was employed for combinatorial decomposition and necessary recombination of original load sequence. Next, based on the modal components generated after decomposition and recombination, combined with optimal hyperparameter combination of CNN-LSTM determined by NGO method, an NGO-CNN-LSTM day-ahead power load combination prediction model with the high prediction accuracy, short training time and fast convergence speed was formulated. Compared with other benchmark models, the obtained results demonstrated that the proposed model has the better adaptability and prediction accuracy, and can provide important technical support for the safe, reliable and economical operation of power system.

, correspAuthors=Sheng-qiang GAO, 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=Lin ZHANG, Sheng-qiang GAO, Yu SONG, Shuai-yu BU, Wei YU), CN=ArticleExt(id=1149776930095985289, articleId=1149776904888217655, tenantId=1146029695717560320, journalId=1146123166801305609, language=CN, title=基于ALIF-VMD二次分解的NGO-CNN-LSTM电力负荷短期组合预测模型, columnId=1156262734506353627, journalTitle=科学技术与工程, columnName=论文·电工技术, runingTitle=null, highlight=null, articleAbstract=

针对电力负荷预测过程中普遍存在的负荷波动变化趋势明显、随机性强,以及预测模型的参数取值不合理导致的精度偏低问题,提出了一种基于ALIF-VMD(adaptive local iterative filtering - variational mode decomposition)二次分解和北方苍鹰优化算法(northern goshawk optimization,NGO)优化CNN-LSTM(convolutional neural networks - long short-term memory)的电力负荷组合预测模型,在使用交叉映射收敛方法(convergent cross-mapping,CCM)准确识别电力负荷的关键影响因素的基础上,创新性地联合使用ALIF、基于NGO的VMD和模糊熵(fuzzy entropy,FE)对原始负荷序列进行组合分解和必要的重组;针对分解和重组后生成的模态分量,结合NGO确定的CNN-LSTM模型最优超参数组合,建立预测精度高、训练时间短、收敛速度快的NGO-CNN-LSTM日前电力负荷组合预测模型。与其他基准模型的对比结果表明,该模型具有更好的适应性和预测精度,可为电力系统的安全、可靠、经济运行提供重要的技术支撑。

, correspAuthors=高胜强, authorNote=null, correspAuthorsNote=
* 高胜强(1980—),男,汉族,北京人,硕士,高级工程师。研究方向:电力负荷预测、电力系统规划。E-mail:
, copyrightStatement=null, copyrightOwner=null, extLink=null, articleAbsUrl=null, sourceXml=tMqm7hLHfIwdq+GAmc9w9Q==, magXml=lfuJsJkHJQQeZtv/szeptA==, pdfUrl=null, pdf=r2wCBMLlNCwyQB9w4RP/xw==, pdfFileSize=15728855, pdfExtLink=null, richHtmlUrl=null, mobilePdfUrl=null, reviewReport=null, pdfFirstPage=null, abstractGraph=dTbwjqkWZeSbndSNzRqMiQ==, abstractGraphContent=null, abstractVideo=null, citation=null, cebUrl=null, magXmlContent=gpEuaijYlhmgX+yLUvZOYg==, mapNumber=null, authorCompany=null, fund=null, authors=

张琳(1977—),男,汉族,北京人,硕士,高级工程师。研究方向:电力负荷预测、电力系统规划。E-mail:

, authorsList=张琳, 高胜强, 宋煜, 卜帅羽, 余伟)}, authors=[Author(id=1218843909418042344, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, orderNo=0, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=13611041997@139.com, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1218843909539677172, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, authorId=1218843909418042344, language=EN, stringName=Lin ZHANG, firstName=Lin, middleName=null, lastName=ZHANG, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1 State Grid Beijing Electric Power Company, Beijing 100031, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1218843909657117696, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, authorId=1218843909418042344, language=CN, stringName=张琳, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1 国网北京市电力公司, 北京 100031, bio={"content":"

张琳(1977—),男,汉族,北京人,硕士,高级工程师。研究方向:电力负荷预测、电力系统规划。E-mail:

"}, bioImg=null, bioContent=

张琳(1977—),男,汉族,北京人,硕士,高级工程师。研究方向:电力负荷预测、电力系统规划。E-mail:

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1218843909095080899, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, xref=1, ext=[AuthorCompanyExt(id=1218843909107663813, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, companyId=1218843909095080899, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 State Grid Beijing Electric Power Company, Beijing 100031, China), AuthorCompanyExt(id=1218843909120246726, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, companyId=1218843909095080899, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 国网北京市电力公司, 北京 100031)])]), Author(id=1218843909774557194, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, orderNo=1, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=1780404374@qq.com, emailSecond=null, emailThird=null, correspondingAuthor=1, authorType=1, ext={EN=AuthorExt(id=1218843909938135068, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, authorId=1218843909774557194, language=EN, stringName=Sheng-qiang GAO, firstName=Sheng-qiang, middleName=null, lastName=GAO, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, *, address=1 State Grid Beijing Electric Power Company, Beijing 100031, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1218843910160433196, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, authorId=1218843909774557194, language=CN, stringName=高胜强, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, *, address=1 国网北京市电力公司, 北京 100031, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1218843909095080899, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, xref=1, ext=[AuthorCompanyExt(id=1218843909107663813, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, companyId=1218843909095080899, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 State Grid Beijing Electric Power Company, Beijing 100031, China), AuthorCompanyExt(id=1218843909120246726, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, companyId=1218843909095080899, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 国网北京市电力公司, 北京 100031)])]), Author(id=1218843910269485115, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, 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=1218843910399508551, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, authorId=1218843910269485115, language=EN, stringName=Yu SONG, firstName=Yu, middleName=null, lastName=SONG, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2 The College of Environmental Science and Engineering, North China Electric Power University, Beijing 102206, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1218843910533726295, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, authorId=1218843910269485115, language=CN, stringName=宋煜, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2 华北电力大学环境科学与工程学院, 北京 102206, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1218843909233492949, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, xref=2, ext=[AuthorCompanyExt(id=1218843909246075862, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, companyId=1218843909233492949, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 The College of Environmental Science and Engineering, North China Electric Power University, Beijing 102206, China), AuthorCompanyExt(id=1218843909254464473, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, companyId=1218843909233492949, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 华北电力大学环境科学与工程学院, 北京 102206)])]), Author(id=1218843910743441520, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, 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=1218843910907019394, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, authorId=1218843910743441520, language=EN, stringName=Shuai-yu BU, firstName=Shuai-yu, middleName=null, lastName=BU, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1 State Grid Beijing Electric Power Company, Beijing 100031, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1218843911074791575, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, authorId=1218843910743441520, language=CN, stringName=卜帅羽, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1 国网北京市电力公司, 北京 100031, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1218843909095080899, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, xref=1, ext=[AuthorCompanyExt(id=1218843909107663813, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, companyId=1218843909095080899, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 State Grid Beijing Electric Power Company, Beijing 100031, China), AuthorCompanyExt(id=1218843909120246726, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, companyId=1218843909095080899, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 国网北京市电力公司, 北京 100031)])]), Author(id=1218843911200620705, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, 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=1218843911330644149, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, authorId=1218843911200620705, language=EN, stringName=Wei YU, firstName=Wei, middleName=null, lastName=YU, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1 State Grid Beijing Electric Power Company, Beijing 100031, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1218843911464861895, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, authorId=1218843911200620705, language=CN, stringName=余伟, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1 国网北京市电力公司, 北京 100031, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1218843909095080899, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, xref=1, ext=[AuthorCompanyExt(id=1218843909107663813, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, companyId=1218843909095080899, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 State Grid Beijing Electric Power Company, Beijing 100031, China), AuthorCompanyExt(id=1218843909120246726, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, companyId=1218843909095080899, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 国网北京市电力公司, 北京 100031)])])], keywords=[Keyword(id=1218843911720714463, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, language=EN, orderNo=1, keyword=load forecasting), Keyword(id=1218843911896875246, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, language=EN, orderNo=2, keyword=sequence decomposition and recombination), Keyword(id=1218843912014315776, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, language=EN, orderNo=3, keyword=northern goshawk optimization), Keyword(id=1218843912114979081, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, language=EN, orderNo=4, keyword=convolutional neural network-long short-term memory neural network model), Keyword(id=1218843912253391128, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, language=CN, orderNo=1, keyword=负荷预测), Keyword(id=1218843912404386090, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, language=CN, orderNo=2, keyword=序列分解与重组), Keyword(id=1218843912509243706, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, language=CN, orderNo=3, keyword=北方苍鹰算法), Keyword(id=1218843912618295620, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, language=CN, orderNo=4, keyword=卷积神经网络-长短期记忆神经网络模型)], refs=[Reference(id=1218843917064258208, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, doi=null, pmid=null, pmcid=null, year=2024, volume=null, issue=4, pageStart=215, pageEnd=218, url=null, language=null, rfNumber=[1], rfOrder=0, authorNames=王晨, 李又轩, 吴其琦, journalName=水电能源科学, refType=null, unstructuredReference=王晨, 李又轩, 吴其琦, 等. 基于SVM-STL-LSTM的区域短期电力负荷预测研究[J]. 水电能源科学, 2024(4): 215-218., articleTitle=基于SVM-STL-LSTM的区域短期电力负荷预测研究, refAbstract=null), Reference(id=1218843917185893034, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, doi=null, pmid=null, pmcid=null, year=2024, volume=null, issue=4, pageStart=215, pageEnd=218, url=null, language=null, rfNumber=[1], rfOrder=1, authorNames=Wang Chen, Li Youxuan, Wu Qiqi, journalName=Water Resources and Power, refType=null, unstructuredReference=Wang Chen, Li Youxuan, Wu Qiqi, et al. Research on regional short-term load forecasting based on SVM-STL-LSTM[J]. Water Resources and Power, 2024(4): 215-218., articleTitle=Research on regional short-term load forecasting based on SVM-STL-LSTM, refAbstract=null), Reference(id=1218843917286556339, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, doi=null, pmid=null, pmcid=null, year=2024, volume=null, issue=6, pageStart=2574, pageEnd=2583, url=null, language=null, rfNumber=[2], rfOrder=2, authorNames=石卓见, 冉启武, 徐福聪, journalName=电网技术, refType=null, unstructuredReference=石卓见, 冉启武, 徐福聪. 基于聚合二次模态分解及Informer的短期负荷预测[J]. 电网技术, 2024(6): 2574-2583., articleTitle=基于聚合二次模态分解及Informer的短期负荷预测, refAbstract=null), Reference(id=1218843917408191165, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, doi=null, pmid=null, pmcid=null, year=2024, volume=null, issue=6, pageStart=2574, pageEnd=2583, url=null, language=null, rfNumber=[2], rfOrder=3, authorNames=Shi Zhuojian, Ran Qiwu, Xu Fucong, journalName=Power System Technology, refType=null, unstructuredReference=Shi Zhuojian, Ran Qiwu, Xu Fucong. Short-term load forecasting based on aggregated secondary decomposition and Informer[J]. Power System Technology, 2024(6): 2574-2583., articleTitle=Short-term load forecasting based on aggregated secondary decomposition and Informer, refAbstract=null), Reference(id=1218843917546603205, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, doi=null, pmid=null, pmcid=null, year=2024, volume=33, issue=5, pageStart=47, pageEnd=56, url=null, language=null, rfNumber=[3], rfOrder=4, authorNames=陈万志, 张思维, 王天元, journalName=计算机系统应用, refType=null, unstructuredReference=陈万志, 张思维, 王天元. 强化数据预处理的BLSTNet-CBAM短期电力负荷预测[J]. 计算机系统应用, 2024, 33(5): 47-56., articleTitle=强化数据预处理的BLSTNet-CBAM短期电力负荷预测, refAbstract=null), Reference(id=1218843917676626634, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, doi=null, pmid=null, pmcid=null, year=2024, volume=33, issue=5, pageStart=47, pageEnd=56, url=null, language=null, rfNumber=[3], rfOrder=5, authorNames=Chen Wanzhi, Zhang Siwei, Wang Tianyuan, journalName=Computer Systems & Applications, refType=null, unstructuredReference=Chen Wanzhi, Zhang Siwei, Wang Tianyuan. Enhanced data preprocessing for BLSTNet-CBAM short-term power load forecasting[J]. Computer Systems & Applications, 2024, 33(5): 47-56., articleTitle=Enhanced data preprocessing for BLSTNet-CBAM short-term power load forecasting, refAbstract=null), Reference(id=1218843917806650067, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=1, pageEnd=8, url=null, language=null, rfNumber=[4], rfOrder=6, authorNames=葛亚明, 仇晨光, 谢丽荣, journalName=现代电力, refType=null, unstructuredReference=葛亚明, 仇晨光, 谢丽荣, 等. 基于K-means聚类与LSTM模型的多能源耦合电力负荷预测[J/OL]. 现代电力: 1-8[2024-04-29]. https://doi.org/10.19725/j.cnki.1007-2322.2023.0110., articleTitle=基于K-means聚类与LSTM模型的多能源耦合电力负荷预测, refAbstract=null), Reference(id=1218843917919896282, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=1, pageEnd=8, url=null, language=null, rfNumber=[4], rfOrder=7, authorNames=Ge Yaming, Chou Chenguang, Xie Lirong, journalName=Modern Electric Power, refType=null, unstructuredReference=Ge Yaming, Chou Chenguang, Xie Lirong, et al. Research on Multi-energy coupled power load prediction based on K-means clustering and LSTM model[J]. Modern Electric Power, 1-8[2024-04-29]. https://doi.org/10.19725/j.cnki.1007-2322.2023.0110., articleTitle=Research on Multi-energy coupled power load prediction based on K-means clustering and LSTM model, refAbstract=null), Reference(id=1218843918033142495, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, doi=null, pmid=null, pmcid=null, year=2019, volume=43, issue=5, pageStart=53, pageEnd=58, url=null, language=null, rfNumber=[5], rfOrder=8, authorNames=王增平, 赵兵, 纪维佳, journalName=电力系统自动化, refType=null, unstructuredReference=王增平, 赵兵, 纪维佳, 等. 基于GRU-NN模型的短期负荷预测方法[J]. 电力系统自动化, 2019, 43(5): 53-58., articleTitle=基于GRU-NN模型的短期负荷预测方法, refAbstract=null), Reference(id=1218843918117028584, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, doi=null, pmid=null, pmcid=null, year=2019, volume=43, issue=5, pageStart=53, pageEnd=58, url=null, language=null, rfNumber=[5], rfOrder=9, authorNames=Wang Zengping, Zhao Bing, Ji Weijia, journalName=Automation of Electric Power Systems, refType=null, unstructuredReference=Wang Zengping, Zhao Bing, Ji Weijia, et al. Short-term load forecasting method based on GRU-NN model[J]. Automation of Electric Power Systems, 2019, 43(5): 53-58., articleTitle=Short-term load forecasting method based on GRU-NN model, refAbstract=null), Reference(id=1218843918242857713, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=1, pageEnd=11, url=http://kns.cnki.net/kcms/detail/23.1202.th.20240321.1323.002.html, language=null, rfNumber=[6], rfOrder=10, authorNames=刘伟, 蔡东升, 冯付勇, journalName=电测与仪表, refType=null, unstructuredReference=刘伟, 蔡东升, 冯付勇, 等. 基于DCT-CNN-GRU的短期电力负荷预测研究[J/OL]. 电测与仪表: 1-11[2024-04-29]. http://kns.cnki.net/kcms/detail/23.1202.th.20240321.1323.002.html., articleTitle=基于DCT-CNN-GRU的短期电力负荷预测研究, refAbstract=null), Reference(id=1218843918360298233, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=1, pageEnd=11, url=http://kns.cnki.net/kcms/detail/23.1202.th.20240321.1323.002.html, language=null, rfNumber=[6], rfOrder=11, authorNames=Liu Wei, Cai Dongsheng, Feng Fuyong, journalName=Electrical Measurement & Instrumentation, refType=null, unstructuredReference=Liu Wei, Cai Dongsheng, Feng Fuyong, et al. Research on short-term electric load forecasting based on DCT-CNN-GRU[J]. Electrical Measurement & Instrumentation, 1-11[2024-04-29]. http://kns.cnki.net/kcms/detail/23.1202.th.20240321.1323.002.html., articleTitle=Research on short-term electric load forecasting based on DCT-CNN-GRU, refAbstract=null), Reference(id=1218843918469350142, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, doi=null, pmid=null, pmcid=null, year=2023, volume=35, issue=2, pageStart=254, pageEnd=262, url=null, language=null, rfNumber=[7], rfOrder=12, authorNames=张文栋, 刘子琨, 梁涛, journalName=重庆邮电大学学报(自然科学版), refType=null, unstructuredReference=张文栋, 刘子琨, 梁涛, 等. 基于CNN-LSTM的综合能源系统负荷预测模型[J]. 重庆邮电大学学报(自然科学版), 2023, 35(2): 254-262., articleTitle=基于CNN-LSTM的综合能源系统负荷预测模型, refAbstract=null), Reference(id=1218843918637122313, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, doi=null, pmid=null, pmcid=null, year=2023, volume=35, issue=2, pageStart=254, pageEnd=262, url=null, language=null, rfNumber=[7], rfOrder=13, authorNames=Zhang Wendong, Liu Zikun, Liang Tao, journalName=Journal of Chongqing University of Posts and Telecommunications( Natural Seience Edition, refType=null, unstructuredReference=Zhang Wendong, Liu Zikun, Liang Tao, et al. Load prediction model of integrated energy system based on CNN-LSTM[J]. Journal of Chongqing University of Posts and Telecommunications( Natural Seience Edition, 2023, 35(2): 254-262., articleTitle=Load prediction model of integrated energy system based on CNN-LSTM, refAbstract=null), Reference(id=1218843918762951440, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, doi=null, pmid=null, pmcid=null, year=2023, volume=null, issue=null, pageStart=1, pageEnd=6, url=null, language=null, rfNumber=[8], rfOrder=14, authorNames=Huy T H B, Vo D N, Nguyen K P, journalName=2023 Asia Meeting on Environment and Electrical Engineering (EEE-AM), refType=null, unstructuredReference=Huy T H B, Vo D N, Nguyen K P, et al. Short-term load forecasting in power system using CNN-LSTM neural network[C]// 2023 Asia Meeting on Environment and Electrical Engineering (EEE-AM). Hanoi: IEEE, 2023: 1-6., articleTitle=Short-term load forecasting in power system using CNN-LSTM neural network, refAbstract=null), Reference(id=1218843918876197651, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=1, pageEnd=9, url=http://kns.cnki.net/kcms/detail/51.1714.TB.20240328.1617.002.html, language=null, rfNumber=[9], rfOrder=15, authorNames=白宇, 薛贵军, 谢文举, journalName=中国测试, refType=null, unstructuredReference=白宇, 薛贵军, 谢文举, 等. 改进GJO优化CNN-BiLSTM的热负荷预测模型[J/OL]. 中国测试: 1-9[2024-04-29]. http://kns.cnki.net/kcms/detail/51.1714.TB.20240328.1617.002.html., articleTitle=改进GJO优化CNN-BiLSTM的热负荷预测模型, refAbstract=null), Reference(id=1218843919010415390, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=1, pageEnd=9, url=http://kns.cnki.net/kcms/detail/51.1714.TB.20240328.1617.002.html, language=null, rfNumber=[9], rfOrder=16, authorNames=Bai Yu, Xue Guijun, Xie Wenju, journalName=China Measurement & Test, refType=null, unstructuredReference=Bai Yu, Xue Guijun, Xie Wenju, et al. Improved GJO optimized CNN-BiLSTM heat load prediction model[J]. China Measurement & Test. 1-9[2024-04-29]. http://kns.cnki.net/kcms/detail/51.1714.TB.20240328.1617.002.html., articleTitle=Improved GJO optimized CNN-BiLSTM heat load prediction model, refAbstract=null), Reference(id=1218843919153021731, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, doi=null, pmid=null, pmcid=null, year=2023, volume=35, issue=null, pageStart=2679, pageEnd=2700, url=null, language=null, rfNumber=[10], rfOrder=17, authorNames=Kiruthiga D, Manikandan V, journalName=Neural Computing & Applications, refType=null, unstructuredReference=Kiruthiga D, Manikandan V. Levy flight-particle swarm optimization-assisted BiLSTM plus dropout deep learning model for short-term load forecasting[J]. Neural Computing & Applications, 2023, 35: 2679-2700., articleTitle=Levy flight-particle swarm optimization-assisted BiLSTM plus dropout deep learning model for short-term load forecasting, refAbstract=null), Reference(id=1218843919249490729, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, doi=null, pmid=null, pmcid=null, year=2024, volume=24, issue=22, pageStart=9408, pageEnd=9416, url=null, language=null, rfNumber=[11], rfOrder=18, authorNames=周磊, 竺筱晶, journalName=科学技术与工程, refType=null, unstructuredReference=周磊, 竺筱晶. 基于MA-CNN-LSTM和自注意力机制的单变量短期电力负荷预测[J]. 科学技术与工程, 2024, 24(22): 9408-9416., articleTitle=基于MA-CNN-LSTM和自注意力机制的单变量短期电力负荷预测, refAbstract=null), Reference(id=1218843919379514158, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, doi=null, pmid=null, pmcid=null, year=2024, volume=24, issue=22, pageStart=9408, pageEnd=9416, url=null, language=null, rfNumber=[11], rfOrder=19, authorNames=Zhou Lei, Zhu Xiaojing, journalName=Science Technology and Engineering, refType=null, unstructuredReference=Zhou Lei, Zhu Xiaojing. Univariate short-term electrical load based on MA-CNN-LSTM-Self attention[J]. Science Technology and Engineering, 2024, 24(22): 9408-9416., articleTitle=Univariate short-term electrical load based on MA-CNN-LSTM-Self attention, refAbstract=null), Reference(id=1218843919509537586, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, doi=null, pmid=null, pmcid=null, year=2024, volume=24, issue=7, pageStart=2759, pageEnd=2766, url=null, language=null, rfNumber=[12], rfOrder=20, authorNames=张淑娴, 江文韬, 陈玉花, journalName=科学技术与工程, refType=null, unstructuredReference=张淑娴, 江文韬, 陈玉花, 等. 基于二次模态分解的LSTM短期电力负荷预测[J]. 科学技术与工程, 2024, 24(7): 2759-2766., articleTitle=基于二次模态分解的LSTM短期电力负荷预测, refAbstract=null), Reference(id=1218843919635366712, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, doi=null, pmid=null, pmcid=null, year=2024, volume=24, issue=7, pageStart=2759, pageEnd=2766, url=null, language=null, rfNumber=[12], rfOrder=21, authorNames=Zhang Shuxian, Jiang Wentao, Chen Yuhua, journalName=Science Technology and Engineering, refType=null, unstructuredReference=Zhang Shuxian, Jiang Wentao, Chen Yuhua, et al. Short-term power load forecasting based on quadratic mode decomposition through LSTM[J]. Science Technology and Engineering, 2024, 24(7): 2759-2766., articleTitle=Short-term power load forecasting based on quadratic mode decomposition through LSTM, refAbstract=null), Reference(id=1218843919731835710, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, doi=null, pmid=null, pmcid=null, year=2023, volume=10, issue=null, pageStart=1076529, pageEnd=null, url=null, language=null, rfNumber=[13], rfOrder=22, authorNames=Wang N, Li Z, journalName=Frontiers in Energy Research, refType=null, unstructuredReference=Wang N, Li Z. Short term power load forecasting based on BES-VMD and CNN-Bi-LSTM method with error correction. Front[J]. Frontiers in Energy Research, 2023, 10: 1076529., articleTitle=Short term power load forecasting based on BES-VMD and CNN-Bi-LSTM method with error correction. Front, refAbstract=null), Reference(id=1218843919853470530, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, doi=null, pmid=null, pmcid=null, year=2023, volume=46, issue=9, pageStart=70, pageEnd=71, url=null, language=null, rfNumber=[14], rfOrder=23, authorNames=程友杰, journalName=山西冶金, refType=null, unstructuredReference=程友杰. 基于ALIF+MCKD方法的冷轧机轴承振动故障信号处理[J]. 山西冶金, 2023, 46(9): 70-71, 74., articleTitle=基于ALIF+MCKD方法的冷轧机轴承振动故障信号处理, refAbstract=null), Reference(id=1218843919949939527, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, doi=null, pmid=null, pmcid=null, year=2023, volume=46, issue=9, pageStart=70, pageEnd=71, url=null, language=null, rfNumber=[14], rfOrder=24, authorNames=Cheng Youjie, journalName=Shanxi Metallurgy, refType=null, unstructuredReference=Cheng Youjie. Vibration fault signal processing of bearing for cold rolling mill based on ALIF+MCKD method[J]. Shanxi Metallurgy, 2023, 46(9): 70-71, 74., articleTitle=Vibration fault signal processing of bearing for cold rolling mill based on ALIF+MCKD method, refAbstract=null), Reference(id=1218843920079962959, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, doi=null, pmid=null, pmcid=null, year=2023, volume=null, issue=4, pageStart=54, pageEnd=60, url=null, language=null, rfNumber=[15], rfOrder=25, authorNames=邓博文, 肖伸平, 廖世英, journalName=控制与信息技术, refType=null, unstructuredReference=邓博文, 肖伸平, 廖世英. 二次分解组合CNN-LSTM的短期负荷预测[J]. 控制与信息技术, 2023(4): 54-60., articleTitle=二次分解组合CNN-LSTM的短期负荷预测, refAbstract=null), Reference(id=1218843920193209172, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, doi=null, pmid=null, pmcid=null, year=2023, volume=null, issue=4, pageStart=54, pageEnd=60, url=null, language=null, rfNumber=[15], rfOrder=26, authorNames=Deng Bowen, Xiao Shenping, Liao Shiying, journalName=Control and Information Technology, refType=null, unstructuredReference=Deng Bowen, Xiao Shenping, Liao Shiying. Short-term load forecasting based on CNN-LSTM with quadratic decomposition combined[J]. Control and Information Technology, 2023(4): 54-60., articleTitle=Short-term load forecasting based on CNN-LSTM with quadratic decomposition combined, refAbstract=null), Reference(id=1218843920306455382, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, doi=null, pmid=null, pmcid=null, year=2024, volume=24, issue=16, pageStart=6734, pageEnd=6741, url=null, language=null, rfNumber=[16], rfOrder=27, authorNames=肖威, 方娜, 邓心, journalName=科学技术与工程, refType=null, unstructuredReference=肖威, 方娜, 邓心. 基于VMD-LSTM-IPSO-GRU的电力负荷预测[J]. 科学技术与工程, 2024, 24(16): 6734-6741., articleTitle=基于VMD-LSTM-IPSO-GRU的电力负荷预测, refAbstract=null), Reference(id=1218843920415507291, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, doi=null, pmid=null, pmcid=null, year=2024, volume=24, issue=16, pageStart=6734, pageEnd=6741, url=null, language=null, rfNumber=[16], rfOrder=28, authorNames=Xiao Wei, Fang Na, Deng Xin, journalName=Science Technology and Engineering, refType=null, unstructuredReference=Xiao Wei, Fang Na, Deng Xin. Short-term load forecasting based on VMD-LSTM-IPSO-GRU[J]. Science Technology and Engineering, 2024, 24(16): 6734-6741., articleTitle=Short-term load forecasting based on VMD-LSTM-IPSO-GRU, refAbstract=null), Reference(id=1218843920570696553, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, doi=null, pmid=null, pmcid=null, year=2023, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[17], rfOrder=29, authorNames=李玉莲, journalName=基于收敛交叉映射方法的3E系统因果推断, refType=null, unstructuredReference=李玉莲. 基于收敛交叉映射方法的3E系统因果推断[D]. 武汉: 武汉科技大学, 2023., articleTitle=null, refAbstract=null), Reference(id=1218843920688137069, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, doi=null, pmid=null, pmcid=null, year=2023, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[17], rfOrder=30, authorNames=Li Yulian, journalName=Causal inference to the 3E system via convergent cross mapping, refType=null, unstructuredReference=Li Yulian. Causal inference to the 3E system via convergent cross mapping[D]. Wuhan: Wuhan University of Science and Technology, 2023., articleTitle=null, refAbstract=null), Reference(id=1218843920839132022, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, doi=null, pmid=null, pmcid=null, year=2020, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[18], rfOrder=31, authorNames=陈俊, journalName=基于格兰杰方法和收敛交叉映射方法的因果分析研究, refType=null, unstructuredReference=陈俊. 基于格兰杰方法和收敛交叉映射方法的因果分析研究[D]. 上海: 华东师范大学, 2020., articleTitle=null, refAbstract=null), Reference(id=1218843920931406717, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, doi=null, pmid=null, pmcid=null, year=2020, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[18], rfOrder=32, authorNames=Chen Jun, journalName=Research on causality analysis based on granger method and convergent cross mapping method, refType=null, unstructuredReference=Chen Jun. Research on causality analysis based on granger method and convergent cross mapping method[D]. Shanghai: East China Normal University, 2020., articleTitle=null, refAbstract=null), Reference(id=1218843921053041541, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, doi=null, pmid=null, pmcid=null, year=2023, volume=51, issue=5, pageStart=196, pageEnd=201, url=null, language=null, rfNumber=[19], rfOrder=33, authorNames=陈向俊, 傅军平, 于晓, journalName=机床与液压, refType=null, unstructuredReference=陈向俊, 傅军平, 于晓, 等. 基于ALIF和ISOMAP的机械设备故障识别方法研究[J]. 机床与液压, 2023, 51(5): 196-201., articleTitle=基于ALIF和ISOMAP的机械设备故障识别方法研究, refAbstract=null), Reference(id=1218843921153704840, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, doi=null, pmid=null, pmcid=null, year=2023, volume=51, issue=5, pageStart=196, pageEnd=201, url=null, language=null, rfNumber=[19], rfOrder=34, authorNames=Chen Xiangjun, Fu Junping, Yu Xiao, journalName=Machine Tool & Hydraulics, refType=null, unstructuredReference=Chen Xiangjun, Fu Junping, Yu Xiao, et al. Research on fault recognition method of mechanical equipment based on ALIF and ISOMAP[J]. Machine Tool & Hydraulics, 2023, 51(5): 196-201., articleTitle=Research on fault recognition method of mechanical equipment based on ALIF and ISOMAP, refAbstract=null), Reference(id=1218843921279533964, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, doi=null, pmid=null, pmcid=null, year=2022, volume=43, issue=9, pageStart=499, pageEnd=507, url=null, language=null, rfNumber=[20], rfOrder=35, 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=1218843921396974482, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, doi=null, pmid=null, pmcid=null, year=2022, volume=43, issue=9, pageStart=499, pageEnd=507, url=null, language=null, rfNumber=[20], rfOrder=36, 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 method for integrated energy system based on ALIF-LSTM and MULTI-TASK learing[J]. Acta Energiae Solaris Sinica, 2022, 43(9): 499-507., articleTitle=Short-term load forecasting method for integrated energy system based on ALIF-LSTM and MULTI-TASK learing, refAbstract=null), Reference(id=1218843921535386522, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, doi=null, pmid=null, pmcid=null, year=2023, volume=23, issue=6, pageStart=2428, pageEnd=2437, url=null, language=null, rfNumber=[21], rfOrder=37, authorNames=宋江涛, 崔双喜, 刘洪广, journalName=科学技术与工程, refType=null, unstructuredReference=宋江涛, 崔双喜, 刘洪广. 基于二次分解NGO-VMD残差项与长短时记忆神经网络的超短期风功率预测[J]. 科学技术与工程, 2023, 23(6): 2428-2437., articleTitle=基于二次分解NGO-VMD残差项与长短时记忆神经网络的超短期风功率预测, refAbstract=null), Reference(id=1218843921631855515, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, doi=null, pmid=null, pmcid=null, year=2023, volume=23, issue=6, pageStart=2428, pageEnd=2437, url=null, language=null, rfNumber=[21], rfOrder=38, authorNames=Song Jiangtao, Cui Shuangxi, Liu Hongguang, journalName=Science Technology and Engineering, refType=null, unstructuredReference=Song Jiangtao, Cui Shuangxi, Liu Hongguang. Ultra-short-term wind power prediction based on the quadratic decomposition NGO-VMD residual term and long short-term memory(LSTM)[J]. Science Technology and Engineering, 2023, 23(6): 2428-2437., articleTitle=Ultra-short-term wind power prediction based on the quadratic decomposition NGO-VMD residual term and long short-term memory(LSTM), refAbstract=null), Reference(id=1218843921740907425, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, doi=null, pmid=null, pmcid=null, year=2023, volume=13, issue=11, pageStart=135, pageEnd=141, url=null, language=null, rfNumber=[22], rfOrder=39, authorNames=蒲维, 杨毅强, 张渊博, journalName=智能计算机与应用, refType=null, unstructuredReference=蒲维, 杨毅强, 张渊博, 等. 基于NGO-VMD-FCBF-Informer的电力负荷组合预测模型[J]. 智能计算机与应用, 2023, 13(11): 135-141., articleTitle=基于NGO-VMD-FCBF-Informer的电力负荷组合预测模型, refAbstract=null), Reference(id=1218843921833182120, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, doi=null, pmid=null, pmcid=null, year=2023, volume=13, issue=11, pageStart=135, pageEnd=141, url=null, language=null, rfNumber=[22], rfOrder=40, authorNames=Pu Wei, Yang Yiqiang, Zhang Yuanbo, journalName=Intelligent Computer and Applications, refType=null, unstructuredReference=Pu Wei, Yang Yiqiang, Zhang Yuanbo, et al. Powerload forecasting based on NGO-VMD-FCBF-informer[J]. Intelligent Computer and Applications, 2023, 13(11): 135-141., articleTitle=Powerload forecasting based on NGO-VMD-FCBF-informer, refAbstract=null), Reference(id=1218843921938039722, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, doi=null, pmid=null, pmcid=null, year=2024, volume=52, issue=8, pageStart=123, pageEnd=133, url=null, language=null, rfNumber=[23], rfOrder=41, authorNames=刘杰, 从兰美, 夏远洋, journalName=电力系统保护与控制, refType=null, unstructuredReference=刘杰, 从兰美, 夏远洋, 等. 基于DBO-VMD和IWOA-BILSTM神经网络组合模型的短期电力负荷预测[J]. 电力系统保护与控制, 2024, 52(8): 123-133., articleTitle=基于DBO-VMD和IWOA-BILSTM神经网络组合模型的短期电力负荷预测, refAbstract=null), Reference(id=1218843922030314411, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, doi=null, pmid=null, pmcid=null, year=2024, volume=52, issue=8, pageStart=123, pageEnd=133, url=null, language=null, rfNumber=[23], rfOrder=42, authorNames=Liu Jie, Cong Meilan, Xia Yuanyang, journalName=Power System Protection and Control, refType=null, unstructuredReference=Liu Jie, Cong Meilan, Xia Yuanyang, et al. Short-term power load prediction based on DBO-VMD and an IWOA-BILSTM neural network combination model[J]. Power System Protection and Control, 2024, 52(8): 123-133., articleTitle=Short-term power load prediction based on DBO-VMD and an IWOA-BILSTM neural network combination model, refAbstract=null), Reference(id=1218843922122589105, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, doi=null, pmid=null, pmcid=null, year=2022, volume=50, issue=13, pageStart=126, pageEnd=133, url=null, language=null, rfNumber=[24], rfOrder=43, authorNames=杨海柱, 田馥铭, 张鹏, journalName=电力系统保护与控制, refType=null, unstructuredReference=杨海柱, 田馥铭, 张鹏, 等. 基于CEEMD-FE和AOA-LSSVM的短期电力负荷预测[J]. 电力系统保护与控制, 2022, 50(13): 126-133., articleTitle=基于CEEMD-FE和AOA-LSSVM的短期电力负荷预测, refAbstract=null), Reference(id=1218843922240029620, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, doi=null, pmid=null, pmcid=null, year=2022, volume=50, issue=13, pageStart=126, pageEnd=133, url=null, language=null, rfNumber=[24], rfOrder=44, authorNames=Yang Haizhu, Tian Fuming, Zhang Peng, journalName=Power System Protection and Control, refType=null, unstructuredReference=Yang Haizhu, Tian Fuming, Zhang Peng, et al. Short-term load forecasting based on CEEMD-FE-AOA-LSSVM[J]. Power System Protection and Control, 2022, 50(13): 126-133., articleTitle=Short-term load forecasting based on CEEMD-FE-AOA-LSSVM, refAbstract=null), Reference(id=1218843922319721398, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, doi=null, pmid=null, pmcid=null, year=2023, volume=42, issue=2, pageStart=80, pageEnd=87, url=null, language=null, rfNumber=[25], rfOrder=45, authorNames=高凯悦, 牟莉, journalName=国外电子测量技术, refType=null, unstructuredReference=高凯悦, 牟莉. 基于二次分解和GRU-attention的时间序列预测研究[J]. 国外电子测量技术, 2023, 42(2): 80-87., articleTitle=基于二次分解和GRU-attention的时间序列预测研究, refAbstract=null), Reference(id=1218843922407801784, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, doi=null, pmid=null, pmcid=null, year=2023, volume=42, issue=2, pageStart=80, pageEnd=87, url=null, language=null, rfNumber=[25], rfOrder=46, authorNames=Gao Kaiyue, Mou Li, journalName=Foreign Electronic Measurement Technology, refType=null, unstructuredReference=Gao Kaiyue, Mou Li. Research on time series forecasting based on quadratic decomposition and GRU-attention[J]. Foreign Electronic Measurement Technology, 2023, 42(2): 80-87, articleTitle=Research on time series forecasting based on quadratic decomposition and GRU-attention, refAbstract=null), Reference(id=1218843922487493564, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, doi=null, pmid=null, pmcid=null, year=2022, volume=54, issue=4, pageStart=86, pageEnd=94, url=null, language=null, rfNumber=[26], rfOrder=47, authorNames=刘旭丽, 莫毓昌, 吴哲, journalName=郑州大学学报(理学版), refType=null, unstructuredReference=刘旭丽, 莫毓昌, 吴哲, 等. 基于DWT-CNN-LSTM的超短期光伏发电功率预测[J]. 郑州大学学报(理学版), 2022, 54(4): 86-94., articleTitle=基于DWT-CNN-LSTM的超短期光伏发电功率预测, refAbstract=null), Reference(id=1218843922571379648, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, doi=null, pmid=null, pmcid=null, year=2022, volume=54, issue=4, pageStart=86, pageEnd=94, url=null, language=null, rfNumber=[26], rfOrder=48, authorNames=Liu Xuli, Mo Yuchang, Wu Zhe, journalName=Journal of Zhengzhou University(Natural Science Edition), refType=null, unstructuredReference=Liu Xuli, Mo Yuchang, Wu Zhe, et al. Super-short-term photovoltaic power forecasting based on DWT-CNN-LSTM[J]. Journal of Zhengzhou University(Natural Science Edition), 2022, 54(4): 86-94., articleTitle=Super-short-term photovoltaic power forecasting based on DWT-CNN-LSTM, refAbstract=null)], funds=[Fund(id=1218843916850348688, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, awardId=520206240001, language=CN, fundingSource=国网北京市电力公司科技项目(520206240001), fundOrder=null, country=null)], companyList=[AuthorCompany(id=1218843909095080899, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, xref=1, ext=[AuthorCompanyExt(id=1218843909107663813, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, companyId=1218843909095080899, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 State Grid Beijing Electric Power Company, Beijing 100031, China), AuthorCompanyExt(id=1218843909120246726, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, companyId=1218843909095080899, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 国网北京市电力公司, 北京 100031)]), AuthorCompany(id=1218843909233492949, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, xref=2, ext=[AuthorCompanyExt(id=1218843909246075862, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, companyId=1218843909233492949, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 The College of Environmental Science and Engineering, North China Electric Power University, Beijing 102206, China), AuthorCompanyExt(id=1218843909254464473, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, companyId=1218843909233492949, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 华北电力大学环境科学与工程学院, 北京 102206)])], figs=[ArticleFig(id=1218843912848982364, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, language=EN, label=Fig.1, caption=Overall technical process, figureFileSmall=wZZ3visllQsk6px7uI0hrQ==, figureFileBig=kXBI0jAKLTTl8YLd9eHfBQ==, tableContent=null), ArticleFig(id=1218843912962228582, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, language=CN, label=图1, caption=总体技术流程, figureFileSmall=wZZ3visllQsk6px7uI0hrQ==, figureFileBig=kXBI0jAKLTTl8YLd9eHfBQ==, tableContent=null), ArticleFig(id=1218843913125806451, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, language=EN, label=Fig.2, caption=The hourly load data information involved into the predicted days of five targeted regions, figureFileSmall=HVHCfK3ku8SJtQQk/EiGgg==, figureFileBig=Lyj+YfQ7Et/dH6aKwhxuOg==, tableContent=null), ArticleFig(id=1218843913243246971, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, language=CN, label=图2, caption=目标区域的待预测日逐时负荷信息, figureFileSmall=HVHCfK3ku8SJtQQk/EiGgg==, figureFileBig=Lyj+YfQ7Et/dH6aKwhxuOg==, tableContent=null), ArticleFig(id=1218843913478128002, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, language=EN, label=Fig.3, caption=Schematic diagram of ALIF decomposition results in five urban areas, figureFileSmall=S2bRBLL0aENB9yDWltpTbQ==, figureFileBig=XSWZaZytkcTzW19S84vLSA==, tableContent=null), ArticleFig(id=1218843913599762828, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, language=CN, label=图3, caption=5个城区的ALIF一次分解结果示意图, figureFileSmall=S2bRBLL0aENB9yDWltpTbQ==, figureFileBig=XSWZaZytkcTzW19S84vLSA==, tableContent=null), ArticleFig(id=1218843913729786260, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, language=EN, label=Fig.4, caption=The iterative optimization process of VMD's two parameters based on NGO algorithm, figureFileSmall=0lUp4QkXqAF+4W6ErGW+Vg==, figureFileBig=kYm9kpS2CdxEZJX/Fsoawg==, tableContent=null), ArticleFig(id=1218843913838838171, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, language=CN, label=图4, caption=北方苍鹰优化VMD的迭代寻优过程示意图, figureFileSmall=0lUp4QkXqAF+4W6ErGW+Vg==, figureFileBig=kYm9kpS2CdxEZJX/Fsoawg==, tableContent=null), ArticleFig(id=1218843913956278697, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, language=EN, label=Fig.5, caption=Quadratic decomposition results of complex components in five urban areas based on NGO-VMD, figureFileSmall=FHAAHvkIDGVWRBlGKFif8g==, figureFileBig=u81TGnAtRDTJ6m/tXSJT6Q==, tableContent=null), ArticleFig(id=1218843914077913528, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, language=CN, label=图5, caption=基于NGO-VMD的五城区复杂分量二次分解结果, figureFileSmall=FHAAHvkIDGVWRBlGKFif8g==, figureFileBig=u81TGnAtRDTJ6m/tXSJT6Q==, tableContent=null), ArticleFig(id=1218843914212131265, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, language=EN, label=Fig.6, caption=The iterative optimization process of three hyperparameters of predicted model, figureFileSmall=2kxaf1DaAZWSQHLVgVDTSg==, figureFileBig=L44MWsWgMMiCb62Yu4/eiA==, tableContent=null), ArticleFig(id=1218843914329571790, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, language=CN, label=图6, caption=各区域预测模型的超参数迭代寻优过程, figureFileSmall=2kxaf1DaAZWSQHLVgVDTSg==, figureFileBig=L44MWsWgMMiCb62Yu4/eiA==, tableContent=null), ArticleFig(id=1218843914463789528, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, language=EN, label=Fig.7, caption=The comparison situation between predicted value and actual value of three models corresponding to five districts, figureFileSmall=y4QEKMhT2MsJHE4ue5fI2w==, figureFileBig=rJJZWAWw7nCdalXO5idfzA==, tableContent=null), ArticleFig(id=1218843914598007261, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, language=CN, label=图7, caption=研究区域的三套模型的预测值与实际值对比情况, figureFileSmall=y4QEKMhT2MsJHE4ue5fI2w==, figureFileBig=rJJZWAWw7nCdalXO5idfzA==, tableContent=null), ArticleFig(id=1218843914782556654, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, language=EN, label=Fig.8, caption=The comparison of evaluation criteria of prediction models under three types of data preprocessing scenarios, figureFileSmall=RQzEO44bTzpPhRCQ8eWM0A==, figureFileBig=T8W0e9loaprlTAlTeHR61Q==, tableContent=null), ArticleFig(id=1218843914891608563, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, language=CN, label=图8, caption=基于3种数据预处理场景的模型评价指标对比情况, figureFileSmall=RQzEO44bTzpPhRCQ8eWM0A==, figureFileBig=T8W0e9loaprlTAlTeHR61Q==, tableContent=null), ArticleFig(id=1218843915009049084, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, language=EN, label=Table 1, caption=

The identified key factors for five target areas

, figureFileSmall=null, figureFileBig=null, tableContent=
区域 温度 工作日 电价 上一时刻负荷 湿度 体感温度 前一天负荷
城区 0.915 -0.013 0.579 4 0.995 1 0.601 3 0.857 3 0.465 0
朝阳区 0.928 -0.012 0.640 3 0.995 2 0.626 8 0.889 7 0.502 3
丰台区 0.943 -0.005 0.601 0 0.995 4 0.590 5 0.886 1 0.487 1
海淀区 0.922 0.012 7 0.598 9 0.995 3 0.620 6 0.869 4 0.449 2
石景山区 0.922 0.014 1 0.615 9 0.995 3 0.615 9 0.880 5 0.514 9
), ArticleFig(id=1218843915151655433, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, language=CN, label=表1, caption=

5个目标区域的关键因素识别结果

, figureFileSmall=null, figureFileBig=null, tableContent=
区域 温度 工作日 电价 上一时刻负荷 湿度 体感温度 前一天负荷
城区 0.915 -0.013 0.579 4 0.995 1 0.601 3 0.857 3 0.465 0
朝阳区 0.928 -0.012 0.640 3 0.995 2 0.626 8 0.889 7 0.502 3
丰台区 0.943 -0.005 0.601 0 0.995 4 0.590 5 0.886 1 0.487 1
海淀区 0.922 0.012 7 0.598 9 0.995 3 0.620 6 0.869 4 0.449 2
石景山区 0.922 0.014 1 0.615 9 0.995 3 0.615 9 0.880 5 0.514 9
), ArticleFig(id=1218843915332010516, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, language=EN, label=Table 2, caption=

The calculated fuzzy entropy values for each IMF component in five regions

, figureFileSmall=null, figureFileBig=null, tableContent=
区域 IMF1 IMF2 IMF3 IMF4 IMF5 IMF6 IMF7
城区 1.0 1.1 0.9 0.6 0.6 0.4 0.6
朝阳区 4.9 1.4 1.2 0.9 0.7 0.6 0.8
丰台区 1.7 1.0 0.9 0.8 0.7 0.6 0.8
海淀区 1.9 1.3 0.9 0.8 0.5 0.4 0.6
石景山区 1.7 0.9 0.8 0.7 0.6 0.5 0.7
), ArticleFig(id=1218843915436868124, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, language=CN, label=表2, caption=

各区一次分解的各分量的模糊熵值计算结果

, figureFileSmall=null, figureFileBig=null, tableContent=
区域 IMF1 IMF2 IMF3 IMF4 IMF5 IMF6 IMF7
城区 1.0 1.1 0.9 0.6 0.6 0.4 0.6
朝阳区 4.9 1.4 1.2 0.9 0.7 0.6 0.8
丰台区 1.7 1.0 0.9 0.8 0.7 0.6 0.8
海淀区 1.9 1.3 0.9 0.8 0.5 0.4 0.6
石景山区 1.7 0.9 0.8 0.7 0.6 0.5 0.7
), ArticleFig(id=1218843915520754209, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, language=EN, label=Table 3, caption=

The calculated fuzzy entropy values for second decomposition components in each region

, figureFileSmall=null, figureFileBig=null, tableContent=
区域 IMF1 IMF2 IMF3 IMF4 IMF5 IMF6 IMF7 IMF8 IMF9 IMF10
城区 1.02 0.82 1.05 1.19 1.15 0.97 0.59 0.92 0.90 0.72
朝阳区 1.13 0.76 1.06 1.29 1.10 1.07 0.94 1.09 0.86 0.83
丰台区 0.37 0.90 1.12 0.65 0.50 0.37 0.31 0.30 0.32 0.22
海淀区 1.12 0.70 1.26 1.58 1.30 1.00 1.02 0.94
石景山区 0.38 0.46 0.39 0.40 0.27 0.27 0.24 0.24 0.39 0.45
), ArticleFig(id=1218843915625611819, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, language=CN, label=表3, caption=

各区的二次分解分量的模糊熵值计算结果

, figureFileSmall=null, figureFileBig=null, tableContent=
区域 IMF1 IMF2 IMF3 IMF4 IMF5 IMF6 IMF7 IMF8 IMF9 IMF10
城区 1.02 0.82 1.05 1.19 1.15 0.97 0.59 0.92 0.90 0.72
朝阳区 1.13 0.76 1.06 1.29 1.10 1.07 0.94 1.09 0.86 0.83
丰台区 0.37 0.90 1.12 0.65 0.50 0.37 0.31 0.30 0.32 0.22
海淀区 1.12 0.70 1.26 1.58 1.30 1.00 1.02 0.94
石景山区 0.38 0.46 0.39 0.40 0.27 0.27 0.24 0.24 0.39 0.45
), ArticleFig(id=1218843915743052340, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, language=EN, label=Table 4, caption=

The recombination results of final components of five districts

, figureFileSmall=null, figureFileBig=null, tableContent=
区域 新序列 原分量
城区 IMF1 二次分解(IMF1、IMF3、IMF4、IMF5)
IMF2 二次分解(IMF2、IMF6、IMF8、IMF9)
IMF3 一次分解(IMF5、IMF7)、二次分解(IMF10)
IMF4 一次分解(IMF4、IMF6)、二次分解(IMF7)
朝阳区 IMF1 一次分解(IMF2、IMF3)、二次分解(IMF1、IMF4)
IMF2 二次分解(IMF3、IMF5、IMF6、IMF8)
IMF3 一次分解(IMF4、IMF7)、二次分解(IMF7、IMF9)
IMF4 一次分解(IMF5、IMF6)、二次分解(IMF2、IMF10)
丰台区 IMF1 一次分解(IMF2、IMF3)、二次分解(IMF2、IMF3)
IMF2 一次分解(IMF4、IMF5、IMF7)
IMF3 一次分解(IMF6)、二次分解(IMF4、IMF5)
IMF4 二次分解(IMF1、IMF6、IMF7、IMF8、IMF9、IMF10)
海淀区 IMF1 二次分解(IMF3、IMF4、IMF5)
IMF2 二次分解(IMF1、IMF6、IMF7)
IMF3 一次分解(IMF3、IMF4)、二次分解(IMF8)
IMF4 一次分解(IMF5、IMF6、IMF7)、二次分解(IMF2)
石景山区 IMF1 一次分解(IMF2、IMF3、IMF4、IMF7)
IMF2 一次分解(IMF5、IMF6)、二次分解(IMF2、IMF10)
IMF3 二次分解(IMF1、IMF3、IMF4、IMF9)
IMF4 二次分解(IMF5、IMF6、IMF7、IMF8)
), ArticleFig(id=1218843915864687166, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, language=CN, label=表4, caption=

各区最终分量的重组结果

, figureFileSmall=null, figureFileBig=null, tableContent=
区域 新序列 原分量
城区 IMF1 二次分解(IMF1、IMF3、IMF4、IMF5)
IMF2 二次分解(IMF2、IMF6、IMF8、IMF9)
IMF3 一次分解(IMF5、IMF7)、二次分解(IMF10)
IMF4 一次分解(IMF4、IMF6)、二次分解(IMF7)
朝阳区 IMF1 一次分解(IMF2、IMF3)、二次分解(IMF1、IMF4)
IMF2 二次分解(IMF3、IMF5、IMF6、IMF8)
IMF3 一次分解(IMF4、IMF7)、二次分解(IMF7、IMF9)
IMF4 一次分解(IMF5、IMF6)、二次分解(IMF2、IMF10)
丰台区 IMF1 一次分解(IMF2、IMF3)、二次分解(IMF2、IMF3)
IMF2 一次分解(IMF4、IMF5、IMF7)
IMF3 一次分解(IMF6)、二次分解(IMF4、IMF5)
IMF4 二次分解(IMF1、IMF6、IMF7、IMF8、IMF9、IMF10)
海淀区 IMF1 二次分解(IMF3、IMF4、IMF5)
IMF2 二次分解(IMF1、IMF6、IMF7)
IMF3 一次分解(IMF3、IMF4)、二次分解(IMF8)
IMF4 一次分解(IMF5、IMF6、IMF7)、二次分解(IMF2)
石景山区 IMF1 一次分解(IMF2、IMF3、IMF4、IMF7)
IMF2 一次分解(IMF5、IMF6)、二次分解(IMF2、IMF10)
IMF3 二次分解(IMF1、IMF3、IMF4、IMF9)
IMF4 二次分解(IMF5、IMF6、IMF7、IMF8)
), ArticleFig(id=1218843916007293513, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, language=EN, label=Table 5, caption=

The optimal hyperparameters combination of prediction models corresponding to five targeted regions

, figureFileSmall=null, figureFileBig=null, tableContent=
区域 正则化系数 学习率 隐层节点
城区 0.008 3 0.000 896 49
朝阳区 0.000 000 000 1 0.000 1 249
丰台区 0.003 8 0.000 588 117
海淀区 0.002 4 0.001 565 114
石景山区 0.009 7 0.001 574 400
), ArticleFig(id=1218843916141511249, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, language=CN, label=表5, caption=

各区域预测模型的最优超参数组合

, figureFileSmall=null, figureFileBig=null, tableContent=
区域 正则化系数 学习率 隐层节点
城区 0.008 3 0.000 896 49
朝阳区 0.000 000 000 1 0.000 1 249
丰台区 0.003 8 0.000 588 117
海淀区 0.002 4 0.001 565 114
石景山区 0.009 7 0.001 574 400
), ArticleFig(id=1218843916212814427, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, language=EN, label=Table 6, caption=

The evaluation results of prediction accuracy of three types of models

, figureFileSmall=null, figureFileBig=null, tableContent=
区划 模型类型 评价指标
MAE RMSE R2
城区 LSTM 635.34 840.75 0.56
CNN-LSTM 467.88 727.12 0.66
NGO-CNN-LSTM 392.45 703.34 0.68
朝阳区 LSTM 1 057.32 1 268.49 0.61
CNN-LSTM 590.49 944.83 0.78
NGO-CNN-LSTM 477.95 819.44 0.84
丰台区 LSTM 278.78 435.95 0.76
CNN-LSTM 245.39 377.63 0.82
NGO-CNN-LSTM 209.86 347.84 0.85
海淀区 LSTM 505.88 745.92 0.73
CNN-LSTM 372.96 634.79 0.80
NGO-CNN-LSTM 338.71 624.23 0.81
石景山区 LSTM 87.49 120.48 0.64
CNN-LSTM 66.10 101.47 0.74
NGO-CNN-LSTM 55.26 87.98 0.81
), ArticleFig(id=1218843916309283430, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, language=CN, label=表6, caption=

三套模型的预测精度评估结果

, figureFileSmall=null, figureFileBig=null, tableContent=
区划 模型类型 评价指标
MAE RMSE R2
城区 LSTM 635.34 840.75 0.56
CNN-LSTM 467.88 727.12 0.66
NGO-CNN-LSTM 392.45 703.34 0.68
朝阳区 LSTM 1 057.32 1 268.49 0.61
CNN-LSTM 590.49 944.83 0.78
NGO-CNN-LSTM 477.95 819.44 0.84
丰台区 LSTM 278.78 435.95 0.76
CNN-LSTM 245.39 377.63 0.82
NGO-CNN-LSTM 209.86 347.84 0.85
海淀区 LSTM 505.88 745.92 0.73
CNN-LSTM 372.96 634.79 0.80
NGO-CNN-LSTM 338.71 624.23 0.81
石景山区 LSTM 87.49 120.48 0.64
CNN-LSTM 66.10 101.47 0.74
NGO-CNN-LSTM 55.26 87.98 0.81
), ArticleFig(id=1218843916443501169, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, language=EN, label=Table 7, caption=

The prediction accuracy comparison of prediction models based on three types of data preprocessing methods

, figureFileSmall=null, figureFileBig=null, tableContent=
区划 模型类型 评价指标
无预处理 一次分解 二次分解
MAE RMSE R2 MAE RMSE R2 MAE RMSE R2
城区 模型1 890.56 1 053.12 0.45 863.32 1 047.38 0.44 635.34 840.75 0.56
模型2 578.65 793.03 0.61 565.19 798.79 0.61 467.88 727.12 0.66
模型3 441.59 722.00 0.67 427.58 736.24 0.65 392.45 703.34 0.68
朝阳区 模型1 1 289.34 1 548.44 0.42 1 102.41 1 342.24 0.56 1 057.32 1 268.49 0.61
模型2 921.24 1 174.87 0.66 783.21 1 080.77 0.72 590.49 944.83 0.78
模型3 581.26 930.97 0.79 518.49 938.45 0.79 477.95 819.44 0.84
丰台区 模型1 319.96 463.23 0.73 295.31 441.75 0.76 435.95 0.76 435.95
模型2 274.66 435.95 0.76 260.30 379.32 0.82 377.63 0.82 377.63
模型3 244.79 394.35 0.81 232.57 389.55 0.81 347.84 0.85 347.84
海淀区 模型1 717.69 998.69 0.55 506.19 760.04 0.72 505.88 745.92 0.73
模型2 445.93 683.02 0.78 397.05 643.15 0.80 372.96 634.79 0.80
模型3 371.95 637.31 0.80 350.06 618.64 0.81 338.71 624.23 0.81
石景山区 模型1 98.48 129.35 0.59 89.09 124.15 0.62 87.49 120.48 0.64
模型2 86.19 113.38 0.69 78.78 112.80 0.68 66.10 101.47 0.74
模型3 65.95 101.71 0.74 65.27 99.35 0.75 55.26 87.98 0.81
), ArticleFig(id=1218843916569330300, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149776904888217655, language=CN, label=表7, caption=

基于三类数据预处理方法的预测模型精度对比

, figureFileSmall=null, figureFileBig=null, tableContent=
区划 模型类型 评价指标
无预处理 一次分解 二次分解
MAE RMSE R2 MAE RMSE R2 MAE RMSE R2
城区 模型1 890.56 1 053.12 0.45 863.32 1 047.38 0.44 635.34 840.75 0.56
模型2 578.65 793.03 0.61 565.19 798.79 0.61 467.88 727.12 0.66
模型3 441.59 722.00 0.67 427.58 736.24 0.65 392.45 703.34 0.68
朝阳区 模型1 1 289.34 1 548.44 0.42 1 102.41 1 342.24 0.56 1 057.32 1 268.49 0.61
模型2 921.24 1 174.87 0.66 783.21 1 080.77 0.72 590.49 944.83 0.78
模型3 581.26 930.97 0.79 518.49 938.45 0.79 477.95 819.44 0.84
丰台区 模型1 319.96 463.23 0.73 295.31 441.75 0.76 435.95 0.76 435.95
模型2 274.66 435.95 0.76 260.30 379.32 0.82 377.63 0.82 377.63
模型3 244.79 394.35 0.81 232.57 389.55 0.81 347.84 0.85 347.84
海淀区 模型1 717.69 998.69 0.55 506.19 760.04 0.72 505.88 745.92 0.73
模型2 445.93 683.02 0.78 397.05 643.15 0.80 372.96 634.79 0.80
模型3 371.95 637.31 0.80 350.06 618.64 0.81 338.71 624.23 0.81
石景山区 模型1 98.48 129.35 0.59 89.09 124.15 0.62 87.49 120.48 0.64
模型2 86.19 113.38 0.69 78.78 112.80 0.68 66.10 101.47 0.74
模型3 65.95 101.71 0.74 65.27 99.35 0.75 55.26 87.98 0.81
)], 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.1671-1815.2404503, detailUrlEn=https://castjournals.cast.org.cn/joweb/kxjsygc/EN/10.12404/j.issn.1671-1815.2404503, pdfUrlCn=https://castjournals.cast.org.cn/joweb/kxjsygc/CN/PDF/10.12404/j.issn.1671-1815.2404503, pdfUrlEn=https://castjournals.cast.org.cn/joweb/kxjsygc/EN/PDF/10.12404/j.issn.1671-1815.2404503, aliStartDate=null, aliEndDate=null, collectionFlag=false, citedCount=null, citedUrl=null, reference=null)
收藏切换
基于ALIF-VMD二次分解的NGO-CNN-LSTM电力负荷短期组合预测模型
收藏切换
PDF下载
张琳 1 , 高胜强 1, * , 宋煜 2 , 卜帅羽 1 , 余伟 1
科学技术与工程 | 论文·电工技术 2025,25(11): 4583-4597
收起
收藏切换
科学技术与工程 | 论文·电工技术 2025, 25(11): 4583-4597
基于ALIF-VMD二次分解的NGO-CNN-LSTM电力负荷短期组合预测模型
全屏
张琳1 , 高胜强1, * , 宋煜2, 卜帅羽1, 余伟1
作者信息
  • 1 国网北京市电力公司, 北京 100031
  • 2 华北电力大学环境科学与工程学院, 北京 102206
  • 张琳(1977—),男,汉族,北京人,硕士,高级工程师。研究方向:电力负荷预测、电力系统规划。E-mail:

通讯作者:

* 高胜强(1980—),男,汉族,北京人,硕士,高级工程师。研究方向:电力负荷预测、电力系统规划。E-mail:
NGO-CNN-LSTM Power Load Short-term Combination Forecasting Model Based on ALIF-VMD Quadratic Decomposition
Lin ZHANG1 , Sheng-qiang GAO1, * , Yu SONG2, Shuai-yu BU1, Wei YU1
Affiliations
  • 1 State Grid Beijing Electric Power Company, Beijing 100031, China
  • 2 The College of Environmental Science and Engineering, North China Electric Power University, Beijing 102206, China
出版时间: 2025-04-18 doi: 10.12404/j.issn.1671-1815.2404503
文章导航
收藏切换

针对电力负荷预测过程中普遍存在的负荷波动变化趋势明显、随机性强,以及预测模型的参数取值不合理导致的精度偏低问题,提出了一种基于ALIF-VMD(adaptive local iterative filtering - variational mode decomposition)二次分解和北方苍鹰优化算法(northern goshawk optimization,NGO)优化CNN-LSTM(convolutional neural networks - long short-term memory)的电力负荷组合预测模型,在使用交叉映射收敛方法(convergent cross-mapping,CCM)准确识别电力负荷的关键影响因素的基础上,创新性地联合使用ALIF、基于NGO的VMD和模糊熵(fuzzy entropy,FE)对原始负荷序列进行组合分解和必要的重组;针对分解和重组后生成的模态分量,结合NGO确定的CNN-LSTM模型最优超参数组合,建立预测精度高、训练时间短、收敛速度快的NGO-CNN-LSTM日前电力负荷组合预测模型。与其他基准模型的对比结果表明,该模型具有更好的适应性和预测精度,可为电力系统的安全、可靠、经济运行提供重要的技术支撑。

负荷预测  /  序列分解与重组  /  北方苍鹰算法  /  卷积神经网络-长短期记忆神经网络模型

Aiming at obvious load fluctuation trend, strong randomness and low accuracy caused by unreasonable parameter values of the prediction model involved into the power load forecasting process, a combined prediction model composing of ALIF (adaptive local iterative filtering), VMD (variational mode decomposition), NGO (northern goshawk optimization) and CNN-LSTM (convolutional neural networks - long short-term memory) was established. Firstly, CCM (convergent cross-mapping) method was used to identify the key factors affecting the power load. Secondly, an innovative combination of ALIF, NGO-based VMD and FE (fuzzy entropy) was employed for combinatorial decomposition and necessary recombination of original load sequence. Next, based on the modal components generated after decomposition and recombination, combined with optimal hyperparameter combination of CNN-LSTM determined by NGO method, an NGO-CNN-LSTM day-ahead power load combination prediction model with the high prediction accuracy, short training time and fast convergence speed was formulated. Compared with other benchmark models, the obtained results demonstrated that the proposed model has the better adaptability and prediction accuracy, and can provide important technical support for the safe, reliable and economical operation of power system.

load forecasting  /  sequence decomposition and recombination  /  northern goshawk optimization  /  convolutional neural network-long short-term memory neural network model
张琳, 高胜强, 宋煜, 卜帅羽, 余伟. 基于ALIF-VMD二次分解的NGO-CNN-LSTM电力负荷短期组合预测模型. 科学技术与工程, 2025 , 25 (11) : 4583 -4597 . DOI: 10.12404/j.issn.1671-1815.2404503
Lin ZHANG, Sheng-qiang GAO, Yu SONG, Shuai-yu BU, Wei YU. NGO-CNN-LSTM Power Load Short-term Combination Forecasting Model Based on ALIF-VMD Quadratic Decomposition[J]. Science Technology and Engineering, 2025 , 25 (11) : 4583 -4597 . DOI: 10.12404/j.issn.1671-1815.2404503
近年来,随着经济的快速发展和人民生活水平的逐步提高,中国的电力需求持续快速增长,国家统计局发布的国民经济和社会发展统计公报显示,2023年的全社会用电量约为92 241亿千瓦时,同比增长6.7%,电力消费呈现多元化发展趋势。电力负荷预测是电力系统规划的重要组成部分,电力负荷的精准预测,有助于提高电能的分配效率和利用率,对于确保电力系统的安全运行,实现电力公司的能源供应规划优化设计和降本增效具有重要意义[1]
目前,常用的短期电力负荷预测方法主要包括统计学习和机器学习两大类。其中,统计学习方法的预测过程容易受制于随机因素的影响,导致其在面对复杂的研究对象时,预测性能较差且缺乏稳定性。相较于统计学习方法,机器学习算法存在多变量的交互作用关系考虑不够全面和基于大数据集的预测精度波动性较大的问题[2-3]。近年来,深度学习在负荷预测领域受到国内外诸多学者的广泛关注。例如,葛亚明等[4]利用长短期记忆神经网络(long short-term memory network,LSTM)建立了计及多能源相互作用机制的电力负荷预测模型。仿真结果表明,在全面考虑多种影响因子的基础上,LSTM能够实现园区电力负荷准确预测的同时,模型的训练时间和计算量也会随之增加。作为LSTM的变体,门控循环单元网络(gated recurrent unit,GRU)的网络结构更加简单且计算时间缩短,因此被广泛用于解决电力负荷预测问题。王增平等[5]使用GRU建立具有时序性的历史负荷序列预测模型。由于模型中缺乏卷积计算过程,导致模型的特征提取能力不足。由此可见,单一的深度学习方法可以在一定程度上实现电力负荷的准确预测,但其仍面临计算复杂度高和效率低下等问题。卷积神经网络(convolutional neural networks,CNN)因其具有强大的特征提取能力而与其他预测模型联合使用,可以妥善解决电力负荷预测的精度偏低问题,组合预测模型应运而生。刘伟等[6]在使用离散余弦将时域信息变换为频域信息的基础上,结合CNN的特征提取结果,构建GRU负荷预测模型。结果表明,CNN-GRU能够显著提高预测精度。张文栋等[7]在使用科普拉理论对冷、热、电3种负荷进行相关性分析的基础上,建立了CNN-LSTM预测模型。在某工业园区的实际应用结果很好地反映了该组合预测模型的良好预测性能。Huy等[8]建立了基于CNN-LSTM的短期负荷预测模型,探明了负荷数据的时空关联关系,相较于独立的CNN和LSTM模型,组合模型的预测精度显著提高。尽管组合预测模型能够实现各类模型的优势互补,但其关键参数的取值往往依赖人为主观设定,耗费大量人工调参时间的同时,降低模型的收敛速度和影响模型的预测表现。因此,运用智能优化算法确定模型的关键参数的最优组合成为解决这一问题的重要途径。白宇等[9]在利用CNN-BiLSTM(bi-directional long short-term memory)对原始的负荷数据信息进行时间和空间特征提取的基础上,使用改进的金豺优化算法(improved golden jackal optimization,IGJO)对CNN-BiLSTM的超参数进行寻优并建立了IGJO-CNN-BiLSTM组合预测模型。结果表明,基于参数优化的预测模型的各项精度指标均得到显著提升。Kiruthiga等[10]建立了基于BiLSTM-Dropout的负荷预测模型,并使用莱维飞行和粒子群算法确定了模型的最优隐藏单元个数和学习率等超参数。相较于两类基准模型,基于优化后的组合模型的平均绝对误差和均方根误差分别降低了5.62%、71.02%和2.24%、50.48%。
受制于天气条件、日期性质、分时电价、需求响应策略等多重因素的影响,原始负荷数据序列普遍呈现出随机性、波动性和无序性,不利于直接建立预测模型,因此,许多学者将信号分解技术应用于电力负荷预测领域。例如,周磊等[11]在使用多重滑动平均(moving average, MA)方法将原始负荷数据分解为多个序列的基础上,融合自注意力机制(self-attention)与CNN-LSTM模型建立了MA-CNN-LSTM-self attention电力负荷预测模型,实现了电力负荷的精准预测。张淑娴等[12]使用自适应噪声完备经验模态分解(complete ensemble empirical mode decomposition with adaptive noise,CEEMDAN)和变分模态分解(variational mode decomposition,VMD)相结合的二次模态分解技术将原始电力负荷分解为多个子序列,运用LSTM模型生成的预测结果表明,信号分解技术能够降低原始负荷的波动性,提高预测精度。Wang等[13]使用基于秃鹰搜索(bald eagle search,BES)优化算法的VMD将电力负荷分解为不同分量,在考虑误差修正的基础上,针对各分量分别建立了CNN-BiLSTM负荷预测模型,有效提高了负荷预测精度。尽管包括VMD、CEEMDAN、EMD等在内的常用信号分解技术的引入可以有效降低电力负荷原始数据的波动性,但是,在分解过程中容易出现模态混叠、处理突发信号能力弱、自适应差等问题。相反,自适应处理的迭代滤波方法(adaptive local iterative filtering,ALIF)可以对非线性波动信号进行高效处理,能够很好地抑制模态混叠且具有更高的分解效率,但是,该方法目前主要用于轴承故障诊断领域,在负荷预测方面应用有限[14]。此外,对原始序列进行分解以后生成的多个子序列的叠加预测,大幅增加模型计算量的同时,容易导致各子序列预测误差的累积,进而影响预测精度。因此,很多研究通过引入样本熵、模糊熵等方法对分解后的序列进行重构。邓博文等[15]在对原始负荷数据进行二次模态分解的基础上,使用样本熵和K均值将分解的子序列重构为3个序列,并输入CNN-LSTM模型进行预测。结果表明,该方法具有更高的精度和拟合度。肖威等[16]在使用VMD将负荷数据分解为一系列的低频和高频模态分量的基础上,分别建立了LSTM和GRU预测模型。仿真结果表明,该方法有效降低数据波动性,具有更高的预测精度。综上所述,联合使用信号分解、序列重构、超参数优化和智能预测算法构建的组合预测模型能够充分发挥各类模型的优势、降低原始负荷序列的非平稳性的同时,显著提高预测精度。因此,本文中在使用交叉映射收敛(convergent cross-mapping,CCM)方法准确识别电力负荷的关键影响因素的基础上,首先,创新性地联合使用ALIF和模糊熵(fuzzy entropy,FE)对原始负荷序列进行一次分解和必要的重组;其次,运用基于北方苍鹰优化算法(northern goshawk optimization,NGO)的VMD对熵值较高的分量进行二次分解,并利用FE对一次分解的剩余分量和二次分解得到的所有分量进行重组;再次,针对分解和重组后生成的模态分量,运用NGO确定CNN-LSTM模型的最优超参数组合;最后,建立了预测精度高、训练时间短、收敛速度快的NGO-CNN-LSTM日前电力负荷短期组合预测模型,实现了北京市8个中心城区的电力负荷短期精准预测。
图1展示了融合影响因素识别、序列分解和重构、超参数优化和人工智能预测的日前电力负荷组合预测模型的构建过程。首先,使用CCM方法选取温度、电价、湿度、上一时刻负荷等作为电力负荷的关键影响因素;其次,运用ALIF对电力负荷进行一次分解,并计算各个分量的FE值,对FE值偏高且较为接近的分量进行重构;再次,使用NGO-VMD对重构后的高熵值分量进行二次分解,根据FE值将一次分解的剩余分量和二次分解得到的所有分量进行二次重构,生成最终的模态分量集;最后,使用NGO对CNN-LSTM的关键参数(包括学习率、隐含层节点和正则化系数)进行优化,建立各个分量的NGO-CNN-LSTM组合预测模型,叠加各分量的预测值来实现电力负荷的日前短期精准预测。本文中选择北京的8个中心城区(即西城区、东城区、宣武区和崇文区组成的四城区、朝阳区、丰台区、海淀区和石景山区)作为研究目标,分别建立了基于序列分解和重构的LSTM、CNN-LSTM和NGO-CNN-LSTM电力负荷预测模型,利用平均绝对误差(mean absolute error,MAE)、均方根误差(root mean squared error,RMSE)和拟合优度值R2作为评价指标,对比各个模型的预测表现看出,基于ALIF-VMD二次分解的NGO-CNN-LSTM组合预测模型具有较好的预测性能,可为电力系统的安全、可靠、经济运行提供重要的技术支撑。
收敛交叉映射算法是Sugihara[17]在2012年提出的一种比较新颖的基于动态系统理论的时间序列因果关系识别方法。该方法具有不受数据分布限制、对噪声和混淆因素具有强鲁棒性以及能够准确识别非线性关系等优势,适于表征复杂系统内部组成要素的因果关系。CCM利用重构相空间具有微分同胚的特性对相空间进行交互预测,寻找在一个变量的相位空间中嵌入的另一个变量的映射,可以准确辨识变量之间的非直接、非对称影响关系,合理判定电力负荷的关键影响因素。具体计算步骤如下所示。
(1)数据预处理。在确定长度为L的两个时间序列X=[x1,x2,…,xt]和Y=[y1,y2,…,yt]的基础上,分别对其进行归一化处理,具体形式为
x ~ i= x i m a x ( x i ), ∀i
y ~ i= y i m a x ( y i ), ∀i
式中: x ~ i y ~ i分别为序列XY的变量xiyi经过归一化后的数值;max(xi)和max(yi)分别为序列XY的所有变量xiyi的最大值。
(2)重构流形。假设重构流形的维度为E,重构过程的采样间隔为τ,重构得到XY的吸引子流形 M x My,即
X(t)=< x ~(t), x ~(t-τ),…, x ~[t-(E-1)τ]>
Y(t)=< y ~(t), y ~(t-τ),…, y ~[t-(E-1)τ]>
式中:MxMy为微分同胚,因此,创建Y(t)的交叉映射估计 Y ( t )Mx,并利用影子流形Mx估计时间序列Y(t)。
(3)预测流形。根据Takens引理,定义流形交叉映射后的 Y ( t )Mx,具体表达式为
Y ( t )Mx= i = 1 E + 1wiY(ti), i=1,2,…,E+1
wi= m i i = 1 E + 1 m j, j=1,2,…,E+1
mi=exp{-d[ x ~(t), x ~(ti)]/d[ x ~(t), x ~(t1)]}
式中:wi为权重;d[ x ~(t), x ~(ti)]为 x ~(t)与 x ~(ti)之间的欧几里得距离。
(4)计算CCM相关系数。计算 Y ( t )Mxy(t)之间的CCM相关系数,具体表达式为
ρ= i = 1 L [ Y ( i ) - Y ¯ ] [ Y ( i ) M X - Y | M X ¯ ] i = 1 L [ Y ( i ) - Y ¯ ] 2 i = 1 L [ Y ( i ) M X - Y | M X ¯ ] 2
式(8)中:ρ为基于皮尔逊相关系数的因果关系强度,反映了XY两个时间序列之间的相关程度;L为样本长度,随着L的增加, Y ( t )Mx逐渐收敛于大于0小于1的确定数值,表明YX存在因果关系。当ρ>0.6时[18],可认为变量间具有显著的因果关系,即YX具有显著因果相关性。
ALIF分解方法在运用迭代滤波方法(iterative filtering,IF)完成原始序列的初始分解的基础上,引入福克-普朗克方程构造可以局部计算滤波器长度的滤波函数,利用自适应数据驱动的滤波器提升初始分解分量的平滑度,进一步强化序列的分解效果。该方法具备高效处理非线性信号、妥善解决信号模态混叠和端点效应等优势,具体步骤如下[19]:
(1)基于IF算法的滑动算子和波动算子计算。
IF算法包含内、外两个循环。一方面,通过内循环获得滑动算子和波动算子。对于待处理信号x(t)和滤波函数y(t),可根据其卷积结果求得滑动算子Γ[x(t)],计算公式[20]
Γ[x(t)]= - l l x(t+τ)y(t)dτ
l=2 β T n
式中:t为时间;l为滤波区间;τ为时间延迟;y(t)为滤波区间为l的滤波函数;β为相关系数,其范围为[1.5,2.0];T为待分解的序列长度;n为序列极值点的数量。根据式(9)和式(10),可以计算得到波动算子,具体表达式为
k[x(t)]=x(t)-Γ[x(t)]
另一方面,当波动算子满足分量计算条件时,进入外循环计算剩余分量r(t);否则,重复上述步骤直至达到分量计算条件,具体形式为
r(t)=x(t)-c(t)
式(12)中:c(t)为提取的本征模态函数(intrinsic mode function,IMF)分量。当r(t)具有明显趋势特征时,迭代滤波计算过程结束;否则,r(t)作为新一轮分解信号继续循环。
(2)基于自适应滤波函数的初始分量平滑度提升。
考虑到IF算法预设的固定滤波函数难以保证初始分解得到的分量具备一定的平滑度,利用福克-普朗克方程构造具备自适应调整功能的滤波函数来提升分量的平滑程度,该方程的具体形式为
pt=-ε1 [ g 1 ( x ) p ] x2 [ g 2 2 ( x ) p ] x x
式(13)中:ε1ε2代表稳态系数,取值范围为(0, 1);函数g1(x)和g2(x)为在x1<0<x2条件下满足:g1(x)=g2(x)=0且∀x∈(x1,x2)存在g1(x)>0,g1(x)<0<g2(x)的可导函数。
由于 [ g 2 2 ( x ) p ] x x会使方程的解从区间向两端移动;相反,-ε1 [ g 1 ( x ) p ] x会使方程解从两端向中心移动,二者达到平衡时可得
-ε1 [ g 1 ( x ) p ] x2 [ g 2 2 ( x ) p ] x x=0
如果上述方程存在非零解p(x),且满足条件式(15)时,可以确定p(x)为所求的滤波函数;当x∈(x1,x2)区间变换时,既可以得到不同的滤波函数,实现滤波函数的自适应计算。
p ( x ) = 0 , x ( x 1 , x 2 ) p ( x ) 0 , x ( x 1 , x 2 )
NGO算法是一种模拟北方苍鹰捕猎行为的元启发式算法,相较于其他优化算法,可以很好地兼顾全局及局部寻优,准确度和稳定性显著提升[21]。该算法以苍鹰种群作为搜索者,针对目标猎物制定相应的捕获方案,主要包括目标猎物识别和捕获两个阶段,具体过程如下:
(1)通过全局随机搜索在识别阶段确定最佳猎物,并对猎物进行攻击,其数学表达式[22]
Pi=Xk
x i , j n e w , P 1= x i , j + r ( p i , j - I x i , j ) , F P i < F i x i , j + r ( x i , j - p i , j ) , F P i F i
Xi= x i , j n e w , P 1 , F i n e w , P 1 < F i X i , F i n e w , P 1 F i
式中:Pi为第i只北方苍鹰的目标猎物的位置;Xk为第k只北方苍鹰向量构成的矩阵; x i , j n e w , P 1为新状态下第一阶段的第i个解决方案的第j个维度; F i n e w , P 1为第一阶段的目标函数值;r代表[0,1]之间的随机数;I取值为1或2。
(2)在捕获阶段,北方苍鹰与猎物进行追逐,最终成功捕获猎物,该过程即为局部寻优,具体方程为
x i , j n e w , P 2=xi,j+R(2r-1)xi,j
R=0.02 1 - t T
Xi= x i , j n e w , P 2 , F i , j n e w , P 2 < F i X i , F i , j n e w , P 2 F i
式中:R代表种群的攻击范围;t为迭代次数;T为最大迭代次数; F i n e w , P 2为第二阶段的目标函数。
以往的研究结果表明,采用VMD方法对原始序列进行分解时,分解的模态数量k的选择和惩罚因子α的设置对于分解效果产生很大影响[23]。其中,k值过大容易引发模态重复或产生额外的噪声,增加计算成本;k值过小则导致分解不够彻底,影响后续的预测精度。至于惩罚因子α,直接关系到模型的复杂度和噪声的容忍度,影响VMD的分解性能。为了降低VMD的两个关键参数人为设置过程中的主观性和随机性对分解结果造成的不利影响,本文中创新性地使用NGO算法结合包络熵对模态分解数k和惩罚因子α进行全局搜索,确定其最优组合,主要步骤如下:
(1)首先,初始化NGO参数和北方苍鹰的位置,选定包络熵EP作为适应度函数,具体公式为
EP=- i = 1 npilgpi
pi= h i i = 1 n h i
式中:hi为信号xi(i=1,2,…,n)经过希尔伯特解调后的包络熵;该值偏大,意味着序列包含噪声较多,分解效果不好;反之,包含较少的噪声,分解结果可以接受;pihi归一化后的序列。
(2)其次,计算每只北方苍鹰初始位置的适应度,选取最优解作为kα的初始值;然后,随机挑选猎物,通过式(16)~式(21),进行全局和局部寻优,更新北方苍鹰的位置。
(3)最后,迭代计算适应度值,更新苍鹰位置并保存最优解,持续循环直至包络熵值趋于不变或者达到最大迭代次数,输出最优解。
作为样本熵(sample entropy,SE)的改进方法,模糊熵算法在保留SE可以量化评估时间序列复杂度的特性的基础上,通过引入模糊隶属度函数作为熵的阈值依据,很好地改善了SE对于参数取值过于敏感的问题,具有更好的鲁棒性[24]。该方法的计算流程如下:
(1)完成原始序列的相空间重构。
针对给定的长度为N的序列{xi(i=1,2,…,n)},通过嵌入维数m对其进行相空间重构,得到时间序列S
Sm[i]={u(i),u(i+1),…,u(i+m-1)}-u0(i)
式(24)中:Sm[i]为重构后的新序列,i=1,2,…,N-m+1;u0(i)为m个连续变量u(i)的均值,其表达式为
u0(i)= 1 m j = 0 m - 1u(i+j)
(2)在定义两个向量s(i)与s(j)的距离为两个向量元素差值绝对值最大值的基础上,引入模糊隶属度函数,计算两个向量的相似程度相关表达式为
d i j m=d[sm(i),sm(j)]= m a x p = 1,2 , , m[ u ( i + p - 1 ) - u 0 ( i )- u ( j + p - 1 ) - u 0 ( j )]
A i j m= 1 , d i j m = 0 e x p - l n d i j m r 2 , d i j m > 0
式中:1≤i,jN-m+1,且ij;r为相似容限参数,定义为原一维时间序列标准差的R倍,即r = ,δ为原一维时间序列的标准差。
(3)首先,定义平均相似度函数 C i m(r),即
C i m(r)=(N-m)-1 j = 1 , j i N - m + 1 A i j m
然后,计算得到m维度下的关系维度φm(r),公式为
φm(r)=(N-m)-1 i = 1 N - m + 1 C i m(r)
最后,嵌入维数增加1,对m+1维向量重复上述步骤,得到m+1维度下的关系维度φm+1(r),即
φm+1(r)= 1 N - m i = 1 N - m C i m + 1(r)
式中:N为序列长度。
(4)计算得到FE值,并根据熵值的接近程度,完成分量的分类和重组。
FE(m,r,N)=lnφm(r)-lnφm+1(r)
CNN-LSTM混合模型由输入层、卷积层、池化层、LSTM层和输出层组成,主要参数包括卷积核的大小与个数、隐含层节点数、正则化系数和学习率等。该模型充分融合了CNN和LSTM两个模型的优势,有效缩短模型训练时间的同时,保证了训练梯度的稳定和利用效率,具有较强的误差修正能力,在预测领域得到了广泛应用[25]。具体操作流程如下:
(1)基于CNN的特征提取。CNN通过卷积运算提取模型的输入特征,并将其输入至LSTM模块。表达式为
x j l=f(xt-1 k j l), j∈J
p j l + 1=max[ x j l(t)], j∈J
式中:xt-1为第t-1个输入; k j l代表滤波,其中,l为卷积层数,J为滤波器的类型;f表示激活函数; x j l(t)为上一卷积层的特征映射; p j l + 1为输出的特征维度;max()指的是最大池化操作。
(2)基于LSTM的数据挖掘分析和预测结果生成。LSTM利用门控机制对CNN层提取的输入特征进行深入分析和耦合,并在输出层生成最终预测结果。计算公式为
ft=σ(Wf[ht-1,xt]+bf)
it=σ(Wi[ht-1,xt]+bI)
Ot=σ(Wo[ht-1,xt]+bo)
C ~ t=tanh(Wc[ht-1,xt]+bc)
Ct=ftCt-1+it C ~ t
ht=OttanhCt
式中:ft为遗忘门;it为输入门;Ot为输出门; C ~ t表征候选细胞状态;xt代表t时刻的输入;ht-1Ct-1分别为上一时刻的隐层输出和细胞状态;htCt分别为当前的隐层输出和细胞状态;WfWiWobfbibo分别代表遗忘门、输入门、输出门的权重及偏置;σ为sigmoid函数。
考虑到CNN-LSTM的模型参数设置多以主观经验判断为主,严重影响组合模型网络结构的合理性[26]。为了充分发挥CNN-LSTM的预测性能,本文中使用前述1.3.2节介绍的NGO算法对CNN-LSTM模型的学习率、隐含层节点和正则化系数3个重要参数进行迭代寻优,确定它们的最优组合,并据此构建预测模型。具体步骤如下:
(1)将上述3个关键参数编码为北方苍鹰个体的初始位置,在算法初始化过程中生成一系列的随机参数组合。
(2)在初始化条件下,以预测结果的误差最小化为目标函数进行北方苍鹰群个体位置的适应度计算,根据计算结果进行北方苍鹰位置的迭代和更新。
(3)重复步骤(2)至最大迭代次数,输出的北方苍鹰个体位置即为参数的最优解。
(4)在确定最优的参数组合的基础上,结合ALIF、NGO-VMD和FE分解和重组后生成的IMF分量,分别训练得到每个IMF分量对应的NGO-CNN-LSTM预测模型,对所有的IMF分量预测结果进行求和,即可得到最终的预测结果。
为了考察本文所提组合预测方法的有效性,选定北京市的8个中心城区(即西城区、东城区、宣武区和崇文区组成的四城区、朝阳区、丰台区、海淀区和石景山区)作为研究目标,开展合计5个城区的日前短期负荷预测。在收集和整理5个城区的2019年夏季6—8月的全天电力负荷、气象条件和实时电价等数据信息(数据采集间隔为1 h),形成历史样本集的基础上,随机选取一周作为待预测日,结果如下:四城区(以下简称为城区)、朝阳区、丰台区、海淀区和石景山区的待预测日分别为7月22—28日、7月8—14日、6月24—30日、8月4—10日和8月25—31日。图2所示为5个城区的待预测日的历史负荷信息。其余数据作为训练样本,建立短期的电力负荷组合预测模型,并通过与其他基准模型进行对比,检验其实用性和可行性。
考虑到CCM方法在关键因素的选取方面具有准确识别变量之间的非线性关系、有效处理时滞效应和面向异常值的强鲁棒性等优点,本文中采用CCM方法识别和确定了5个目标区域的关键影响因素,具体结果如表1所示。
表1所示,温度、上一时刻负荷和体感温度对原始负荷施加了很大的影响,它们的关系呈现为强相关性;电价和湿度为较强相关性;前一天负荷呈现中相关性;工作日几乎跟负荷没有关联关系,为弱相关性。因此,各区均选取温度、上一时刻负荷、体感温度、电价和湿度作为关键因素;其中,朝阳区和石景山区的前一天负荷的相关性系数大于0.5,也考虑作为后续构建的预测模型的输入变量。
为了有效克服原始电力负荷序列的非平稳性和波动性给负荷预测模型的精度带来的不利影响,联合使用ALIF、NGO优化的VMD和模糊熵对原始序列进行分解和必要的重构,生成变化趋势平稳、更具规律性,易于预测的IMF分量。图3所示为利用ALIF对5个区域的原始电力负荷数据分别进行一次分解得到的分量的情况。
为了便于对上述ALIF方法分解得到的初始IMF分量进行必要的二次分解和重组,使用模糊熵对一次分解得到的7个分量进行复杂程度评估,表2展示了各区一次分解生成的各分量的模糊熵计算结果。
表2可以看出,各区的原始负荷序列经过一次分解后生成的各分量的模糊熵值存在较大差异,但整体呈现递减趋势,即分量的复杂程度逐渐减小。另外,各区分量的复杂度偏高即模糊熵值偏大的分量有所不同,朝阳区、丰台区和石景山区的IMF1分量、城区的IMF1-3分量和海淀区的IMF1-2分量模糊熵值明显高于其他分量,表明其仍具有明显的波动性和非平稳性,给后续的高精度分量预测模型构建带来不利影响,有必要对其进行二次分解。
如前所述,由于各区一次分解后得到的个别分量熵值偏高,仍具有较高的复杂度,考虑到VMD的分解效果主要受到模态数K值和惩罚因子α的影响,K值过大可能产生模态混叠现象或者额外的噪声;相反,K值过小容易造成分解不彻底。至于惩罚因子α,由于其直接影响模态分量的带宽,若该值设置过小,影响特征提取的同时,导致模态混叠;相反,虽然可以有效避免模态混叠,但容易造成局部信息丢失。因此,在将5个区域的熵值偏高即复杂程度较高的分量序列分别进行重组形成5个单一分量的基础上,利用NGO确定VMD的两个关键参数的最优组合,对5个区域的分量序列进行二次分解。
NGO的初始参数设置情况如下:初始种群数量为5;最大迭代次数为20次;模态数K值和惩罚因子α的上下界分别为[4,10]和[800,2 000]。图4所示为5个区域的基于NGO的VMD超参数寻优过程。各区的高熵值分量二次分解结果如图5所示。表3展示了二次分解得到的分量的熵值计算结果。
表3所示,相较于ALIF一次分解得到的分量的模糊熵值计算结果,经过二次分解以后得到的分量(海淀区的二次分解最优分量数为8个,其余4个区域的最优分量数为10个)的熵值明显降低,表明NGO-VMD具有进一步降低分量的非平稳性的优势。为了有效降低分量预测模型的计算量和提高模型的预测精度,选定模糊熵值作为评判指标,将一次分解和二次分解的分量进行重构(重构结果详见表4),最终构建5个目标区域的各个重构分量的NGO-CNN-LSTM预测模型。
为了更好地验证NGO-CNN-LSTM组合预测模型的性能优势,在联合使用ALIF、NGO-VMD和FE对原始负荷进行分解和必要的重组的基础上,分别构建了LSTM、CNN-LSTM、NGO-CNN-LSTM三套电力负荷日前短期预测模型,分别预测得到了5个城区的夏季待预测日的发电量。表5提供了运用NGO算法确定的CNN-LSTM的最优超参数组合。NGO的初始参数设置情况如下:初始种群数量为6;最大迭代次数为30次;预先设定的正则化系数、学习率和隐藏节点数的搜索范围分别为[1×10-10,1×10-2]、[0.000 1,0.002]和[10,400]。图6所示为5个区域的CNN-LSTM的最优超参数组合的生成过程。表6展示了三套模型的预测精度。图7所示为三套模型的预测结果与实际值的对比情况。
图7所示,从整体角度出发,基于序列分解和重组的三套模型生成的预测值与实际值的曲线形状基本保持一致;另外,受制于各个模型的预测性能的固有差异,各个地区建立的模型的预测表现略有不同。传统LSTM模型因其时间记忆特性和在处理时间序列数据方面的优势,展示出良好的预测性能,如表6所示,城区LSTM的MAE和RMSE的均值分别为635.34和840.75;对比单一LSTM模型,CNN-LSTM组合模型凭借其较强的数据特征提取能力,预测性能明显提升,MAE、RMSE分别下降了167.46和113.63,拟合效果(R2)提升了0.1;在此基础上,使用北方苍鹰优化算法对CNN-LSTM模型的网络结构参数进行全局优化搜索有助于进一步提升预测精度,相较于CNN-LSTM组合模型,MAE和RMSE分别下降了75.43和23.78,拟合效果提高了0.02。类似的现象在其他区域也有所体现,以朝阳区为例,相较于LSTM和CNN-LSTM模型,NGO-CNN-LSTM模型的MAE分别降低811.39和122.54,RMSE分别降低了729和125.39,拟合效果(R2)分别提升了0.23和0.06。
为了进一步体现本文提出的基于二次分解方法的优越性,分别设置无预处理、一次分解和二次分解3种预处理情景,并据此构建了对应的预测模型。表7提供了3种预处理场景下的预测模型精度。图8展示了3种情景条件下的预测模型的评价指标对比情况。
图8所示,相较于其他两种数据预处理方法,采用二次分解进行数据预处理构建的三套预测模型的精度更高,拟合效果更佳。以城区的NGO-CNN-LSTM为例,相较于不处理和只进行一次分解构建两类情景,MAE分别降低了49.14和35.13,RMSE分别降低了18.66和32.9,拟合优度值R2分别提升了0.01和0.03。这是因为,基于二次分解的数据预处理方法能够对原始负荷序列进行有效的降噪处理,降低原始数据的波动性和非平稳特性。此外,尽管不处理和一次分解两种预处理方法可以展现出一定的预测性能,但其固有的缺乏数据的波动性处理和数据的非平稳性削减程度不够的劣势,导致上述两种方法无法满足更高的预测精度要求。上述情况在LSTM和CNN-LSTM两类模型也有所体现,由表7可知,以城区为例,相较于不处理和只进行一次分解,基于二次分解的LSTM模型的MAE分别降低了255.22和227.98,RMSE分别降低了212.37和206.63,拟合优度值R2分别提升了0.11和0.12;CNN-LSTM模型的MAE分别降低了110.77和97.31,RMSE分别降低了65.91和71.67,拟合优度值R2分别提升了0.05和0.03。由此可见,基于二次模态分解和预测模型的超参数优化均为模型预测精度提高的关键环节,未来应该持续关注两类方法的研究进展,以期进一步提升电力负荷组合预测模型的技术性和实用性。
为了进一步提升电力负荷的预测精度,本文提出了一种基于ALIF、NGO-VMD和FE的二次模态分解和重组的NGO-CNN-LSTM日前电力负荷短期组合预测模型,在运用CCM方法准确识别电力负荷的关键影响因素的基础上,创新性地使用ALIF、NGO-VMD和模糊熵相结合的二次模态分解方法对原始电力负荷数据进行降噪处理,利用具备强大全局优化搜索能力和较强鲁棒性的北方苍鹰优化算法确定CNN-LSTM模型关键参数的最佳组合,最终构建了基于二次分解的NGO-CNN-LSTM电力负荷组合预测模型。在北京市8个中心城区的电力负荷短期预测的应用结果表明,相较于其他类型的预测模型,本文提出的组合预测模型的精度更高且稳定性更好,具备一定的先进性和实用性。后续工作可以考虑为电力负荷的关键影响因素分配不同的权重系数,建立计及因素影响程度的相似日选取方法,进一步提升电力负荷预测模型的预测性能;另外,深入开展电力负荷的区间和概率预测,准确表征负荷预测结果的潜在不确定性,为其大规模的推广和应用奠定很好的基础。
  • 国网北京市电力公司科技项目(520206240001)
参考文献 引证文献
排序方式:
[1]
王晨, 李又轩, 吴其琦, 等. 基于SVM-STL-LSTM的区域短期电力负荷预测研究[J]. 水电能源科学, 2024(4): 215-218.
Wang Chen, Li Youxuan, Wu Qiqi, et al. Research on regional short-term load forecasting based on SVM-STL-LSTM[J]. Water Resources and Power, 2024(4): 215-218.
[2]
石卓见, 冉启武, 徐福聪. 基于聚合二次模态分解及Informer的短期负荷预测[J]. 电网技术, 2024(6): 2574-2583.
Shi Zhuojian, Ran Qiwu, Xu Fucong. Short-term load forecasting based on aggregated secondary decomposition and Informer[J]. Power System Technology, 2024(6): 2574-2583.
[3]
陈万志, 张思维, 王天元. 强化数据预处理的BLSTNet-CBAM短期电力负荷预测[J]. 计算机系统应用, 2024, 33(5): 47-56.
Chen Wanzhi, Zhang Siwei, Wang Tianyuan. Enhanced data preprocessing for BLSTNet-CBAM short-term power load forecasting[J]. Computer Systems & Applications, 2024, 33(5): 47-56.
[4]
葛亚明, 仇晨光, 谢丽荣, 等. 基于K-means聚类与LSTM模型的多能源耦合电力负荷预测[J/OL]. 现代电力: 1-8[2024-04-29]. https://doi.org/10.19725/j.cnki.1007-2322.2023.0110.
Ge Yaming, Chou Chenguang, Xie Lirong, et al. Research on Multi-energy coupled power load prediction based on K-means clustering and LSTM model[J]. Modern Electric Power, 1-8[2024-04-29]. https://doi.org/10.19725/j.cnki.1007-2322.2023.0110.
[5]
王增平, 赵兵, 纪维佳, 等. 基于GRU-NN模型的短期负荷预测方法[J]. 电力系统自动化, 2019, 43(5): 53-58.
Wang Zengping, Zhao Bing, Ji Weijia, et al. Short-term load forecasting method based on GRU-NN model[J]. Automation of Electric Power Systems, 2019, 43(5): 53-58.
[6]
刘伟, 蔡东升, 冯付勇, 等. 基于DCT-CNN-GRU的短期电力负荷预测研究[J/OL]. 电测与仪表: 1-11[2024-04-29]. http://kns.cnki.net/kcms/detail/23.1202.th.20240321.1323.002.html. http://kns.cnki.net/kcms/detail/23.1202.th.20240321.1323.002.html
Liu Wei, Cai Dongsheng, Feng Fuyong, et al. Research on short-term electric load forecasting based on DCT-CNN-GRU[J]. Electrical Measurement & Instrumentation, 1-11[2024-04-29]. http://kns.cnki.net/kcms/detail/23.1202.th.20240321.1323.002.html. http://kns.cnki.net/kcms/detail/23.1202.th.20240321.1323.002.html
[7]
张文栋, 刘子琨, 梁涛, 等. 基于CNN-LSTM的综合能源系统负荷预测模型[J]. 重庆邮电大学学报(自然科学版), 2023, 35(2): 254-262.
Zhang Wendong, Liu Zikun, Liang Tao, et al. Load prediction model of integrated energy system based on CNN-LSTM[J]. Journal of Chongqing University of Posts and Telecommunications( Natural Seience Edition, 2023, 35(2): 254-262.
[8]
Huy T H B, Vo D N, Nguyen K P, et al. Short-term load forecasting in power system using CNN-LSTM neural network[C]// 2023 Asia Meeting on Environment and Electrical Engineering (EEE-AM). Hanoi: IEEE, 2023: 1-6.
[9]
白宇, 薛贵军, 谢文举, 等. 改进GJO优化CNN-BiLSTM的热负荷预测模型[J/OL]. 中国测试: 1-9[2024-04-29]. http://kns.cnki.net/kcms/detail/51.1714.TB.20240328.1617.002.html. http://kns.cnki.net/kcms/detail/51.1714.TB.20240328.1617.002.html
Bai Yu, Xue Guijun, Xie Wenju, et al. Improved GJO optimized CNN-BiLSTM heat load prediction model[J]. China Measurement & Test. 1-9[2024-04-29]. http://kns.cnki.net/kcms/detail/51.1714.TB.20240328.1617.002.html. http://kns.cnki.net/kcms/detail/51.1714.TB.20240328.1617.002.html
[10]
Kiruthiga D, Manikandan V. Levy flight-particle swarm optimization-assisted BiLSTM plus dropout deep learning model for short-term load forecasting[J]. Neural Computing & Applications, 2023, 35: 2679-2700.
[11]
周磊, 竺筱晶. 基于MA-CNN-LSTM和自注意力机制的单变量短期电力负荷预测[J]. 科学技术与工程, 2024, 24(22): 9408-9416.
Zhou Lei, Zhu Xiaojing. Univariate short-term electrical load based on MA-CNN-LSTM-Self attention[J]. Science Technology and Engineering, 2024, 24(22): 9408-9416.
[12]
张淑娴, 江文韬, 陈玉花, 等. 基于二次模态分解的LSTM短期电力负荷预测[J]. 科学技术与工程, 2024, 24(7): 2759-2766.
Zhang Shuxian, Jiang Wentao, Chen Yuhua, et al. Short-term power load forecasting based on quadratic mode decomposition through LSTM[J]. Science Technology and Engineering, 2024, 24(7): 2759-2766.
[13]
Wang N, Li Z. Short term power load forecasting based on BES-VMD and CNN-Bi-LSTM method with error correction. Front[J]. Frontiers in Energy Research, 2023, 10: 1076529.
[14]
程友杰. 基于ALIF+MCKD方法的冷轧机轴承振动故障信号处理[J]. 山西冶金, 2023, 46(9): 70-71, 74.
Cheng Youjie. Vibration fault signal processing of bearing for cold rolling mill based on ALIF+MCKD method[J]. Shanxi Metallurgy, 2023, 46(9): 70-71, 74.
[15]
邓博文, 肖伸平, 廖世英. 二次分解组合CNN-LSTM的短期负荷预测[J]. 控制与信息技术, 2023(4): 54-60.
Deng Bowen, Xiao Shenping, Liao Shiying. Short-term load forecasting based on CNN-LSTM with quadratic decomposition combined[J]. Control and Information Technology, 2023(4): 54-60.
[16]
肖威, 方娜, 邓心. 基于VMD-LSTM-IPSO-GRU的电力负荷预测[J]. 科学技术与工程, 2024, 24(16): 6734-6741.
Xiao Wei, Fang Na, Deng Xin. Short-term load forecasting based on VMD-LSTM-IPSO-GRU[J]. Science Technology and Engineering, 2024, 24(16): 6734-6741.
[17]
李玉莲. 基于收敛交叉映射方法的3E系统因果推断[D]. 武汉: 武汉科技大学, 2023.
Li Yulian. Causal inference to the 3E system via convergent cross mapping[D]. Wuhan: Wuhan University of Science and Technology, 2023.
[18]
陈俊. 基于格兰杰方法和收敛交叉映射方法的因果分析研究[D]. 上海: 华东师范大学, 2020.
Chen Jun. Research on causality analysis based on granger method and convergent cross mapping method[D]. Shanghai: East China Normal University, 2020.
[19]
陈向俊, 傅军平, 于晓, 等. 基于ALIF和ISOMAP的机械设备故障识别方法研究[J]. 机床与液压, 2023, 51(5): 196-201.
Chen Xiangjun, Fu Junping, Yu Xiao, et al. Research on fault recognition method of mechanical equipment based on ALIF and ISOMAP[J]. Machine Tool & Hydraulics, 2023, 51(5): 196-201.
[20]
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 method for integrated energy system based on ALIF-LSTM and MULTI-TASK learing[J]. Acta Energiae Solaris Sinica, 2022, 43(9): 499-507.
[21]
宋江涛, 崔双喜, 刘洪广. 基于二次分解NGO-VMD残差项与长短时记忆神经网络的超短期风功率预测[J]. 科学技术与工程, 2023, 23(6): 2428-2437.
Song Jiangtao, Cui Shuangxi, Liu Hongguang. Ultra-short-term wind power prediction based on the quadratic decomposition NGO-VMD residual term and long short-term memory(LSTM)[J]. Science Technology and Engineering, 2023, 23(6): 2428-2437.
[22]
蒲维, 杨毅强, 张渊博, 等. 基于NGO-VMD-FCBF-Informer的电力负荷组合预测模型[J]. 智能计算机与应用, 2023, 13(11): 135-141.
Pu Wei, Yang Yiqiang, Zhang Yuanbo, et al. Powerload forecasting based on NGO-VMD-FCBF-informer[J]. Intelligent Computer and Applications, 2023, 13(11): 135-141.
[23]
刘杰, 从兰美, 夏远洋, 等. 基于DBO-VMD和IWOA-BILSTM神经网络组合模型的短期电力负荷预测[J]. 电力系统保护与控制, 2024, 52(8): 123-133.
Liu Jie, Cong Meilan, Xia Yuanyang, et al. Short-term power load prediction based on DBO-VMD and an IWOA-BILSTM neural network combination model[J]. Power System Protection and Control, 2024, 52(8): 123-133.
[24]
杨海柱, 田馥铭, 张鹏, 等. 基于CEEMD-FE和AOA-LSSVM的短期电力负荷预测[J]. 电力系统保护与控制, 2022, 50(13): 126-133.
Yang Haizhu, Tian Fuming, Zhang Peng, et al. Short-term load forecasting based on CEEMD-FE-AOA-LSSVM[J]. Power System Protection and Control, 2022, 50(13): 126-133.
[25]
高凯悦, 牟莉. 基于二次分解和GRU-attention的时间序列预测研究[J]. 国外电子测量技术, 2023, 42(2): 80-87.
Gao Kaiyue, Mou Li. Research on time series forecasting based on quadratic decomposition and GRU-attention[J]. Foreign Electronic Measurement Technology, 2023, 42(2): 80-87
[26]
刘旭丽, 莫毓昌, 吴哲, 等. 基于DWT-CNN-LSTM的超短期光伏发电功率预测[J]. 郑州大学学报(理学版), 2022, 54(4): 86-94.
Liu Xuli, Mo Yuchang, Wu Zhe, et al. Super-short-term photovoltaic power forecasting based on DWT-CNN-LSTM[J]. Journal of Zhengzhou University(Natural Science Edition), 2022, 54(4): 86-94.
2025年第25卷第11期
PDF下载
332
131
引用本文
BibTeX
文章信息
doi: 10.12404/j.issn.1671-1815.2404503
  • 接收时间:2024-06-17
  • 首发时间:2025-07-09
  • 出版时间:2025-04-18
补充材料
相关文章
文章信息
作者
出版历史
  • 收稿日期:2024-06-17
  • 修回日期:2024-10-29
基金
国网北京市电力公司科技项目(520206240001)
作者信息
    1 国网北京市电力公司, 北京 100031
    2 华北电力大学环境科学与工程学院, 北京 102206

通讯作者:

* 高胜强(1980—),男,汉族,北京人,硕士,高级工程师。研究方向:电力负荷预测、电力系统规划。E-mail:
参考文献
分享链接
https://castjournals.cast.org.cn/joweb/kxjsygc/CN/10.12404/j.issn.1671-1815.2404503
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