Article(id=1149780466875720021, tenantId=1146029695717560320, journalId=1146123166801305609, issueId=1149780466032669506, articleNumber=null, orderNo=null, doi=10.12404/j.issn.1671-1815.2403394, pmid=null, cstr=null, oa=null, hot=null, price=null, onlineType=0, articleFormat=0, articleType=null, articleTypeStr=research-article, receivedDate=1715097600000, receivedDateStr=2024-05-08, revisedDate=1736438400000, revisedDateStr=2025-01-10, acceptedDate=null, acceptedDateStr=null, onlineDate=1752058625190, onlineDateStr=2025-07-09, pubDate=1744041600000, pubDateStr=2025-04-08, doiRegisterDate=null, doiRegisterDateStr=null, onlineIssueDate=1752058625190, onlineIssueDateStr=2025-07-09, onlineJustAcceptDate=null, onlineJustAcceptDateStr=null, onlineFirstDate=null, onlineFirstDateStr=null, sourceXml=null, magXml=null, createTime=1752058625190, creator=13701087609, updateTime=1752058625190, updator=13701087609, issue=Issue{id=1149780466032669506, tenantId=1146029695717560320, journalId=1146123166801305609, year='2025', volume='25', issue='10', pageStart='3969', pageEnd='4395', issueExtLink='null', onlineDate='null', pubDate='null', beforeIssueId=null, nextIssueId=null, price=null, status=1, issueComplete=1, articleOrder=1, issueType=-1, specialIssue=0, createTime=1752058624990, creator=13701087609, updateTime=1768456644259, updator=13701087609, preIssue=null, nextIssue=null, ext={EN=IssueExt(id=1218558743898411553, tenantId=1146029695717560320, journalId=1146123166801305609, issueId=1149780466032669506, language=EN, specialIssueTitle=, coverIllustrator=, specialIssueEditor=, specialIssueAbout=), CN=IssueExt(id=1218558743898411554, tenantId=1146029695717560320, journalId=1146123166801305609, issueId=1149780466032669506, language=CN, specialIssueTitle=, coverIllustrator=, specialIssueEditor=, specialIssueAbout=)}, issueFiles=null}, startPage=4145, endPage=4155, ext={EN=ArticleExt(id=1149780467072852311, articleId=1149780466875720021, tenantId=1146029695717560320, journalId=1146123166801305609, language=EN, title=Probabilistic Prediction of Photovoltaic Power Interval Based on Feature Mining and Improved TCN-BiGRU, columnId=1156262733675876713, journalTitle=Science Technology and Engineering, columnName=Papers·Electrical Technology, runingTitle=null, highlight=null, articleAbstract=

Photovoltaic power generation has an important place in the energy sector. In order to accurately quantify the uncertainty and fluctuation range of PV(photovoltaic) power and to improve the comprehensiveness of interval forecasts, a probabilistic prediction method for PV power intervals based on feature mining with improved TCN-BiGRU was proposed. First, the maximum information coefficient and symbolic transfer entropy causal analysis were utilized to screen the meteorological features, remove redundant information, and construct global horizontal radiation trend features, seasonal features, and weather clustering features to provide more effective information. Subsequently, the TCN-BiGRU model was improved by combining the temporal pattern attention mechanism and quantile regression methods to construct a combined model for interval prediction. Finally, the probabilistic prediction results are generated using the KDE method of empirical bandwidth selection with scatter measure semi-polar optimization. The proposed method is analyzed by real PV plant data, which verifies the high reliability and applicability of the proposed method in PV power interval probability prediction.

, correspAuthors=Hao WU, 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=Yuan CAI, Hao WU, Dan TANG), CN=ArticleExt(id=1149780498567880844, articleId=1149780466875720021, tenantId=1146029695717560320, journalId=1146123166801305609, language=CN, title=基于特征挖掘与改进TCN-BiGRU的光伏功率区间概率预测, columnId=1156262734506353627, journalTitle=科学技术与工程, columnName=论文·电工技术, runingTitle=null, highlight=null, articleAbstract=

光伏发电在能源领域中具有重要地位。为了准确量化光伏发电功率的不确定性和波动范围,并提高区间预测的综合性能,提出了一种基于特征挖掘与改进TCN-BiGRU的光伏功率区间概率预测方法。首先,利用最大信息系数和符号传递熵因果分析,对气象特征进行筛选,剔除冗余信息,并构造全球水平辐射趋势特征、季节性特征和天气聚类特征以提供更多有效信息。随后,结合时间模式注意力机制和分位数回归方法对TCN-BiGRU模型进行改进,构建组合模型进行区间预测。最后,采用散度度量半极差优化经验带宽选择的核密度估计(kernel density estimation,KDE)方法生成概率预测结果。通过真实光伏电站数据进行分析,验证了所提方法在光伏功率区间概率预测中具有较高的可靠性和适用性。

, correspAuthors=吴浩, authorNote=null, correspAuthorsNote=
* 吴浩(1980—),男,汉族,四川成都人,博士,教授。研究方向:电力系统保护与智能控制。E-mail:
, copyrightStatement=null, copyrightOwner=null, extLink=null, articleAbsUrl=null, sourceXml=JtAosYM4X5PTuvjADidHAw==, magXml=GrUdfO3S3unWsg2B/2KOhQ==, pdfUrl=null, pdf=UEn9mU5YKW10PtozPNj32w==, pdfFileSize=8971920, pdfExtLink=null, richHtmlUrl=null, mobilePdfUrl=null, reviewReport=null, pdfFirstPage=null, abstractGraph=VUH6vigpWJ81Ckr3ra7YVQ==, abstractGraphContent=null, abstractVideo=null, citation=null, cebUrl=null, magXmlContent=NURlA7sr5fh0Tu4QYJZJJw==, mapNumber=null, authorCompany=null, fund=null, authors=

蔡源(1999—),男,汉族,四川大竹人,硕士研究生。研究方向:光伏输出功率预测。E-mail:

, authorsList=蔡源, 吴浩, 唐丹)}, authors=[Author(id=1218525101344084885, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, orderNo=0, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=1398768033@qq.com, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1218525101482496928, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, authorId=1218525101344084885, language=EN, stringName=Yuan CAI, firstName=Yuan, middleName=null, lastName=CAI, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1 School of Automation and Information Engineering, Sichuan University of Science & Engineering, Yibin 644000, China
2 Sichuan Key Laboratory of Artificial Intelligence, Yibin 644000, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1218525101616714665, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, authorId=1218525101344084885, language=CN, stringName=蔡源, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1 四川轻化工大学自动化与信息工程学院, 宜宾 644000
2 人工智能四川省重点实验室, 宜宾 644000, bio={"content":"

蔡源(1999—),男,汉族,四川大竹人,硕士研究生。研究方向:光伏输出功率预测。E-mail:

"}, bioImg=null, bioContent=

蔡源(1999—),男,汉族,四川大竹人,硕士研究生。研究方向:光伏输出功率预测。E-mail:

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1218525101016929156, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, xref=1, ext=[AuthorCompanyExt(id=1218525101037900677, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, companyId=1218525101016929156, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 School of Automation and Information Engineering, Sichuan University of Science & Engineering, Yibin 644000, China), AuthorCompanyExt(id=1218525101054677894, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, companyId=1218525101016929156, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 四川轻化工大学自动化与信息工程学院, 宜宾 644000)]), AuthorCompany(id=1218525101230838670, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, xref=2, ext=[AuthorCompanyExt(id=1218525101239227279, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, companyId=1218525101230838670, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 Sichuan Key Laboratory of Artificial Intelligence, Yibin 644000, China), AuthorCompanyExt(id=1218525101247615888, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, companyId=1218525101230838670, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 人工智能四川省重点实验室, 宜宾 644000)])]), Author(id=1218525101738349489, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, orderNo=1, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=11305076@qq.com, emailSecond=null, emailThird=null, correspondingAuthor=1, authorType=1, ext={EN=AuthorExt(id=1218525101876761530, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, authorId=1218525101738349489, language=EN, stringName=Hao WU, firstName=Hao, middleName=null, lastName=WU, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, *, address=1 School of Automation and Information Engineering, Sichuan University of Science & Engineering, Yibin 644000, China
2 Sichuan Key Laboratory of Artificial Intelligence, Yibin 644000, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1218525101994202049, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, authorId=1218525101738349489, language=CN, stringName=吴浩, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, *, address=1 四川轻化工大学自动化与信息工程学院, 宜宾 644000
2 人工智能四川省重点实验室, 宜宾 644000, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1218525101016929156, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, xref=1, ext=[AuthorCompanyExt(id=1218525101037900677, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, companyId=1218525101016929156, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 School of Automation and Information Engineering, Sichuan University of Science & Engineering, Yibin 644000, China), AuthorCompanyExt(id=1218525101054677894, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, companyId=1218525101016929156, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 四川轻化工大学自动化与信息工程学院, 宜宾 644000)]), AuthorCompany(id=1218525101230838670, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, xref=2, ext=[AuthorCompanyExt(id=1218525101239227279, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, companyId=1218525101230838670, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 Sichuan Key Laboratory of Artificial Intelligence, Yibin 644000, China), AuthorCompanyExt(id=1218525101247615888, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, companyId=1218525101230838670, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 人工智能四川省重点实验室, 宜宾 644000)])]), Author(id=1218525102078088135, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, 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=1218525102220694480, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, authorId=1218525102078088135, language=EN, stringName=Dan TANG, firstName=Dan, middleName=null, lastName=TANG, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1 School of Automation and Information Engineering, Sichuan University of Science & Engineering, Yibin 644000, China
2 Sichuan Key Laboratory of Artificial Intelligence, Yibin 644000, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1218525102296191955, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, authorId=1218525102078088135, language=CN, stringName=唐丹, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1 四川轻化工大学自动化与信息工程学院, 宜宾 644000
2 人工智能四川省重点实验室, 宜宾 644000, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1218525101016929156, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, xref=1, ext=[AuthorCompanyExt(id=1218525101037900677, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, companyId=1218525101016929156, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 School of Automation and Information Engineering, Sichuan University of Science & Engineering, Yibin 644000, China), AuthorCompanyExt(id=1218525101054677894, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, companyId=1218525101016929156, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 四川轻化工大学自动化与信息工程学院, 宜宾 644000)]), AuthorCompany(id=1218525101230838670, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, xref=2, ext=[AuthorCompanyExt(id=1218525101239227279, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, companyId=1218525101230838670, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 Sichuan Key Laboratory of Artificial Intelligence, Yibin 644000, China), AuthorCompanyExt(id=1218525101247615888, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, companyId=1218525101230838670, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 人工智能四川省重点实验室, 宜宾 644000)])])], keywords=[Keyword(id=1218525102552044513, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, language=EN, orderNo=1, keyword=feature mining), Keyword(id=1218525102665290730, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, language=EN, orderNo=2, keyword=TCN-BiGRU), Keyword(id=1218525102749176820, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, language=EN, orderNo=3, keyword=quantile regression), Keyword(id=1218525102904366074, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, language=EN, orderNo=4, keyword=kernel density estimation), Keyword(id=1218525103021806592, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, language=EN, orderNo=5, keyword=interval probability prediction), Keyword(id=1218525103156023303, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, language=CN, orderNo=1, keyword=特征挖掘), Keyword(id=1218525103269269517, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, language=CN, orderNo=2, keyword=TCN-BiGRU), Keyword(id=1218525103344766993, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, language=CN, orderNo=3, keyword=分位数回归), Keyword(id=1218525103449624599, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, language=CN, orderNo=4, keyword=核密度估计), Keyword(id=1218525103571259420, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, language=CN, orderNo=5, keyword=区间概率预测)], refs=[Reference(id=1218525108168216947, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, doi=null, pmid=null, pmcid=null, year=2023, volume=null, issue=10, pageStart=74, pageEnd=83, url=null, language=null, rfNumber=[1], rfOrder=0, authorNames=龙勇, 宋敏, journalName=改革, refType=null, unstructuredReference=龙勇, 宋敏. 全球能源安全大变局下保障我国能源安全的思路与方略[J]. 改革, 2023(10): 74-83., articleTitle=全球能源安全大变局下保障我国能源安全的思路与方略, refAbstract=null), Reference(id=1218525108302434692, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, doi=null, pmid=null, pmcid=null, year=2023, volume=null, issue=10, pageStart=74, pageEnd=83, url=null, language=null, rfNumber=[1], rfOrder=1, authorNames=Long Yong, Song Min, journalName=Reform, refType=null, unstructuredReference=Long Yong, Song Min. Ideas and strategies for ensuring China's energy security in the context of global energy security upheaval[J]. Reform, 2023(10): 74-83., articleTitle=Ideas and strategies for ensuring China's energy security in the context of global energy security upheaval, refAbstract=null), Reference(id=1218525108445041044, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, doi=null, pmid=null, pmcid=null, year=2023, volume=null, issue=7, pageStart=71, pageEnd=null, url=null, language=null, rfNumber=[2], rfOrder=2, authorNames=国家能源局, journalName=电力勘测设计, refType=null, unstructuredReference=国家能源局. 我国可再生能源发电总装机突破13亿千瓦[J]. 电力勘测设计, 2023(7): 71., articleTitle=我国可再生能源发电总装机突破13亿千瓦, refAbstract=null), Reference(id=1218525108558287264, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, doi=null, pmid=null, pmcid=null, year=2023, volume=null, issue=7, pageStart=71, pageEnd=null, url=null, language=null, rfNumber=[2], rfOrder=3, authorNames=National Energy Agency, journalName=Electric Power Survey and Design, refType=null, unstructuredReference=National Energy Agency. China's renewable energy power generation total installed capacity exceeds 1.3 billion kilowatts[J]. Electric Power Survey and Design, 2023(7): 71., articleTitle=China's renewable energy power generation total installed capacity exceeds 1.3 billion kilowatts, refAbstract=null), Reference(id=1218525108675727787, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, doi=null, pmid=null, pmcid=null, year=2023, volume=23, issue=27, pageStart=11664, pageEnd=11672, url=null, language=null, rfNumber=[3], rfOrder=4, authorNames=张弛, 朱宗玖, journalName=科学技术与工程, refType=null, unstructuredReference=张弛, 朱宗玖. 基于误差修正的短期光伏功率预测[J]. 科学技术与工程, 2023, 23(27): 11664-11672., articleTitle=基于误差修正的短期光伏功率预测, refAbstract=null), Reference(id=1218525108805751223, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, doi=null, pmid=null, pmcid=null, year=2023, volume=23, issue=27, pageStart=11664, pageEnd=11672, url=null, language=null, rfNumber=[3], rfOrder=5, authorNames=Zhang Chi, Zhu Zongjiu, journalName=Science Technology and Engineering, refType=null, unstructuredReference=Zhang Chi, Zhu Zongjiu. Short-term PV power prediction based on error correction[J]. Science Technology and Engineering, 2023, 23(27): 11664-11672., articleTitle=Short-term PV power prediction based on error correction, refAbstract=null), Reference(id=1218525108977717703, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, doi=null, pmid=null, pmcid=null, year=2016, volume=136, issue=null, pageStart=78, pageEnd=111, url=null, language=null, rfNumber=[4], rfOrder=6, authorNames=Antonanzas J, Osorio N, Escobar R, journalName=Solar Energy, refType=null, unstructuredReference=Antonanzas J, Osorio N, Escobar R, et al. Review of photovoltaic power forecasting[J]. Solar Energy, 2016, 136: 78-111., articleTitle=Review of photovoltaic power forecasting, refAbstract=null), Reference(id=1218525109090963926, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, doi=null, pmid=null, pmcid=null, year=2022, volume=14, issue=24, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[5], rfOrder=7, authorNames=Iheanetu K J, journalName=Sustainability, refType=null, unstructuredReference=Iheanetu K J. Solar photovoltaic power forecasting: a review[J]. Sustainability, 2022, 14(24). DOI: 10.3390/su142417005., articleTitle=Solar photovoltaic power forecasting: a review, refAbstract=null), Reference(id=1218525109195821535, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, doi=null, pmid=null, pmcid=null, year=2023, volume=23, issue=34, pageStart=14596, pageEnd=14602, url=null, language=null, rfNumber=[6], rfOrder=8, authorNames=刘运超, 杨宁, 崔承刚, journalName=科学技术与工程, refType=null, unstructuredReference=刘运超, 杨宁, 崔承刚, 等. 基于时空信息的区域内光伏电站功率预测[J]. 科学技术与工程, 2023, 23(34): 14596-14602., articleTitle=基于时空信息的区域内光伏电站功率预测, refAbstract=null), Reference(id=1218525109338427884, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, doi=null, pmid=null, pmcid=null, year=2023, volume=23, issue=34, pageStart=14596, pageEnd=14602, url=null, language=null, rfNumber=[6], rfOrder=9, authorNames=Liu Yunchao, Yang Ning, Cui Chenggang, journalName=Science Technology and Engineering, refType=null, unstructuredReference=Liu Yunchao, Yang Ning, Cui Chenggang, et al. Power prediction of regional photovoltaic plants based on spatio-temporal information[J]. Science Technology and Engineering, 2023, 23(34): 14596-14602., articleTitle=Power prediction of regional photovoltaic plants based on spatio-temporal information, refAbstract=null), Reference(id=1218525109455868408, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, doi=null, pmid=null, pmcid=null, year=2023, volume=266, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[7], rfOrder=10, authorNames=Fan H, Zhen Z, Liu N, journalName=Energy, refType=null, unstructuredReference=Fan H, Zhen Z, Liu N, et al. Fluctuation pattern recognition based ultra-short-term wind power probabilistic forecasting method[J]. Energy, 2023, 266.DOI: 10.1016/j.energy.2022.126420., articleTitle=Fluctuation pattern recognition based ultra-short-term wind power probabilistic forecasting method, refAbstract=null), Reference(id=1218525109598474761, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, doi=null, pmid=null, pmcid=null, year=2022, volume=47, issue=4, pageStart=986, pageEnd=997, url=null, language=null, rfNumber=[8], rfOrder=11, authorNames=宋绍剑, 罗世坚, 李国进, journalName=广西大学学报(自然科学版), refType=null, unstructuredReference=宋绍剑, 罗世坚, 李国进, 等. 基于MPA-LSTM模型和Bootstrap方法的短期光伏功率区间预测[J]. 广西大学学报(自然科学版), 2022, 47(4): 986-997., articleTitle=基于MPA-LSTM模型和Bootstrap方法的短期光伏功率区间预测, refAbstract=null), Reference(id=1218525109690749461, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, doi=null, pmid=null, pmcid=null, year=2022, volume=47, issue=4, pageStart=986, pageEnd=997, url=null, language=null, rfNumber=[8], rfOrder=12, authorNames=Song Shaojian, Luo Shijian, Li Guojin, journalName=Journal of Guangxi University(Natural Science Edition), refType=null, unstructuredReference=Song Shaojian, Luo Shijian, Li Guojin, et al. A short-term PV power interval prediction based on MPA-LSTM network model and Bootstrap method[J]. Journal of Guangxi University(Natural Science Edition), 2022, 47(4): 986-997., articleTitle=A short-term PV power interval prediction based on MPA-LSTM network model and Bootstrap method, refAbstract=null), Reference(id=1218525109808189985, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, doi=null, pmid=null, pmcid=null, year=2023, volume=275, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[9], rfOrder=13, authorNames=Yu M, Niu D, Wang K, journalName=Energy, refType=null, unstructuredReference=Yu M, Niu D, Wang K, et al. Short-term photovoltaic power point-interval forecasting based on double-layer decomposition and WOA-BiLSTM-Attention and considering weather classification[J]. Energy, 2023, 275.DOI: 10.1016/j.energy.2023.127348., articleTitle=Short-term photovoltaic power point-interval forecasting based on double-layer decomposition and WOA-BiLSTM-Attention and considering weather classification, refAbstract=null), Reference(id=1218525109900464685, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, doi=null, pmid=null, pmcid=null, year=2023, volume=47, issue=23, pageStart=152, pageEnd=161, url=null, language=null, rfNumber=[10], rfOrder=14, authorNames=赵耀, 高少炜, 李东东, journalName=电力系统自动化, refType=null, unstructuredReference=赵耀, 高少炜, 李东东, 等. 基于天气相似聚类与QRNN的短期光伏功率区间概率预测[J]. 电力系统自动化, 2023, 47(23): 152-161., articleTitle=基于天气相似聚类与QRNN的短期光伏功率区间概率预测, refAbstract=null), Reference(id=1218525109971767863, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, doi=null, pmid=null, pmcid=null, year=2023, volume=47, issue=23, pageStart=152, pageEnd=161, url=null, language=null, rfNumber=[10], rfOrder=15, authorNames=Zhao Yao, Gao Shaowei, Li Dongdong, journalName=Power System Automation, refType=null, unstructuredReference=Zhao Yao, Gao Shaowei, Li Dongdong, et al. Short-term interval probability prediction of photovoltaic power based on weather similarity clustering and quantile regression neural network[J]. Power System Automation, 2023, 47(23): 152-161., articleTitle=Short-term interval probability prediction of photovoltaic power based on weather similarity clustering and quantile regression neural network, refAbstract=null), Reference(id=1218525110093402690, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, doi=null, pmid=null, pmcid=null, year=2020, volume=44, issue=4, pageStart=1343, pageEnd=1350, url=null, language=null, rfNumber=[11], rfOrder=16, authorNames=庞昊, 高金峰, 杜耀恒, journalName=电网技术, refType=null, unstructuredReference=庞昊, 高金峰, 杜耀恒. 基于时间卷积网络分位数回归的短期负荷概率密度预测方法[J]. 电网技术, 2020, 44(4): 1343-1350., articleTitle=基于时间卷积网络分位数回归的短期负荷概率密度预测方法, refAbstract=null), Reference(id=1218525110345060948, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, doi=null, pmid=null, pmcid=null, year=2020, volume=44, issue=4, pageStart=1343, pageEnd=1350, url=null, language=null, rfNumber=[11], rfOrder=17, authorNames=Pang Hao, Gao Jinfeng, Du Yaoheng, journalName=Grid Technology, refType=null, unstructuredReference=Pang Hao, Gao Jinfeng, Du Yaoheng. A short-term load probability density prediction based on quantile regression of time convolution network[J]. Grid Technology, 2020, 44(4): 1343-1350., articleTitle=A short-term load probability density prediction based on quantile regression of time convolution network, refAbstract=null), Reference(id=1218525110479278686, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, doi=null, pmid=null, pmcid=null, year=2024, volume=236, issue=null, pageStart=121286, pageEnd=null, url=null, language=null, rfNumber=[12], rfOrder=18, authorNames=Wang J, He M, Jiang W, journalName=Expert Systems with Applications, refType=null, unstructuredReference=Wang J, He M, Jiang W. Causal carbon price interval prediction using lower upper bound estimation combined with asymmetric multi-objective evolutionary algorithm and long short-term memory[J]. Expert Systems with Applications, 2024, 236: 121286., articleTitle=Causal carbon price interval prediction using lower upper bound estimation combined with asymmetric multi-objective evolutionary algorithm and long short-term memory, refAbstract=null), Reference(id=1218525110626079336, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, doi=null, pmid=null, pmcid=null, year=2023, volume=32, issue=5, pageStart=943, pageEnd=952, url=null, language=null, rfNumber=[13], rfOrder=19, authorNames=张铮, 高森, 张泽扬, journalName=上海海洋大学学报, refType=null, unstructuredReference=张铮, 高森, 张泽扬. 基于特征构造预处理与TCN-BiGRU的池塘溶解氧预测模型[J]. 上海海洋大学学报, 2023, 32(5): 943-952., articleTitle=基于特征构造预处理与TCN-BiGRU的池塘溶解氧预测模型, refAbstract=null), Reference(id=1218525110781268603, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, doi=null, pmid=null, pmcid=null, year=2023, volume=32, issue=5, pageStart=943, pageEnd=952, url=null, language=null, rfNumber=[13], rfOrder=20, authorNames=Zhang Zheng, Gao Sen, Zhang Zeyang, journalName=Journal of Shanghai Ocean University, refType=null, unstructuredReference=Zhang Zheng, Gao Sen, Zhang Zeyang. Prediction model of dissolved oxygen in pond based on feature construction pretreatment and TCN-BiGRU[J]. Journal of Shanghai Ocean University, 2023, 32(5): 943-952., articleTitle=Prediction model of dissolved oxygen in pond based on feature construction pretreatment and TCN-BiGRU, refAbstract=null), Reference(id=1218525110890320519, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, doi=null, pmid=null, pmcid=null, year=2024, volume=288, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[14], rfOrder=21, authorNames=Liu T, Qi S, Qiao X, journalName=Energy, refType=null, unstructuredReference=Liu T, Qi S, Qiao X, et al. A hybrid short-term wind power point-interval prediction model based on combination of improved preprocessing methods and entropy weighted GRU quantile regression network[J]. Energy, 2024, 288.DOI: 10.1016/j.energy.2023.129904., articleTitle=A hybrid short-term wind power point-interval prediction model based on combination of improved preprocessing methods and entropy weighted GRU quantile regression network, refAbstract=null), Reference(id=1218525110995178127, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, doi=null, pmid=null, pmcid=null, year=2024, volume=36, issue=3, pageStart=151, pageEnd=158, url=null, language=null, rfNumber=[15], rfOrder=22, authorNames=欧阳卫年, 赵紫昱, 陈渊睿, journalName=电力系统及其自动化学报, refType=null, unstructuredReference=欧阳卫年, 赵紫昱, 陈渊睿. 自样本特征构造的1DCNN-BiLSTM网侧光伏功率预测[J]. 电力系统及其自动化学报, 2024, 36(3): 151-158., articleTitle=自样本特征构造的1DCNN-BiLSTM网侧光伏功率预测, refAbstract=null), Reference(id=1218525111112618649, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, doi=null, pmid=null, pmcid=null, year=2024, volume=36, issue=3, pageStart=151, pageEnd=158, url=null, language=null, rfNumber=[15], rfOrder=23, authorNames=Ouyang Weinian, Zhao Ziyu, Chen Yuanrui, journalName=Journal of Power System and Automation, refType=null, unstructuredReference=Ouyang Weinian, Zhao Ziyu, Chen Yuanrui. 1DCNN-BiLSTM method of grid-side photovoltaic power prediction with self-sampled feature construction[J]. Journal of Power System and Automation, 2024, 36(3): 151-158., articleTitle=1DCNN-BiLSTM method of grid-side photovoltaic power prediction with self-sampled feature construction, refAbstract=null), Reference(id=1218525111242642086, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, doi=null, pmid=null, pmcid=null, year=2023, volume=44, issue=8, pageStart=467, pageEnd=476, url=null, language=null, rfNumber=[16], rfOrder=24, authorNames=王旭光, 张可, 李潇, journalName=太阳能学报, refType=null, unstructuredReference=王旭光, 张可, 李潇, 等. 基于变量相关注意力机制的短期风速预测[J]. 太阳能学报, 2023, 44(8): 467-476., articleTitle=基于变量相关注意力机制的短期风速预测, refAbstract=null), Reference(id=1218525111385248436, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, doi=null, pmid=null, pmcid=null, year=2023, volume=44, issue=8, pageStart=467, pageEnd=476, url=null, language=null, rfNumber=[16], rfOrder=25, authorNames=Wang Xuguang, Zhang Ke, Li Xiao, journalName=Journal of Solar Energy, refType=null, unstructuredReference=Wang Xuguang, Zhang Ke, Li Xiao, et al. Short-term wind speed forecasting based on variable correlation attention mechanism[J]. Journal of Solar Energy, 2023, 44(8): 467-476., articleTitle=Short-term wind speed forecasting based on variable correlation attention mechanism, refAbstract=null), Reference(id=1218525111469134524, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, doi=null, pmid=null, pmcid=null, year=2008, volume=100, issue=15, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[17], rfOrder=26, authorNames=Staniek M, Lehnertz K, journalName=Physical Review Letters, refType=null, unstructuredReference=Staniek M, Lehnertz K. Symbolic transfer entropy[J]. Physical Review Letters, 2008, 100(15). DOI: 10.1103/PhysRevLett.100.158101., articleTitle=Symbolic transfer entropy, refAbstract=null), Reference(id=1218525111641101001, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, doi=null, pmid=null, pmcid=null, year=2002, volume=52, issue=9, pageStart=1096, pageEnd=1101, url=null, language=null, rfNumber=[18], rfOrder=27, authorNames=McKendry I G, journalName=Journal of the Air & Waste Management Association, refType=null, unstructuredReference=McKendry I G. Evaluation of artificial neural networks for fine particulate pollution (PM10 and PM2.5) forecasting[J]. Journal of the Air & Waste Management Association, 2002, 52(9): 1096-1101., articleTitle=Evaluation of artificial neural networks for fine particulate pollution (PM10 and PM2.5) forecasting, refAbstract=null), Reference(id=1218525111779513047, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, doi=null, pmid=null, pmcid=null, year=2023, volume=51, issue=18, pageStart=81, pageEnd=93, url=null, language=null, rfNumber=[19], rfOrder=28, authorNames=季玉琦, 严亚帮, 和萍, journalName=电力系统保护与控制, refType=null, unstructuredReference=季玉琦, 严亚帮, 和萍, 等. 基于K-Medoids聚类与栅格法提取负荷曲线特征的CNN-LSTM短期负荷预测[J]. 电力系统保护与控制, 2023, 51(18): 81-93., articleTitle=基于K-Medoids聚类与栅格法提取负荷曲线特征的CNN-LSTM短期负荷预测, refAbstract=null), Reference(id=1218525111930508000, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, doi=null, pmid=null, pmcid=null, year=2023, volume=51, issue=18, pageStart=81, pageEnd=93, url=null, language=null, rfNumber=[19], rfOrder=29, authorNames=Ji Yuqi, Yan Yabang, He Ping, journalName=Power System Protection and Control, refType=null, unstructuredReference=Ji Yuqi, Yan Yabang, He Ping, et al. CNN-LSTM short-term load forecasting basedon the K-Medoids clustering and grid method to extract load curve features[J]. Power System Protection and Control, 2023, 51(18): 81-93., articleTitle=CNN-LSTM short-term load forecasting basedon the K-Medoids clustering and grid method to extract load curve features, refAbstract=null), Reference(id=1218525112031171305, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, doi=null, pmid=null, pmcid=null, year=2017, volume=null, issue=null, pageStart=1, pageEnd=6, url=null, language=null, rfNumber=[20], rfOrder=30, authorNames=Kapoor A, Singhal A, journalName=3rd International Conference on Computational Intelligence & Communication Technology (CICT), refType=null, unstructuredReference=Kapoor A, Singhal A. A comparative study of K-Means, K-Means++ and Fuzzy C-Means clustering algorithms[C]// 3rd International Conference on Computational Intelligence & Communication Technology (CICT). New York: IEEE, 2017: 1-6., articleTitle=A comparative study of K-Means, K-Means++ and Fuzzy C-Means clustering algorithms, refAbstract=null), Reference(id=1218525112131834613, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, doi=null, pmid=null, pmcid=null, year=2018, volume=39, issue=11, pageStart=3232, pageEnd=3238, url=null, language=null, rfNumber=[21], rfOrder=31, authorNames=吴红斌, 何叶, 赵波, journalName=太阳能学报, refType=null, unstructuredReference=吴红斌, 何叶, 赵波, 等. 基于改进K-means聚类算法的风电场动态等值[J]. 太阳能学报, 2018, 39(11): 3232-3238., articleTitle=基于改进K-means聚类算法的风电场动态等值, refAbstract=null), Reference(id=1218525112236692222, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, doi=null, pmid=null, pmcid=null, year=2018, volume=39, issue=11, pageStart=3232, pageEnd=3238, url=null, language=null, rfNumber=[21], rfOrder=32, authorNames=Wu Hongbin, He Ye, Zhao Bo, journalName=Journal of Solar Energy, refType=null, unstructuredReference=Wu Hongbin, He Ye, Zhao Bo, et al. Research on dynamic equivalent of wind farm based onimproved K-means clustering algorithm[J]. Journal of Solar Energy, 2018, 39(11): 3232-3238., articleTitle=Research on dynamic equivalent of wind farm based onimproved K-means clustering algorithm, refAbstract=null), Reference(id=1218525112354132743, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, doi=null, pmid=null, pmcid=null, year=2020, volume=24, issue=null, pageStart=16453, pageEnd=16482, url=null, language=null, rfNumber=[22], rfOrder=33, authorNames=Hewage P, Behera A, Trovati M, journalName=Soft Computing, refType=null, unstructuredReference=Hewage P, Behera A, Trovati M, et al. Temporal convolutional neural (TCN) network for an effective weather forecasting using time-series data from the local weather station[J]. Soft Computing, 2020, 24: 16453-16482., articleTitle=Temporal convolutional neural (TCN) network for an effective weather forecasting using time-series data from the local weather station, refAbstract=null), Reference(id=1218525112450601748, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, doi=null, pmid=null, pmcid=null, year=2020, volume=21, issue=6, pageStart=2099, pageEnd=2111, url=null, language=null, rfNumber=[23], rfOrder=34, authorNames=Lin X, Quan Z, Wang Z J, journalName=Briefings in Bioinformatics, refType=null, unstructuredReference=Lin X, Quan Z, Wang Z J, et al. A novel molecular representation with BiGRU neural networks for learning atom[J]. Briefings in Bioinformatics, 2020, 21(6): 2099-2111., articleTitle=A novel molecular representation with BiGRU neural networks for learning atom, refAbstract=null), Reference(id=1218525112538682137, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, doi=null, pmid=null, pmcid=null, year=2019, volume=108, issue=null, pageStart=1421, pageEnd=1441, url=null, language=null, rfNumber=[24], rfOrder=35, authorNames=Shih S Y, Sun F K, Lee H, journalName=Machine Learning, refType=null, unstructuredReference=Shih S Y, Sun F K, Lee H. Temporal pattern attention for multivariate time series forecasting[J]. Machine Learning, 2019, 108: 1421-1441., articleTitle=Temporal pattern attention for multivariate time series forecasting, refAbstract=null), Reference(id=1218525112664511268, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, doi=null, pmid=null, pmcid=null, year=2022, volume=43, issue=2, pageStart=479, pageEnd=485, url=null, language=null, rfNumber=[25], rfOrder=36, authorNames=王晓东, 鞠邦国, 刘颖明, journalName=太阳能学报, refType=null, unstructuredReference=王晓东, 鞠邦国, 刘颖明, 等. 基于QR-NFGLSTM与核密度估计的风电功率概率预测[J]. 太阳能学报, 2022, 43(2): 479-485., articleTitle=基于QR-NFGLSTM与核密度估计的风电功率概率预测, refAbstract=null), Reference(id=1218525112760980270, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, doi=null, pmid=null, pmcid=null, year=2022, volume=43, issue=2, pageStart=479, pageEnd=485, url=null, language=null, rfNumber=[25], rfOrder=37, authorNames=Wang Xiaodong, Ju Bangguo, Liu Yingming, journalName=Journal of Solar Energy, refType=null, unstructuredReference=Wang Xiaodong, Ju Bangguo, Liu Yingming, et al. Probability prediction of wind power based on QR-NFGLSTM and kernel density estimation[J]. Journal of Solar Energy, 2022, 43(2): 479-485., articleTitle=Probability prediction of wind power based on QR-NFGLSTM and kernel density estimation, refAbstract=null), Reference(id=1218525112907780924, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, doi=null, pmid=null, pmcid=null, year=2022, volume=48, issue=11, pageStart=4372, pageEnd=4388, url=null, language=null, rfNumber=[26], rfOrder=38, authorNames=王开艳, 杜浩东, 贾嵘, journalName=高电压技术, refType=null, unstructuredReference=王开艳, 杜浩东, 贾嵘, 等. 基于相似日聚类和QR-CNN-BiLSTM模型的光伏功率短期区间概率预测[J]. 高电压技术, 2022, 48(11): 4372-4388., articleTitle=基于相似日聚类和QR-CNN-BiLSTM模型的光伏功率短期区间概率预测, refAbstract=null), Reference(id=1218525113058775881, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, doi=null, pmid=null, pmcid=null, year=2022, volume=48, issue=11, pageStart=4372, pageEnd=4388, url=null, language=null, rfNumber=[26], rfOrder=39, authorNames=Wang Kaiyan, Du Haodong, Jia Rong, journalName=High Voltage Technology, refType=null, unstructuredReference=Wang Kaiyan, Du Haodong, Jia Rong, et al. Short-term interval probability prediction of photovoltaic power based on similar daily clustering and QR-CNN-BiLSTM model[J]. High Voltage Technology, 2022, 48(11): 4372-4388., articleTitle=Short-term interval probability prediction of photovoltaic power based on similar daily clustering and QR-CNN-BiLSTM model, refAbstract=null), Reference(id=1218525113176216406, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, doi=null, pmid=null, pmcid=null, year=2023, volume=44, issue=12, pageStart=274, pageEnd=282, url=null, language=null, rfNumber=[27], rfOrder=40, authorNames=饶志, 王科, 谭俊丰, journalName=太阳能学报, refType=null, unstructuredReference=饶志, 王科, 谭俊丰, 等. 基于最优带宽的广东海上风电出力非参数核密度估计与分析[J]. 太阳能学报, 2023, 44(12): 274-282., articleTitle=基于最优带宽的广东海上风电出力非参数核密度估计与分析, refAbstract=null), Reference(id=1218525113285268321, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, doi=null, pmid=null, pmcid=null, year=2023, volume=44, issue=12, pageStart=274, pageEnd=282, url=null, language=null, rfNumber=[27], rfOrder=41, authorNames=Rao Zhi, Wang Ke, Tan Junfeng, journalName=Journal of Solar Energy, refType=null, unstructuredReference=Rao Zhi, Wang Ke, Tan Junfeng, et al. Non-parametric kernel density estimation and analysis of GUANGDONG offshore wind power output based on optimal bandwidth[J]. Journal of Solar Energy, 2023, 44(12): 274-282., articleTitle=Non-parametric kernel density estimation and analysis of GUANGDONG offshore wind power output based on optimal bandwidth, refAbstract=null)], funds=[Fund(id=1218525106968645903, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, awardId=2021YFG0313, language=CN, fundingSource=四川省科技厅项目(2021YFG0313), fundOrder=null, country=null), Fund(id=1218525107157389595, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, awardId=2022YFS0518, language=CN, fundingSource=四川省科技厅项目(2022YFS0518), fundOrder=null, country=null), Fund(id=1218525107295801633, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, awardId=2022ZHCG0035, language=CN, fundingSource=四川省科技厅项目(2022ZHCG0035), fundOrder=null, country=null), Fund(id=1218525107396464939, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, awardId=2019RYY01, language=CN, fundingSource=人工智能四川省重点实验室项目(2019RYY01), fundOrder=null, country=null), Fund(id=1218525107497128245, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, awardId=2021RC12, language=CN, fundingSource=四川轻化工大学人才引进项目(2021RC12), fundOrder=null, country=null), Fund(id=1218525107581014334, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, awardId=2019YYJC02, language=CN, fundingSource=自贡市科技局项目(2019YYJC02), fundOrder=null, country=null), Fund(id=1218525107702649165, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, awardId=2020YGJC16, language=CN, fundingSource=自贡市科技局项目(2020YGJC16), fundOrder=null, country=null), Fund(id=1218525107841061203, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, awardId=Y2023294, language=CN, fundingSource=四川轻化工大学研究生创新基金(Y2023294), fundOrder=null, country=null)], companyList=[AuthorCompany(id=1218525101016929156, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, xref=1, ext=[AuthorCompanyExt(id=1218525101037900677, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, companyId=1218525101016929156, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 School of Automation and Information Engineering, Sichuan University of Science & Engineering, Yibin 644000, China), AuthorCompanyExt(id=1218525101054677894, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, companyId=1218525101016929156, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 四川轻化工大学自动化与信息工程学院, 宜宾 644000)]), AuthorCompany(id=1218525101230838670, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, xref=2, ext=[AuthorCompanyExt(id=1218525101239227279, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, companyId=1218525101230838670, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 Sichuan Key Laboratory of Artificial Intelligence, Yibin 644000, China), AuthorCompanyExt(id=1218525101247615888, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, companyId=1218525101230838670, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 人工智能四川省重点实验室, 宜宾 644000)])], figs=[ArticleFig(id=1218525103806140457, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, language=EN, label=Fig.1, caption=MIC values for each meteorological feature, figureFileSmall=ppZkFiR6SMZVgchmdJLLbw==, figureFileBig=HUHDNLmVfJzPX1KAvhpPOQ==, tableContent=null), ArticleFig(id=1218525103931969582, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, language=CN, label=图1, caption=各气象特征的MIC值, figureFileSmall=ppZkFiR6SMZVgchmdJLLbw==, figureFileBig=HUHDNLmVfJzPX1KAvhpPOQ==, tableContent=null), ArticleFig(id=1218525104074575929, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, language=EN, label=Fig.2, caption=Results of the analysis of the symbolic transfer entropy, figureFileSmall=5J+Tq52RcYqG1su6iTpdRA==, figureFileBig=Kil9/4QD+lzz/4EU/p0sXg==, tableContent=null), ArticleFig(id=1218525104179433538, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, language=CN, label=图2, caption=符号传递熵的分析结果, figureFileSmall=5J+Tq52RcYqG1su6iTpdRA==, figureFileBig=Kil9/4QD+lzz/4EU/p0sXg==, tableContent=null), ArticleFig(id=1218525104288485451, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, language=EN, label=Fig.3, caption=Comparison of PV power in January and July, figureFileSmall=OUnGMe16QtBT9XEp/RRC/g==, figureFileBig=DV8wKJqEMsbzo19U5TU+3w==, tableContent=null), ArticleFig(id=1218525104405925971, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, language=CN, label=图3, caption=1月与7月光伏功率对比, figureFileSmall=OUnGMe16QtBT9XEp/RRC/g==, figureFileBig=DV8wKJqEMsbzo19U5TU+3w==, tableContent=null), ArticleFig(id=1218525104586281054, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, language=EN, label=Fig.4, caption=Variation of contour coefficients, figureFileSmall=sPmU9pYzstQBdX6HBxtSDQ==, figureFileBig=MnCnXI1Ark/SwZ4qA3CsxQ==, tableContent=null), ArticleFig(id=1218525104678555748, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, language=CN, label=图4, caption=轮廓系数的变化, figureFileSmall=sPmU9pYzstQBdX6HBxtSDQ==, figureFileBig=MnCnXI1Ark/SwZ4qA3CsxQ==, tableContent=null), ArticleFig(id=1218525104816967786, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, language=EN, label=Fig.5, caption=PV power after clustering on similar days, figureFileSmall=XZOyPSBxtUgxTKl64yLNpg==, figureFileBig=aDNLH4+TvfzVgt7USD60Rw==, tableContent=null), ArticleFig(id=1218525104934408307, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, language=CN, label=图5, caption=相似日聚类后的光伏发电功率, figureFileSmall=XZOyPSBxtUgxTKl64yLNpg==, figureFileBig=aDNLH4+TvfzVgt7USD60Rw==, tableContent=null), ArticleFig(id=1218525105081208957, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, language=EN, label=Fig.6, caption=Process for interval probability prediction based on TCN-BiGRU-TPA, figureFileSmall=6dVmZAdgUPG0XdCvqX4QFA==, figureFileBig=344CEREX2BE/YrDe1qfjbQ==, tableContent=null), ArticleFig(id=1218525105223815306, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, language=CN, label=图6, caption=基于TCN-BiGRU-TPA的区间概率预测的架构, figureFileSmall=6dVmZAdgUPG0XdCvqX4QFA==, figureFileBig=344CEREX2BE/YrDe1qfjbQ==, tableContent=null), ArticleFig(id=1218525105324478609, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, language=EN, label=Fig.7, caption=Residual structure of TCN, figureFileSmall=PV7EBaA8HimEvqI60E3oWg==, figureFileBig=cRxmFuZmxPmTglfPE8n45Q==, tableContent=null), ArticleFig(id=1218525105441919129, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, language=CN, label=图7, caption=TCN残差结构, figureFileSmall=PV7EBaA8HimEvqI60E3oWg==, figureFileBig=cRxmFuZmxPmTglfPE8n45Q==, tableContent=null), ArticleFig(id=1218525105584525475, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, language=EN, label=Fig.8, caption=Architecture of the TPA mechanism, figureFileSmall=1hGsyUajQCrMDlt75B/Auw==, figureFileBig=C+NeVLdy7ZT8YyiP3ZqlsQ==, tableContent=null), ArticleFig(id=1218525105697771690, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, language=CN, label=图8, caption=TPA机制架构, figureFileSmall=1hGsyUajQCrMDlt75B/Auw==, figureFileBig=C+NeVLdy7ZT8YyiP3ZqlsQ==, tableContent=null), ArticleFig(id=1218525105836183731, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, language=EN, label=Fig.9, caption=Interval prediction results for each weather, figureFileSmall=pVlefmX2jTMVT5N9++phkg==, figureFileBig=WTy8Hzyj9CXHzhDcJabofA==, tableContent=null), ArticleFig(id=1218525105970401468, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, language=CN, label=图9, caption=各天气下区间预测结果, figureFileSmall=pVlefmX2jTMVT5N9++phkg==, figureFileBig=WTy8Hzyj9CXHzhDcJabofA==, tableContent=null), ArticleFig(id=1218525106066870467, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, language=EN, label=Fig.10, caption=Probability density curves for the proposed methods, figureFileSmall=9y7J/75NVeQv9BWfiz43Uw==, figureFileBig=AKcbS9oWKrrz0TAaMJ3zpA==, tableContent=null), ArticleFig(id=1218525106163339465, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, language=CN, label=图10, caption=本文方法的概率密度曲线, figureFileSmall=9y7J/75NVeQv9BWfiz43Uw==, figureFileBig=AKcbS9oWKrrz0TAaMJ3zpA==, tableContent=null), ArticleFig(id=1218525106247225554, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, language=EN, label=Table 1, caption=

Comparison of interval predictions among methods under different weather conditions

, figureFileSmall=null, figureFileBig=null, tableContent=
天气 方法 PCIP PINAW S
方法1 0.871 0.160 1.315
方法2 0.883 0.124 1.568
晴朗天 方法3 0.908 0.137 1.451
方法4 0.925 0.168 1.614
本文方法 0.922 0.134 1.648
方法1 0.849 0.334 1.325
方法2 0.855 0.370 1.295
多云天 方法3 0.883 0.359 1.326
方法4 0.911 0.404 1.332
本文方法 0.907 0.372 1.347
方法1 0.859 0.456 1.235
方法2 0.865 0.482 1.221
阴雨天 方法3 0.883 0.491 1.237
方法4 0.921 0.521 1.250
本文方法 0.916 0.477 1.277
), ArticleFig(id=1218525106381443294, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, language=CN, label=表1, caption=

不同天气下各方法区间预测结果的比较

, figureFileSmall=null, figureFileBig=null, tableContent=
天气 方法 PCIP PINAW S
方法1 0.871 0.160 1.315
方法2 0.883 0.124 1.568
晴朗天 方法3 0.908 0.137 1.451
方法4 0.925 0.168 1.614
本文方法 0.922 0.134 1.648
方法1 0.849 0.334 1.325
方法2 0.855 0.370 1.295
多云天 方法3 0.883 0.359 1.326
方法4 0.911 0.404 1.332
本文方法 0.907 0.372 1.347
方法1 0.859 0.456 1.235
方法2 0.865 0.482 1.221
阴雨天 方法3 0.883 0.491 1.237
方法4 0.921 0.521 1.250
本文方法 0.916 0.477 1.277
), ArticleFig(id=1218525106473717992, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, language=EN, label=Table 2, caption=

Comparison of probabilistic prediction results

, figureFileSmall=null, figureFileBig=null, tableContent=
方法 CRPS
方法1 0.063 6
方法2 0.059 9
方法3 0.057 6
方法4 0.052 9
本文方法 0.050 6
), ArticleFig(id=1218525106586964210, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149780466875720021, language=CN, label=表2, caption=

概率预测结果的比较

, figureFileSmall=null, figureFileBig=null, tableContent=
方法 CRPS
方法1 0.063 6
方法2 0.059 9
方法3 0.057 6
方法4 0.052 9
本文方法 0.050 6
)], attaches=null, journal=Journal(id=1146119176004939786, delFlag=0, nameCn=科学技术与工程, nameEn=Science Technology and Engineering, nameHistory1=null, nameHistory2=null, issn=1671-1815, eissn=, cn=11-4688/T, coden=null, periodic=4, language=CN, oaType=是, ccby=null, superviseOffice=null, ownerOffice=null, pubOffice=null, editorOffice=null, officeType=null, aims=null, clcCode=null, officeProv=null, officeCity=null, officeAddr=null, officeZip=null, officeEmail=null, officePhone=null, editDirector=null, officeDirector=null, officeDirectorPhone=null, officeStaffNum=null, officeEmpNum=null, coverPicUrl=UKU/O7GSka5polgCTkbIIw==, journalPrice=null, startedYear=null, abbrevIsoEn=Sci Technol Eng, journalRemark=null, publicationField=null, createdTime=null, updatedTime=1754445529766, createdBy=null, updatedBy=13701087609, firstLetterCn=S, firstLetterEn=S, subjectCode=Natural Sciences, subjectName=自然科学, subjectCodeEn=Natural Sciences, subjectNameEn=null, picCn=UKU/O7GSka5polgCTkbIIw==, picEn=5hwlULoNwcbj3xUmVi9MAQ==, jcr=null, cjcr=null, exts=[JournalExt(id=1159791870395564357, language=CN, name=科学技术与工程, nameHistory1=null, nameHistory2=null, managedBy=, sponsoredBy=, publishedBy=, editorOffice=, officeProv=null, officeCity=null, officeAddr=, officeZip=, editDirector=null, officeDirector=null, officePhone=null, coverPicUrl=null, journalRemark=, submitArticleUrl=null, websiteUrl=http://www.stae.com.cn/jsygc/home, createdTime=1754445529793, updatedTime=1754445529793, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=http://www.stae.com.cn/jsygc/site/menus/20090429150146001, submissionAuthorUrl=http://www.stae.com.cn/jsygc/author/login, submissionEditorUrl=http://www.stae.com.cn/jsygc/editor/login, submissionReviewUrl=http://www.stae.com.cn/jsygc/reviewer/login, submissionCeEditorUrl=, submissionAeEditorUrl=, option={"copyright":""}), JournalExt(id=1159791870441701702, language=EN, name=Science Technology and Engineering, nameHistory1=null, nameHistory2=null, managedBy=, sponsoredBy=, publishedBy=, editorOffice=, officeProv=null, officeCity=null, officeAddr=, officeZip=, editDirector=null, officeDirector=null, officePhone=null, coverPicUrl=null, journalRemark=, submitArticleUrl=null, websiteUrl=http://www.stae.com.cn/jsygc/home, createdTime=1754445529804, updatedTime=1754445529804, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=, submissionAuthorUrl=http://www.stae.com.cn/jsygc/author/login, submissionEditorUrl=http://www.stae.com.cn/jsygc/editor/login, submissionReviewUrl=http://www.stae.com.cn/jsygc/reviewer/login, submissionCeEditorUrl=, submissionAeEditorUrl=, option={"copyright":""})], databaseList=null, tenantJournalId=1146123166801305609, websiteList=[Website(id=1148243202391400884, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1146123166801305609, journalNameCn=null, journalNameEn=null, grayFlag=null, tenantId=1146029695717560320, platformId=null, journalGroupId=null, journalGroupNameCn=null, journalGroupNameEn=null, type=1, domain=https://castjournals.cast.org.cn/joweb/kxjsygc/CN, language=CN, createTime=1751692112777, createBy=18614031015, updateTime=1753520965431, updateBy=18614031015, name=科学技术与工程-中文站点, tplId=1146099689490845704, title=科学技术与工程, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1148622798802673703, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202391400884, code=articleTextType, value=kx, createTime=1751782615614, updateTime=1751782615614, creator=18614031015, updator=18614031015), WebsiteProps(id=1148622798781702180, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202391400884, code=banner, value=null, createTime=1751782615609, updateTime=1751782615609, creator=18614031015, updator=18614031015), WebsiteProps(id=1148622798769119267, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202391400884, code=logo, value=https://castjournals.cast.org.cn/joweb/kjdb/CN/file/pic?fileId=j86gbwi+p0Idkyl5SzIlmQ==, createTime=1751782615606, updateTime=1751782615606, creator=18614031015, updator=18614031015), WebsiteProps(id=1148622798794285094, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202391400884, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/kjdb/CN/file/pic, createTime=1751782615612, updateTime=1751782615612, creator=18614031015, updator=18614031015), WebsiteProps(id=1148622798790090789, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202391400884, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_cn_619/, createTime=1751782615611, updateTime=1751782615611, creator=18614031015, updator=18614031015)]), Website(id=1155914124811976731, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1146123166801305609, journalNameCn=null, journalNameEn=null, grayFlag=null, tenantId=1146029695717560320, platformId=null, journalGroupId=null, journalGroupNameCn=null, journalGroupNameEn=null, type=1, domain=https://castjournals.cast.org.cn/joweb/kxjsygc/EN, language=EN, createTime=1753521003206, createBy=18614031015, updateTime=1753521003206, updateBy=18614031015, name=科学技术与工程-英文站点, tplId=1146101810881728533, title=Science Technology and Engineering, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1155914371227308235, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155914124811976731, code=articleTextType, value=kx, createTime=1753521061952, updateTime=1753521061952, creator=18614031015, updator=18614031015), WebsiteProps(id=1155914371210531016, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155914124811976731, code=banner, value=null, createTime=1753521061947, updateTime=1753521061947, creator=18614031015, updator=18614031015), WebsiteProps(id=1155914371202142407, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155914124811976731, code=logo, value=https://castjournals.cast.org.cn/joweb/kjdb/CN/file/pic?fileId=j86gbwi+p0Idkyl5SzIlmQ==, createTime=1753521061945, updateTime=1753521061945, creator=18614031015, updator=18614031015), WebsiteProps(id=1155914371223113930, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155914124811976731, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/kjdb/CN/file/pic, createTime=1753521061950, updateTime=1753521061950, creator=18614031015, updator=18614031015), WebsiteProps(id=1155914371218919625, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155914124811976731, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_cn_619/, createTime=1753521061949, updateTime=1753521061949, creator=18614031015, updator=18614031015)])], journalTitle=科学技术与工程, weixinUrl=null, journalUrl=null, iacademicId=null, status=0, seqNo=null, journalTitleEn=Science Technology and Engineering, journalPhotoCn=UKU/O7GSka5polgCTkbIIw==, journalPhotoEn=5hwlULoNwcbj3xUmVi9MAQ==, journalFirstLetter=S, journalRecommend=null, journalNew=null, journalCollection=null, jcrJf=null, cjcrJf=null, jcrJfStr=null, cjcrJfStr=null, submissionFirstDecision=null, sciSubjectClassification=null, casSubjectClassification=null, citeScore=null, totalCitationFrequency=null, icpCode=null, psCode=null, advertisingLicenseCode=null, copyrightInformation=null, country=null, option=null, provinceCode=null, provinceName=null, collectFlag=false), detailUrlCn=https://castjournals.cast.org.cn/joweb/kxjsygc/CN/10.12404/j.issn.1671-1815.2403394, detailUrlEn=https://castjournals.cast.org.cn/joweb/kxjsygc/EN/10.12404/j.issn.1671-1815.2403394, pdfUrlCn=https://castjournals.cast.org.cn/joweb/kxjsygc/CN/PDF/10.12404/j.issn.1671-1815.2403394, pdfUrlEn=https://castjournals.cast.org.cn/joweb/kxjsygc/EN/PDF/10.12404/j.issn.1671-1815.2403394, aliStartDate=null, aliEndDate=null, collectionFlag=false, citedCount=null, citedUrl=null, reference=null)
收藏切换
基于特征挖掘与改进TCN-BiGRU的光伏功率区间概率预测
收藏切换
PDF下载
蔡源 1, 2 , 吴浩 1, 2, * , 唐丹 1, 2
科学技术与工程 | 论文·电工技术 2025,25(10): 4145-4155
收起
收藏切换
科学技术与工程 | 论文·电工技术 2025, 25(10): 4145-4155
基于特征挖掘与改进TCN-BiGRU的光伏功率区间概率预测
全屏
蔡源1, 2 , 吴浩1, 2, * , 唐丹1, 2
作者信息
  • 1 四川轻化工大学自动化与信息工程学院, 宜宾 644000
  • 2 人工智能四川省重点实验室, 宜宾 644000
  • 蔡源(1999—),男,汉族,四川大竹人,硕士研究生。研究方向:光伏输出功率预测。E-mail:

通讯作者:

* 吴浩(1980—),男,汉族,四川成都人,博士,教授。研究方向:电力系统保护与智能控制。E-mail:
Probabilistic Prediction of Photovoltaic Power Interval Based on Feature Mining and Improved TCN-BiGRU
Yuan CAI1, 2 , Hao WU1, 2, * , Dan TANG1, 2
Affiliations
  • 1 School of Automation and Information Engineering, Sichuan University of Science & Engineering, Yibin 644000, China
  • 2 Sichuan Key Laboratory of Artificial Intelligence, Yibin 644000, China
出版时间: 2025-04-08 doi: 10.12404/j.issn.1671-1815.2403394
文章导航
收藏切换

光伏发电在能源领域中具有重要地位。为了准确量化光伏发电功率的不确定性和波动范围,并提高区间预测的综合性能,提出了一种基于特征挖掘与改进TCN-BiGRU的光伏功率区间概率预测方法。首先,利用最大信息系数和符号传递熵因果分析,对气象特征进行筛选,剔除冗余信息,并构造全球水平辐射趋势特征、季节性特征和天气聚类特征以提供更多有效信息。随后,结合时间模式注意力机制和分位数回归方法对TCN-BiGRU模型进行改进,构建组合模型进行区间预测。最后,采用散度度量半极差优化经验带宽选择的核密度估计(kernel density estimation,KDE)方法生成概率预测结果。通过真实光伏电站数据进行分析,验证了所提方法在光伏功率区间概率预测中具有较高的可靠性和适用性。

特征挖掘  /  TCN-BiGRU  /  分位数回归  /  核密度估计  /  区间概率预测

Photovoltaic power generation has an important place in the energy sector. In order to accurately quantify the uncertainty and fluctuation range of PV(photovoltaic) power and to improve the comprehensiveness of interval forecasts, a probabilistic prediction method for PV power intervals based on feature mining with improved TCN-BiGRU was proposed. First, the maximum information coefficient and symbolic transfer entropy causal analysis were utilized to screen the meteorological features, remove redundant information, and construct global horizontal radiation trend features, seasonal features, and weather clustering features to provide more effective information. Subsequently, the TCN-BiGRU model was improved by combining the temporal pattern attention mechanism and quantile regression methods to construct a combined model for interval prediction. Finally, the probabilistic prediction results are generated using the KDE method of empirical bandwidth selection with scatter measure semi-polar optimization. The proposed method is analyzed by real PV plant data, which verifies the high reliability and applicability of the proposed method in PV power interval probability prediction.

feature mining  /  TCN-BiGRU  /  quantile regression  /  kernel density estimation  /  interval probability prediction
蔡源, 吴浩, 唐丹. 基于特征挖掘与改进TCN-BiGRU的光伏功率区间概率预测. 科学技术与工程, 2025 , 25 (10) : 4145 -4155 . DOI: 10.12404/j.issn.1671-1815.2403394
Yuan CAI, Hao WU, Dan TANG. Probabilistic Prediction of Photovoltaic Power Interval Based on Feature Mining and Improved TCN-BiGRU[J]. Science Technology and Engineering, 2025 , 25 (10) : 4145 -4155 . DOI: 10.12404/j.issn.1671-1815.2403394
随着全球各国对可持续能源的需求日益增长,光伏发电作为一种清洁、可再生的能源技术,其在全球能源结构中的比重持续上升[1]。尤其在中国提出“双碳目标”以促进绿色经济发展后,截至2023年6月底,光伏发电装机容量占中国可再生能源累计装机容量的35.6%,可见,光伏发电在中国也具有十分优秀的应用前景[2]。然而,光伏输出功率受多种环境因素的影响,如辐照度、温度和云层覆盖等,这导致光伏发电量具有高度的不确定性和波动性。当光伏系统大规模并网时,这些波动性和不确定性会对电网的稳定性和运行效率产生显著影响[3]。在此背景下,对光伏功率进行预测变得尤为关键。国外光伏产业发展较早,故拥有丰富的技术成果和经验。例如,太阳能仿真气象数据库Meteonorm以及光功率预测系统Meteodyn Solar广泛应用于光伏工程项目。虽然国内的光伏功率预测研究起步较晚,但近年来在地理优势和政策支持的推动下,也取得了显著进展。例如,国能日新研发的功率预测系统SPSF-3000已经成功应用于实际电力系统,并且在功率预测相关机构中,其市占率最高,展现了其在光伏功率预测方面的实力。
目前,中外的大部分研究集中于光伏输出功率的确定性预测[4]。这类预测致力于提供确切的预测值,对电网的日常运营管理和短期决策制定起到了至关重要的作用。然而,鉴于光伏发电固有的波动性特征,仅依赖确定性预测可能无法全面揭示预测结果中潜在的不确定性和相关风险[5]。而区间概率预测则弥补了确定性预测的不足,通过提供预测值的波动范围以及概率密度分布,来反映功率的不确定性和变化幅度,使区间内包含真实值的概率达到设定的置信水平,而不仅仅是提供具体的预测值,这对于制定更加灵活和鲁棒的电网管理策略具有重要意义[6]
当前的光伏功率区间概率预测主要依赖于人工智能方法与非参数估计法的融合[7],文献[8]使用智能优化算法寻优长短期记忆网络(long short term memory, LSTM)模型参数进行光伏功率点预测,并基于Bootstrap分析预测结果及误差分布,得到光伏功率各个置信度的区间分布情况。实验表明,Bootstrap方法避免了传统参数估计法需预设分布形式的缺点,更准确地适应了功率的分布特性。但Bootstrap需进行反复抽样,因而存在较高的计算量。文献[9]对相似日进行聚类,并利用改进的互补集合经验模态分解与变分模态分解对光伏功率序列进行二次分解,最后同样使用智能优化算法优化双向长短期记忆网络(bidirectional long short term memory, BiLSTM)的超参数,并添加注意力机制以提高点预测结果,最后基于核密度估计进行区间预测。该方法证明了相似日聚类提升了BiLSTM的计算效率,且注意力机制对预测精度的提升具有积极效果。文献[10]结合相似日聚类、分位数回归神经网络(quantile regression neural network,QRNN)和KDE进行光伏功率区间概率预测,并弥补了分位数交叉的缺陷,取得了良好的区间概率预测结果。文献[11]构建了分位数回归时间卷积网络(quantile regression of temporal convolution network,QRTCN)来输出不同条件分位数的预测值,并利用KDE对电力负荷进行概率预测,最后实验对比验证了QRTCN模型的优越性。文献[12]进行了炭价格区间预测研究,并在考虑影响因素时首次引入了符号传递熵(symbolic transfer entropy,STE),分析了碳价格序列与特征序列之间的因果联系。同样在光伏功率预测时,大部分文献只进行了相关性分析,而未考虑因果关系,所以在影响因素分析时加入STE具有较强的应用潜力。文献[13]将时间卷积网络(temporal convolution network,TCN)与双向门控循环单元(bidirectional gate recurrent unit,BiGRU)结合形成TCN-BiGRU模型,实现了对溶解氧的预测,并提出了构造新的输入特征,经实验验证,该方法有效提高了输入特征的质量和预测精度。文献[14]通过优化的去噪技术和模式分解增强数据质量,同时基于信息熵改进的门控循环单元(gated recurrent unit,GRU)和非线性分位数回归(quantile regression,QR)损失函数进行风电功率区间预测。尽管模型在预测准确性方面取得了显著进步,但未能提供关于概率密度的预测,仍需在未来的研究中进一步完善。
鉴于上述文献分析中揭示的优势及不足,本文研究提出一种基于特征挖掘与改进TCN-BiGRU的光伏功率区间概率预测方法。首先,将最大信息系数和STE因果分析结合,对各气象特征筛选,剔除冗余特征,保证输入特征的强关联;然后,构造全球水平辐射趋势特征、季节特征和天气聚类特征,帮助预测模型捕捉更多有效信息,提高模型的鲁棒性;接着,结合时间模式注意力机制(temporal pattern attention,TPA)和QR方法改进TCN-BiGRU模型,加强对关键时间模式的关注,构建QR-TCN-BiGRU-TPA的区间预测模型;最后,利用散度度量半极差优化经验带宽选择的KDE方法,降低拟合估计误差,得到概率密度预测结果。
在光伏发电功率预测研究中,特征挖掘主要是指对输入特征进行分析与构造。特征分析主要是识别影响功率变化的关键因素,如光照强度、温度等。特征构造则是基于原始数据通过数学和统计方法创造新的特征变量,旨在更全面地捕捉影响光伏发电性能的复杂动态[15]
使用澳大利亚Alice Springs某光伏电站的数据,数据每隔15 min记录一次。其中气象因素包括风速(wind speed,WS)、风向(wind direction,WD)、温度(temperature,T)、湿度(humidity,H)、全球水平辐射(global horizontal radiation,GHR)、弥散水平辐射(diffuse horizontal radiation,DHR)、降雨量(rainfall,R)。通过对这些气象特征进行相关性分析和因果分析,选出最优气象特征集。然后结合特征分析的结果,构造新的输入特征来提供更多的有效信息,帮助预测模型更好地理解数据结构和内在联系,提高其在不同运行条件下的适应性和准确性。
最大信息系数(maximum information coefficient,MIC)是一种基于互信息的特征选择方法,其主要用途是量化两个变量之间的关联强度[16]。MIC不仅能够识别变量间的线性关系,而且能够捕捉非线性以及其他更复杂的关系形式,这种普适性使其适用于探索数据之间存在的多样关联性。MIC值的范围在0~1,数值越接近1表示两个变量之间的相关性越强。
在计算气象特征与光伏发电功率的MIC时,首先设各气象特征序列为X,光伏发电功率序列为Y
进行数据清洗后,将XY数据对齐,保证时间序列一致性。随后将变量XY的值域分别划分成多种网格,网格数从1到B,其中B是数据量n的函数,通常取数据量的0.6次方。
接着,对每个网格计算联合概率分布以及边缘概率分布,从而得到每个网格计算变量XY的互信息I[X;Y]。
最后,归一化互信息,并遍历所有网格分割方案,选择互信息最大的那个值作为 MIC。具体计算公式为
MIC(X,Y)= m a x X Y < B I [ X ; Y ] l o g 2 [ m i n ( X , Y ) ]
式(1)中: X Y为网格中沿XY方向的格子数。
对各气象因素与光伏发电功率进行MIC分析后,得到各气象特征的MIC值,如图1所示。由图1可以看出气象特征中的全球水平辐射和弥散水平辐射对光电功率的影响最为显著。风速和相对湿度也有一定的相关性,但远低于辐射量。而温度、风向和降雨量的MIC值接近于0,呈现出较低的相关性。故继续分析特征序列和光电功率之间的因果联系,两者结合选出最终的输入特征集。
符号传递熵基于传递熵的概念,通过使用符号方法简化了计算过程,使其适合分析实际应用中的大规模或复杂时间序列数据。它通过衡量时间序列数据中两个系统之间的单向动态信息流,即信息的传递方向,以揭示两个时间序列之间的因果关系[17]
进行因果分析时,设各气象特征序列为X,光伏发电功率序列为Y。首先,将原始时间序列数据先对元素排序,再对位置索引排序,即可得到转换后的符号序列。
然后,对于每个时间点t,计算不同符号向量出现的概率分布,以及在给定一个系统的过去状态时另一个系统未来状态的条件概率分布。
最后,得到从XY的符号传递熵为
TX,Y= y t + δ , y t , x t np y t + δ , y t , x tln p y t + δ y t , x t p y t + δ y t
式(2)中:TX,Y为序列XY的传递熵;n为序列长度;δ为时间步长;xtyt分别为序列XY的历史观测值;p(yt+δ,yt,xt)为联合概率分布函数;p(yt+δ y t , x t)和p(yt+δ y t)为条件概率分布函数。
对各气象特征进行基于STE的因果分析后,进行随机化检验以保证因果关系的有效性,得到各气象特征的STE值,如图2所示。random为随机化检验的STE值,其趋近于0,验证了因果关系的有效性。在考虑的气象变量中,温度、全球水平辐射和弥散水平辐射对光伏功率的状态变化具有相对更高的因果关联,风速也有一定的影响,风向和降雨量的STE仍和MIC分析结果一致,影响十分微弱。结合分析得出结论,风向和降雨量的影响则几乎可以忽略不计,所以不将它们作为输入特征。
基于GHR与光伏功率的高度关联性,构建GHR趋势变化的特征可以帮助模型更好地理解功率变化趋势。通过计算待预测时刻的前n个时刻GHR的变化量反映出辐射量随时间的增减趋势。计算公式为
ΔI= i = 1 n(Gt-i+1+Gt-i)
式(3)中:ΔI为待预测时刻之前n个时刻的GHR累积变化量;Gt-i为在t时刻之前第i个时刻的GHR值;n为选择用于计算辐射趋势的时间点数量。
光伏发电功率因季节而异,澳大利亚位于南半球,冬夏季时间相反,图3画出了历史数据某年中典型的1月和7月每日的光伏功率平均值变化曲线。由图3可知,1月平均功率值较高,反映了夏季较强的太阳辐射和更长的日照时间,但是波动天数较多,主要是由于夏季的降雨量较多,导致存在有几天的平均功率不高。相比之下,7月的平均功率值较低,符合冬季较弱的太阳辐射和较短的日照时间。此外,7月的功率波动较少,表明冬季气候相对稳定,少有极端天气影响太阳能发电效率。
基于功率的季节差异性,构造季节特征可以提高模型对光伏功率季节性变化的理解能力,进而提高预测的准确性。引入儒略日特征[18]来提供连续变化的季节信号,因为冬天最冷的时间在7月中旬左右,所以设置第195天的儒略日值置为-1,则冬季的儒略日值接近-1,夏季时接近1。每天的儒略日值计算式为
D=-cos 2 π ( J - 195 ) 365
式(4)中:J为儒略日;D为儒略日值。
在不同天气条件下,光伏功率呈现的变化趋势不同,这主要是源于太阳辐射的显著差异,从而影响了光伏板的功率输出。大量研究表明,在光伏功率预测领域中,进行相似天气聚类能够使模型更准确地理解不同天气对光伏发电功率的影响规律,进而提升预测的准确性。
结合特征分析的结果,对历史数据中的GHR序列以日为单位进行聚类分析。本文研究采用文献[19]中的改进K-Mediods聚类方法,其中用余弦相似度替换原算法中的欧氏距离度量指标。然而,K-Mediods算法的中心点的选择对聚类效果的影响仍很大,传统的随机初始化方法容易陷入局部最优解。因此本文借鉴K-means++[20]的思想,改进了聚类中心的初始化方法。先随机选择第一个medoid,对于接下来的每个medoid的选择,计算每个数据点到已选medoids的最小距离。根据这些距离,为每个点计算一个概率,其中距离越大的点被选为新medoid的概率也越大。使用累积概率分布来随机选择下一个medoid。这样可以尽量分散初始中心点的位置,从而缓解由于随机初始化导致的聚类质量不稳定问题。
为了确定聚类个数,选取轮廓系数[21]作为聚类评价指标,随着聚类个数增长的轮廓系数变化如图4所示。从图4中可以观察到,当聚类个数大于3后,随着聚类数量的增加,轮廓系数逐渐降低,聚类个数为3时的轮廓系数最高,并在聚类数超过5个之后出现了显著的下降。为了确保不同聚类间的差异性和同类样本间的相似性最大,选择聚类数为3。经过以天为单位的聚类处理后,不同类别下的历史光伏发电功率如图5所示。
根据图5分析可知,最左边图中的各光伏功率曲线显示出较高的一致性,且天气状况稳定,波动极少,通常对应于晴朗天气条件;中间图中的各曲线总体呈先上升后下降的趋势,但可能因云层覆盖变化而导致功率产出波动频繁,通常对应着多云天气类型;最右侧图中的功率波动显著,且整体功率较低,这表明在此聚类中的日子,光伏系统受到了极端恶劣天气的影响,通常对应着阴雨天气类型。基于以上的聚类结果,采用独热编码构造天气类型特征,分别建立各天气类别的标签。
本文研究结合TCN、BiGRU、TPA、QR和KDE方法提出了区间概率预测模型,主要架构如图6所示。首先,模型基于TCN架构,引入BiGRU层以增强对时间序列动态变化的捕捉能力。随后,加入TPA强化在预测过程中对关键时间模式的关注,通过对序列各部分赋予不同权重,增强对重要时间点的感知能力。接着,利用QR损失函数训练模型,生成不同条件下的分位数,通过取不同分位数水平得到不同置信区间的区间预测结果。最后,应用KDE方法对预测得到的分位数进行概率密度估计,提供概率预测结果。
TCN通过其独特的扩张卷积层提供了广泛的感受野,卷积层的扩张因子以2的幂次增加,并通过因果卷积仅使用历史数据,避免未来信息泄露[22]。此外,TCN的残差连接有助于解决深层网络中的梯度消失问题,使得模型能够训练更深的网络层,残差结构如图7所示。
经过时间卷积层处理后的特征被送入BiGRU层,利用其双向结构来从序列的两个方向捕获信息。BiGRU的门控机制控制着信息的流动,确保模型在记忆长期依赖时还能保持稳定性和高效性[23]
接下来,利用TPA机制进一步处理BIGRU的输出,其结构如图8所示。TPA通过调整不同时间点的特征权重来提高预测的准确性,相较于传统注意力仅对时间步进行加权具有明显优势[24]
对于输入TPA的隐藏特征向量h,它们共同构成了隐藏状态矩阵H,然后使用滤波器C对每一行进行一维卷积*得到时间模式矩阵HC
H i , j C= l = 1 w[Hi,(t-w-1+l)*Cj,T-w+l]
式(5)中: H i , j C为第i个行向量和第j个卷积核作用的时间模式矩阵;l为卷积中的填充长度;Hi,(t-w-1+l)为第i个行向量和第(t-w-1+l)个卷积核的隐藏状态矩阵;T为滤波器提取权重的最大长度,一般为w;Cj,T-w+l为第j个滤波器提取的特征。
通过时间模式矩阵HC的每行与t时刻的隐藏特征向量ht进行运算得到变量权重,进行归一化后得到注意力权重ai,具体计算公式为
f ( H i C , h t ) = ( H i C ) T W a h t a i = s i g m o i d [ f ( H i C , h t ) ]
式(6)中:Wa为权重参数; H i C为时间模式矩阵的第i个行向量;f(·)为评估函数;sigmoid为激活函数。
aiHC进行权重求和,得到权重向量vt,即
vt= i = 1 mai H i C
式(7)中:m为特征向量维数。
将权重向量vt和时刻t的隐藏状态ht线性映射求和,得到综合时序特征状态 h ˙ t
h ˙ t=Wtht+Wvvt
式(8)中:WtWv为对应变量的权重矩阵。
分位数回归是一种用于预测条件分位数的统计方法,通过取0~1的分位数,寻找分位数回归系数的最优估计值,以此来估计任意条件分位数下的目标变量的预测值[25],公式为
QY(τ X)=Xβ(τ)
式(9)中:QY(τ X)为条件分位数τ下的输出;XY分别为输入变量和输出变量;τ为分位数;β(τ)为分位数回归系数。
在基于机器学习的点预测模型训练中,通常将均方根误差作为损失函数以得到最优点预测结果。同理,将预测模型与最小化QR损失函数结合,就可以得到在不同分位数下的最优条件分位数估计结果。根据分位数回归理论,可以得到目标损失函数L
L τ [ y , Q ^ y ( τ ) ] = i = 1 n ( τ - I u ) u u = y i - Q ^ y i ( τ )
式(10)中:i为样本索引; Q ^ y i(τ)为第i个样本的条件分位数τ下的预测值;yi为第i个样本的真实值;I为指示函数,指示了当预测值与真实值的大小关系不同时,损失函数的计算式不同。
I(u)= 0 , u 0 1 , u < 0
通过分位数τ的不同取值,使模型学习到不同分位数条件下输入与输出的复杂映射关系,从而得到各个分位数条件下的最优预测值。
对于概率密度函数的计算,采用非参数估计方法KDE。相较于参数方法,核密度估计不需要对数据分布进行假设,因此在处理复杂数据分布时具有显著优势[26]
将得到的光伏功率某样本点的各分位数预测结果组合为输入向量,利用核密度估计得到其概率密度函数,具体计算式为
g(y)= 1 q b i = 1 qKg y - Q ^ i y b
式(12)中:q为分位数个数;b为带宽; Q ^ i yy的条件分位数个数为i时的预测值;Kg的核函数为高斯核函数。
Kg(z)= 2 2 πexp - z 2 2
式(13)中:z为随机变量。
大多数研究表明,带宽b的选择对拟合效果有着更为显著的影响,为了选择最优带宽,利用散度度量半极差优化经验带宽选择的方法[27],具体计算式为
σ ˙ = m i n σ , I [ φ - 1 ( 0.75 ) - φ - 1 ( 0.25 ) ] b = 1.06 m i n σ ˙ , I 1.34 n - 1 5
式(14)中:σ为标准差; σ ˙为优化后的标准差;φ为标准正态分布函数;n为样本数量;I为散度度量半极差。
本文研究使用澳大利亚Alice Springs某光伏电站的2014—2015年的历史数据为例进行分析,其中包括气象数据和光伏功率数据,功率等级为千瓦,数据每隔15 min记录一次。由于夜间没有光伏出力,故仅保留了每天07:00—19:00的数据。对于数据中的异常值和缺失值进行检测,然后采用均值插补法进行处理,最后进行归一化处理。完成所有预处理工作后,将历史数据按照7∶1∶2的比例分为训练集、验证集和测试集,从区间预测和概率预测方面验证模型有效性。
本文模型在Pytorch框架下进行构建,模型首先定义了分位数列表quantiles的取值范围为[0.05,0.95],以0.05为间隔,用来输出各个分位数下的最优估计值。训练时采用早停机制,设置最大训练轮数为50,如验证集的损失连续5次不增长,则保留最佳模型进行测试。并通过多次试验分析,利用控制变量方法确定了模型的最佳关键参数。其中,残差块TemporalBlock为3,每个TemporalBlock中卷积核的大小kernel_size为3。且TemporalBlock层的扩张因子为2的指数倍增加,即每个TemporalBlock的扩张因子依次为1、2、4。BiGRU层的层数为1,BiGRU层的隐藏单元数量gru_hidden_size为64。学习率为0.001,优化器使用Adam,Dropout为0.2,用于减少过拟合。
为了验证方法的有效性,加入了4种方法与本文方法进行对比,关键参数设置与本文方法相同。对比方法包括:未经过数据分析与构造、未加入BiGRU层和TPA机制的方法1,即QR-TCN;在方法一的基础上加入BiGRU层的方法2,即QR-TCN-BiGRU;在方法二的基础上加入TPA机制的方法3,即未经数据分析与构造的QR-TCN-BiGRU-TPA;在方法3的基础上加入特征分析但未加特征构造的方法4,即加入特征分析的QR-TCN-BiGRU-TPA。
得到区间预测的结果后,使用误差评价指标来量化预测结果的不确定性,才能更灵活地制定决策。本文研究使用常用的预测区间覆盖率(prediction interval coverage probability,PICP)、预测区间平均宽度(prediction interval normalized average,PINAW)以及综合评价指标S来评价区间预测结果。计算公式为
PCIP= 1 N n = 1 Nkn
PINAW= 1 N D n = 1 Nupn-lown
S= 2 P C I P × A P C I P + A
式中:N为样本数量;kn为真实值位于预测区间内部或外部的指示值,当真实值位于预测区间内时kn为1,否则为0;D为真实值的最大值减最小值;up和low为预测区间的上、下界;A为PINAW的倒数;S越大则区间预测效果越好。
使用连续分级概率评分(continuous ranked probability score,CRPS)评估概率预测结果,衡量真实值y与概率分布预测F之间的差异。计算公式为
CRPS[F(y'),y]= y [F(y')-1(y'≥y)]2dy
式(18)中:当y'y时取值1,否则为0。CRPS的值越小,则概率预测效果越好。
为了验证所提方法的区间预测有效性,本文将前述的4种方法与本文方法在90%置信区间的测试结果进行对比,取测试集中3类天气下的平均区间预测结果如表1所示。
表1可知,晴朗天气下的各方法的评价指标相对较好,其中方法4的区间覆盖率PCIP指标表现最好,为0.925,但方法4的平均区间宽度PINAW较高,而本文方法的PINAW则为0.134,虽不如方法2的PINAW窄,但是在区间覆盖率方面高于方法2,且本文方法的综合评价指标S也为最高,为1.648。对比各项评价指标,本文方法的PCIP比方法1提高了5.9%,比方法4的PINAW降低了20.2%,比方法1的S提高了25.3%。在多云天气下,各方法的PCIP普遍下降,本文方法的PCIP达到置信度要求,PCIP相比方法一提高了6.8%,PINAW比方法4降低了8.9%,S仍为最高。同样,在阴雨天气下,本文方法在PCIP和PINAW上仍处于较优的位置,PCIP相比方法一提高了6.6%,PINAW比方法4降低了8.4%,S仍为最高。综合分析可知,对比QR-TCN模型,引入BiGRU层和TPA机制能够有效提高模型的区间覆盖率。虽然本文方法相较于未加征构造的方法4的PCIP更低,但差距不大。在区间平均宽度PINAW指标上,本文方法的PINAW相较于方法4更低,表明在加入特征构造后,模型牺牲了一定的的区间覆盖率,但保证了区间平均宽度更窄,以此来获得更优的综合性能。在各类天气下,本文方法与其他方法相比,本文方法的综合评价指标S均更高,表明了本文方法在综合考虑区间覆盖率和区间宽度方面取得了平衡,满足了90%的置信水平,且减少了区间宽度,相比其他方法具有更好的综合性能。
为了更加直观地对比分析本文方法在不同天气聚类下的区间预测结果,在测试集中的各天气类型中随机选取1 d进行可视化分析,如图9所示。
图9可知,在晴朗天气下,各模型的覆盖率较好,且预测的区间宽度较窄,然而,在多云天和阴雨天情况下,由于功率波动较大,导致生成的区间宽度相对较宽。综合分析,本文方法在不同天气类型下的区间预测结果,满足了置信度要求的区间覆盖率,且具有较小的区间宽度,预测的综合性能较高。
对于概率预测,通过各模型测试集中的预测分位数作为KDE的输入,进行概率密度函数预测后,计算各模型的平均CRPS,具体结果如表2所示。
通过表2可以观察到,本文方法相对于其他4种方法具有最低的CRPS值,突显了本文方法在概率预测方面的高可靠性。具体而言,本文方法相较于方法1的CRPS降低了20.4%,这源于本文方法在方法一的基础上引入了BiGRU层、注意力机制、以及特征分析与构造的技术,加强了模型对特征的学习能力,使得模型能够更有效地考虑特征信息,进而在一定程度上提升了概率预测的性能。同样,为了直观展示本文方法在3类天气下的概率预测结果,绘制了测试集中3 d不同类别天气的12:00的光伏功率概率密度函数,如图10所示。
图10可知,在12:00时刻下,光伏功率在不同天气条件下的实际值大致分布在概率密度曲线的峰值附近,且曲线形状平滑,表明预测模型具有一定的稳定性。通过观察实际值偏离曲线峰值的情况,可以分析预测误差,并结合区间预测结果,以制定更为全面和合理的调度计划。
针对确定性预测无法全面揭示预测结果中潜在的不确定性和相关风险,同时为提高区间概率预测的稳定性,研究了基于特征挖掘与改进TCN-BiGRU的光伏功率区间概率预测模型,以真实电站的数据进行了算例分析与验证,得出以下结论。
(1)对气象特征进行MIC分析和STE因果分析,能够优化气象特征集,筛除冗余特征。构造的各类输入特征提供了更多的有效信息,帮助预测模型更好地理解数据结构和内在联系,提高了模型在各类条件下预测的适应性和准确性。
(2)在TCN-BiGRU模型的基础上加入时间模式注意力机制,通过对序列各部分赋予不同权重,能够强化在预测过程中对关键时间模式的关注,增强区间概率预测结果的可靠性。
(3)在不同天气条件下的区间预测,本文模型均能在区间覆盖率和区间宽度方面取得较好的平衡。同时,利用CRPS指标验证了本文方法的概率预测的可靠性,KDE给出了较好的概率密度分布。本文模型在不同天气类别下能够满足较高的区间覆盖率、较窄的区间宽度以及较低的CRPS指标,证明了模型在不同天气下的泛化能力和适应性。
本文研究表明模型能够对光伏输出功率进行可靠的区间概率预测,但后续仍需采用不同量级或不同天气条件的数据集进行多方面检验,以验证模型的泛化性。
  • 四川省科技厅项目(2021YFG0313)
  • 四川省科技厅项目(2022YFS0518)
  • 四川省科技厅项目(2022ZHCG0035)
  • 人工智能四川省重点实验室项目(2019RYY01)
  • 四川轻化工大学人才引进项目(2021RC12)
  • 自贡市科技局项目(2019YYJC02)
  • 自贡市科技局项目(2020YGJC16)
  • 四川轻化工大学研究生创新基金(Y2023294)
参考文献 引证文献
排序方式:
[1]
龙勇, 宋敏. 全球能源安全大变局下保障我国能源安全的思路与方略[J]. 改革, 2023(10): 74-83.
Long Yong, Song Min. Ideas and strategies for ensuring China's energy security in the context of global energy security upheaval[J]. Reform, 2023(10): 74-83.
[2]
国家能源局. 我国可再生能源发电总装机突破13亿千瓦[J]. 电力勘测设计, 2023(7): 71.
National Energy Agency. China's renewable energy power generation total installed capacity exceeds 1.3 billion kilowatts[J]. Electric Power Survey and Design, 2023(7): 71.
[3]
张弛, 朱宗玖. 基于误差修正的短期光伏功率预测[J]. 科学技术与工程, 2023, 23(27): 11664-11672.
Zhang Chi, Zhu Zongjiu. Short-term PV power prediction based on error correction[J]. Science Technology and Engineering, 2023, 23(27): 11664-11672.
[4]
Antonanzas J, Osorio N, Escobar R, et al. Review of photovoltaic power forecasting[J]. Solar Energy, 2016, 136: 78-111.
[5]
Iheanetu K J. Solar photovoltaic power forecasting: a review[J]. Sustainability, 2022, 14(24). DOI: 10.3390/su142417005.
[6]
刘运超, 杨宁, 崔承刚, 等. 基于时空信息的区域内光伏电站功率预测[J]. 科学技术与工程, 2023, 23(34): 14596-14602.
Liu Yunchao, Yang Ning, Cui Chenggang, et al. Power prediction of regional photovoltaic plants based on spatio-temporal information[J]. Science Technology and Engineering, 2023, 23(34): 14596-14602.
[7]
Fan H, Zhen Z, Liu N, et al. Fluctuation pattern recognition based ultra-short-term wind power probabilistic forecasting method[J]. Energy, 2023, 266.DOI: 10.1016/j.energy.2022.126420.
[8]
宋绍剑, 罗世坚, 李国进, 等. 基于MPA-LSTM模型和Bootstrap方法的短期光伏功率区间预测[J]. 广西大学学报(自然科学版), 2022, 47(4): 986-997.
Song Shaojian, Luo Shijian, Li Guojin, et al. A short-term PV power interval prediction based on MPA-LSTM network model and Bootstrap method[J]. Journal of Guangxi University(Natural Science Edition), 2022, 47(4): 986-997.
[9]
Yu M, Niu D, Wang K, et al. Short-term photovoltaic power point-interval forecasting based on double-layer decomposition and WOA-BiLSTM-Attention and considering weather classification[J]. Energy, 2023, 275.DOI: 10.1016/j.energy.2023.127348.
[10]
赵耀, 高少炜, 李东东, 等. 基于天气相似聚类与QRNN的短期光伏功率区间概率预测[J]. 电力系统自动化, 2023, 47(23): 152-161.
Zhao Yao, Gao Shaowei, Li Dongdong, et al. Short-term interval probability prediction of photovoltaic power based on weather similarity clustering and quantile regression neural network[J]. Power System Automation, 2023, 47(23): 152-161.
[11]
庞昊, 高金峰, 杜耀恒. 基于时间卷积网络分位数回归的短期负荷概率密度预测方法[J]. 电网技术, 2020, 44(4): 1343-1350.
Pang Hao, Gao Jinfeng, Du Yaoheng. A short-term load probability density prediction based on quantile regression of time convolution network[J]. Grid Technology, 2020, 44(4): 1343-1350.
[12]
Wang J, He M, Jiang W. Causal carbon price interval prediction using lower upper bound estimation combined with asymmetric multi-objective evolutionary algorithm and long short-term memory[J]. Expert Systems with Applications, 2024, 236: 121286.
[13]
张铮, 高森, 张泽扬. 基于特征构造预处理与TCN-BiGRU的池塘溶解氧预测模型[J]. 上海海洋大学学报, 2023, 32(5): 943-952.
Zhang Zheng, Gao Sen, Zhang Zeyang. Prediction model of dissolved oxygen in pond based on feature construction pretreatment and TCN-BiGRU[J]. Journal of Shanghai Ocean University, 2023, 32(5): 943-952.
[14]
Liu T, Qi S, Qiao X, et al. A hybrid short-term wind power point-interval prediction model based on combination of improved preprocessing methods and entropy weighted GRU quantile regression network[J]. Energy, 2024, 288.DOI: 10.1016/j.energy.2023.129904.
[15]
欧阳卫年, 赵紫昱, 陈渊睿. 自样本特征构造的1DCNN-BiLSTM网侧光伏功率预测[J]. 电力系统及其自动化学报, 2024, 36(3): 151-158.
Ouyang Weinian, Zhao Ziyu, Chen Yuanrui. 1DCNN-BiLSTM method of grid-side photovoltaic power prediction with self-sampled feature construction[J]. Journal of Power System and Automation, 2024, 36(3): 151-158.
[16]
王旭光, 张可, 李潇, 等. 基于变量相关注意力机制的短期风速预测[J]. 太阳能学报, 2023, 44(8): 467-476.
Wang Xuguang, Zhang Ke, Li Xiao, et al. Short-term wind speed forecasting based on variable correlation attention mechanism[J]. Journal of Solar Energy, 2023, 44(8): 467-476.
[17]
Staniek M, Lehnertz K. Symbolic transfer entropy[J]. Physical Review Letters, 2008, 100(15). DOI: 10.1103/PhysRevLett.100.158101.
[18]
McKendry I G. Evaluation of artificial neural networks for fine particulate pollution (PM10 and PM2.5) forecasting[J]. Journal of the Air & Waste Management Association, 2002, 52(9): 1096-1101.
[19]
季玉琦, 严亚帮, 和萍, 等. 基于K-Medoids聚类与栅格法提取负荷曲线特征的CNN-LSTM短期负荷预测[J]. 电力系统保护与控制, 2023, 51(18): 81-93.
Ji Yuqi, Yan Yabang, He Ping, et al. CNN-LSTM short-term load forecasting basedon the K-Medoids clustering and grid method to extract load curve features[J]. Power System Protection and Control, 2023, 51(18): 81-93.
[20]
Kapoor A, Singhal A. A comparative study of K-Means, K-Means++ and Fuzzy C-Means clustering algorithms[C]// 3rd International Conference on Computational Intelligence & Communication Technology (CICT). New York: IEEE, 2017: 1-6.
[21]
吴红斌, 何叶, 赵波, 等. 基于改进K-means聚类算法的风电场动态等值[J]. 太阳能学报, 2018, 39(11): 3232-3238.
Wu Hongbin, He Ye, Zhao Bo, et al. Research on dynamic equivalent of wind farm based onimproved K-means clustering algorithm[J]. Journal of Solar Energy, 2018, 39(11): 3232-3238.
[22]
Hewage P, Behera A, Trovati M, et al. Temporal convolutional neural (TCN) network for an effective weather forecasting using time-series data from the local weather station[J]. Soft Computing, 2020, 24: 16453-16482.
[23]
Lin X, Quan Z, Wang Z J, et al. A novel molecular representation with BiGRU neural networks for learning atom[J]. Briefings in Bioinformatics, 2020, 21(6): 2099-2111.
[24]
Shih S Y, Sun F K, Lee H. Temporal pattern attention for multivariate time series forecasting[J]. Machine Learning, 2019, 108: 1421-1441.
[25]
王晓东, 鞠邦国, 刘颖明, 等. 基于QR-NFGLSTM与核密度估计的风电功率概率预测[J]. 太阳能学报, 2022, 43(2): 479-485.
Wang Xiaodong, Ju Bangguo, Liu Yingming, et al. Probability prediction of wind power based on QR-NFGLSTM and kernel density estimation[J]. Journal of Solar Energy, 2022, 43(2): 479-485.
[26]
王开艳, 杜浩东, 贾嵘, 等. 基于相似日聚类和QR-CNN-BiLSTM模型的光伏功率短期区间概率预测[J]. 高电压技术, 2022, 48(11): 4372-4388.
Wang Kaiyan, Du Haodong, Jia Rong, et al. Short-term interval probability prediction of photovoltaic power based on similar daily clustering and QR-CNN-BiLSTM model[J]. High Voltage Technology, 2022, 48(11): 4372-4388.
[27]
饶志, 王科, 谭俊丰, 等. 基于最优带宽的广东海上风电出力非参数核密度估计与分析[J]. 太阳能学报, 2023, 44(12): 274-282.
Rao Zhi, Wang Ke, Tan Junfeng, et al. Non-parametric kernel density estimation and analysis of GUANGDONG offshore wind power output based on optimal bandwidth[J]. Journal of Solar Energy, 2023, 44(12): 274-282.
2025年第25卷第10期
PDF下载
349
148
引用本文
BibTeX
文章信息
doi: 10.12404/j.issn.1671-1815.2403394
  • 接收时间:2024-05-08
  • 首发时间:2025-07-09
  • 出版时间:2025-04-08
补充材料
相关文章
文章信息
作者
出版历史
  • 收稿日期:2024-05-08
  • 修回日期:2025-01-10
基金
四川省科技厅项目(2021YFG0313)
四川省科技厅项目(2022YFS0518)
四川省科技厅项目(2022ZHCG0035)
人工智能四川省重点实验室项目(2019RYY01)
四川轻化工大学人才引进项目(2021RC12)
自贡市科技局项目(2019YYJC02)
自贡市科技局项目(2020YGJC16)
四川轻化工大学研究生创新基金(Y2023294)
作者信息
    1 四川轻化工大学自动化与信息工程学院, 宜宾 644000
    2 人工智能四川省重点实验室, 宜宾 644000

通讯作者:

* 吴浩(1980—),男,汉族,四川成都人,博士,教授。研究方向:电力系统保护与智能控制。E-mail:
参考文献
分享链接
https://castjournals.cast.org.cn/joweb/kxjsygc/CN/10.12404/j.issn.1671-1815.2403394
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