Article(id=1261262693726544554, tenantId=1146029695717560320, journalId=1146123166801305609, issueId=1261262687258985194, articleNumber=null, orderNo=null, doi=10.12404/j.issn.1671-1815.2407479, pmid=null, cstr=null, oa=null, hot=null, price=null, onlineType=0, articleFormat=0, articleType=null, articleTypeStr=research-article, receivedDate=1728489600000, receivedDateStr=2024-10-10, revisedDate=1745769600000, revisedDateStr=2025-04-28, acceptedDate=null, acceptedDateStr=null, onlineDate=1778638059311, onlineDateStr=2026-05-13, pubDate=1752768000000, pubDateStr=2025-07-18, doiRegisterDate=null, doiRegisterDateStr=null, onlineIssueDate=1778638059311, onlineIssueDateStr=2026-05-13, onlineJustAcceptDate=null, onlineJustAcceptDateStr=null, onlineFirstDate=null, onlineFirstDateStr=null, sourceXml=null, magXml=null, createTime=1778638059311, creator=13701087609, updateTime=1778638059311, updator=13701087609, issue=Issue{id=1261262687258985194, tenantId=1146029695717560320, journalId=1146123166801305609, year='2025', volume='25', issue='20', pageStart='8317', pageEnd='8759', issueExtLink='null', onlineDate='null', pubDate='null', beforeIssueId=null, nextIssueId=null, price=null, status=1, issueComplete=1, articleOrder=1, issueType=-1, specialIssue=null, createTime=1778638057769, creator=13701087609, updateTime=1778753106634, updator=13701087609, preIssue=null, nextIssue=null, ext={EN=IssueExt(id=1261745237240722095, tenantId=1146029695717560320, journalId=1146123166801305609, issueId=1261262687258985194, language=EN, specialIssueTitle=, coverIllustrator=null, specialIssueEditor=, specialIssueAbout=), CN=IssueExt(id=1261745237240722096, tenantId=1146029695717560320, journalId=1146123166801305609, issueId=1261262687258985194, language=CN, specialIssueTitle=, coverIllustrator=null, specialIssueEditor=, specialIssueAbout=)}, issueFiles=null}, startPage=8514, endPage=8525, ext={EN=ArticleExt(id=1261262696893244102, articleId=1261262693726544554, tenantId=1146029695717560320, journalId=1146123166801305609, language=EN, title=Wind Speed Prediction Based on Modal Decomposition and RIME-CNN-BiLSTM-AM, columnId=1156262733675876713, journalTitle=Science Technology and Engineering, columnName=Papers·Electrical Technology, runingTitle=null, highlight=null, articleAbstract=

Serving as a clean and renewable energy source, wind energy plays a significant role in mitigating the increasingly severe energy crisis. However, the fluctuation and randomness of wind speed pose severe challenges to the stable operation of power systems. To address this issue, a combined short-term wind speed forecasting model named CEEMDAN-RIME-CNN-BiLSTM-AM was proposed, which was based on complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN), rime optimization algorithm (RIME), convolutional neural network (CNN), bidirectional long short-term memory network (BiLSTM), and attention mechanism (AM). Initially, the CEEMDAN algorithm was applied to the original wind speed series to obtain a series of relatively stable sub-modes, thereby reducing the volatility of the wind speed series. Subsequently, the CNN hyperparameters were optimized using the RIME algorithm to establish the CNN-RIME model for adaptive extraction and mining of wind speed data. Then, the BiLSTM-AM model was employed to forecast the processed data. Finally, the forecasting results of each sub-series were superimposed to obtain the final forecasting result. A comparative experiment was conducted using an actual wind speed dataset from a certain location. The proposed model demonstrates good forecasting performance in both single-step and multi-step forecasting, providing a reference for scheduling plans to maximize energy utilization and power supply.

, correspAuthors=Qi-sheng YAN, 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=Ting ZHU, Qi-sheng YAN), CN=ArticleExt(id=1261262713477522293, articleId=1261262693726544554, tenantId=1146029695717560320, journalId=1146123166801305609, language=CN, title=基于模态分解和RIME-CNN-BiLSTM-AM的风速预测方法, columnId=1156262734506353627, journalTitle=科学技术与工程, columnName=论文·电工技术, runingTitle=null, highlight=null, articleAbstract=

作为一种清洁的可再生能源,风能在缓解日益严重的能源危机方面充当着重要作用。然而,风速的波动性和随机性给电力系统的稳定运行带来了严峻的挑战。针对该问题,提出一种基于自适应噪声完备集合经验模态分解(complete ensemble empirical mode decomposition with adaptive noise,CEEMDAN)与霜冰优化算法(rime optimization algorithm,RIME)-卷积神经网络(convolutional neural network,CNN)-双向长短期记忆网络(bidirectional long short-term memory network,BiLSTM)-注意力机制(attention mechanism,AM)的短期风速预测组合模型CEEMDAN-RIME-CNN-BiLSTM-AM。首先,对初始风速序列采用CEEMDAN算法,得到一系列较平稳的子模态,以降低风速序列的波动性;然后,采用RIME霜冰优化算法优化CNN超参数,建立CNN-RIME模型,对风速数据进行自适应提取和挖掘;接着,采用BiLSTM-AM模型对处理后的数据进行预测;最后,将各子序列的预测结果叠加,得到最终预测结果。以某地实际风速数据集进行对比试验,该模型在单步与多步预测中均展现出良好的预测性能,可以为制定调度计划提供参考,以最大程度地提高能源利用率和供电。

, correspAuthors=颜七笙, authorNote=null, correspAuthorsNote=
* 颜七笙(1976—),男,汉族,江西临川人,博士,教授。研究方向:智能计算及应用。E-mail:
, copyrightStatement=null, copyrightOwner=null, extLink=null, articleAbsUrl=null, sourceXml=u4qGhrS9ANiw+BDvMIL0fg==, magXml=dkDoRWWW0+5x3myoZYBcPw==, pdfUrl=null, pdf=OKjhdLKe46bYgKiQZ5D4Dg==, pdfFileSize=12457889, pdfExtLink=null, richHtmlUrl=null, mobilePdfUrl=null, reviewReport=null, pdfFirstPage=null, abstractGraph=TtOyek2kpZ5OHPXjtAafDw==, abstractGraphContent=null, abstractVideo=null, citation=null, cebUrl=null, magXmlContent=Vp6oidMyUQXfX8BjjN9tzw==, mapNumber=null, authorCompany=null, fund=null, authors=

朱婷(2001—),女,汉族,安徽阜阳人,硕士研究生。研究方向:智能计算及应用。E-mail:

, authorsList=朱婷, 颜七笙)}, authors=[Author(id=1261377022295720263, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, orderNo=0, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=1668986880@qq.com, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1261377023470125391, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, authorId=1261377022295720263, language=EN, stringName=Ting ZHU, firstName=Ting, middleName=null, lastName=ZHU, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1 College of Economic and Management, East China University of Technology, Nanchang 330000, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1261377025009434966, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, authorId=1261377022295720263, language=CN, stringName=朱婷, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1 东华理工大学经济与管理学院, 南昌 330000, bio={"content":"

朱婷(2001—),女,汉族,安徽阜阳人,硕士研究生。研究方向:智能计算及应用。E-mail:

"}, bioImg=null, bioContent=

朱婷(2001—),女,汉族,安徽阜阳人,硕士研究生。研究方向:智能计算及应用。E-mail:

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1261377020966125876, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, xref=1, ext=[AuthorCompanyExt(id=1261377020987097397, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, companyId=1261377020966125876, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 College of Economic and Management, East China University of Technology, Nanchang 330000, China), AuthorCompanyExt(id=1261377021066789176, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, companyId=1261377020966125876, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 东华理工大学经济与管理学院, 南昌 330000)])]), Author(id=1261377025659552098, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, orderNo=1, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=yanqs93@126.com, emailSecond=null, emailThird=null, correspondingAuthor=1, authorType=1, ext={EN=AuthorExt(id=1261377026032845158, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, authorId=1261377025659552098, language=EN, stringName=Qi-sheng YAN, firstName=Qi-sheng, middleName=null, lastName=YAN, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, *, address=2 College of Science, East China University of Technology, Nanchang 330013, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1261377026506801517, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, authorId=1261377025659552098, language=CN, stringName=颜七笙, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, *, address=2 东华理工大学理学院, 南昌 330013, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1261377021431693627, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, xref=2, ext=[AuthorCompanyExt(id=1261377021809180990, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, companyId=1261377021431693627, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 College of Science, East China University of Technology, Nanchang 330013, China), AuthorCompanyExt(id=1261377021939204417, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, companyId=1261377021431693627, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 东华理工大学理学院, 南昌 330013)])])], keywords=[Keyword(id=1261377027391799667, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, language=EN, orderNo=1, keyword=wind speed prediction), Keyword(id=1261377029434425723, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, language=EN, orderNo=2, keyword=complete ensemble empirical mode decomposition with adaptive noise(CEEMDAN)), Keyword(id=1261377030315229571, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, language=EN, orderNo=3, keyword=rime optimization algorithm (RIME)), Keyword(id=1261377030621413774, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, language=EN, orderNo=4, keyword=convolutional neural network (CNN)), Keyword(id=1261377031242170773, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, language=EN, orderNo=5, keyword=bidirectional long short-term memory network (BiLSTM)), Keyword(id=1261377031636435359, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, language=EN, orderNo=6, keyword=attention mechanism(AM)), Keyword(id=1261377032424964528, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, language=CN, orderNo=1, keyword=风速预测), Keyword(id=1261377034056548790, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, language=CN, orderNo=2, keyword=自适应噪声完备集合经验模态分解(CEEMDAN)), Keyword(id=1261377034538893764, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, language=CN, orderNo=3, keyword=霜冰优化算法(RIME)), Keyword(id=1261377035319034318, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, language=CN, orderNo=4, keyword=卷积神经网络(CNN)), Keyword(id=1261377035721687508, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, language=CN, orderNo=5, keyword=双向长短期记忆网络(BiLSTM)), Keyword(id=1261377036547965404, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, language=CN, orderNo=6, keyword=注意力机制(AM))], refs=[Reference(id=1261377070450524977, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, doi=null, pmid=null, pmcid=null, year=2024, volume=24, issue=35, pageStart=15065, pageEnd=15073, url=null, language=null, rfNumber=[1], rfOrder=0, authorNames=刘向龙, 蔡文斌, 于源, journalName=科学技术与工程, refType=null, unstructuredReference=刘向龙, 蔡文斌, 于源, . 基于系统调度成本和储能经济收益规划的“源-网-储”协同互动优化调度策略[J]. 科学技术与工程, 2024, 24(35): 15065-15073., articleTitle=基于系统调度成本和储能经济收益规划的“源-网-储”协同互动优化调度策略, refAbstract=null), Reference(id=1261377070840595256, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, doi=null, pmid=null, pmcid=null, year=2024, volume=24, issue=35, pageStart=15065, pageEnd=15073, url=null, language=null, rfNumber=[1], rfOrder=1, authorNames=Liu Xianglong, Cai Wenbin, Yu Yuan, journalName=Science Technology and Engineering, refType=null, unstructuredReference=Liu Xianglong, Cai Wenbin, Yu Yuan, et al. “Source-network-storage” collaborative and interactive optimal dispatch strategy based onsys tem dispatch cost and energy storage economic benefit objectives[J]. Science Technology and Engineering, 2024, 24(35): 15065-15073., articleTitle=“Source-network-storage” collaborative and interactive optimal dispatch strategy based onsys tem dispatch cost and energy storage economic benefit objectives, refAbstract=null), Reference(id=1261377071415214912, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, doi=null, pmid=null, pmcid=null, year=2024, volume=24, issue=34, pageStart=14503, pageEnd=14521, url=null, language=null, rfNumber=[2], rfOrder=2, authorNames=郑晓亮, 杨晓亮, 来文豪, journalName=科学技术与工程, refType=null, unstructuredReference=郑晓亮, 杨晓亮, 来文豪. 新能源发电预测方法分类及研究进展[J]. 科学技术与工程, 2024, 24 (34): 14503-14521., articleTitle=新能源发电预测方法分类及研究进展, refAbstract=null), Reference(id=1261377071838839620, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, doi=null, pmid=null, pmcid=null, year=2024, volume=24, issue=34, pageStart=14503, pageEnd=14521, url=null, language=null, rfNumber=[2], rfOrder=3, authorNames=Zheng Xiaoliang, Yang Xiaoliang, Lai Wenhao, journalName=Science Technology and Engineering, refType=null, unstructuredReference=Zheng Xiaoliang, Yang Xiaoliang, Lai Wenhao. Classification and research progress of new energy power generation forecasting methods[J]. Science Technology and Engineering, 2024, 24(34): 14503-14521., articleTitle=Classification and research progress of new energy power generation forecasting methods, refAbstract=null), Reference(id=1261377072010806088, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, doi=null, pmid=null, pmcid=null, year=2019, volume=237, issue=null, pageStart=1, pageEnd=10, url=null, language=null, rfNumber=[3], rfOrder=4, authorNames=Wang H, Han S, Liu Y, journalName=Applied Energy, refType=null, unstructuredReference=Wang H, Han S, Liu Y, et al. Sequence transfer correction algorithm for numerical weather prediction wind speed and its application in a wind power forecasting system[J]. Applied Energy, 2019, 237: 1-10., articleTitle=Sequence transfer correction algorithm for numerical weather prediction wind speed and its application in a wind power forecasting system, refAbstract=null), Reference(id=1261377072203744078, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, doi=null, pmid=null, pmcid=null, year=2022, volume=50, issue=22, pageStart=49, pageEnd=58, url=null, language=null, rfNumber=[4], rfOrder=5, authorNames=张金良, 刘子毅, journalName=电力系统保护与控制, refType=null, unstructuredReference=张金良, 刘子毅. 基于混合模型的超短期风速区间预测[J]. 电力系统保护与控制, 2022, 50(22): 49-58., articleTitle=基于混合模型的超短期风速区间预测, refAbstract=null), Reference(id=1261377073990517590, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, doi=null, pmid=null, pmcid=null, year=2022, volume=50, issue=22, pageStart=49, pageEnd=58, url=null, language=null, rfNumber=[4], rfOrder=6, authorNames=Zhang Jinliang, Liu Ziyi, journalName=Power System Protection and Control, refType=null, unstructuredReference=Zhang Jinliang, Liu Ziyi. Ultra short term wind speed interval prediction based on a hybrid model[J]. Power System Protection and Control, 2022, 50(22): 49-58., articleTitle=Ultra short term wind speed interval prediction based on a hybrid model, refAbstract=null), Reference(id=1261377074330256221, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, doi=null, pmid=null, pmcid=null, year=2008, volume=null, issue=2, pageStart=374, pageEnd=376, url=null, language=null, rfNumber=[5], rfOrder=7, authorNames=蒋金良, 林广明, journalName=控制理论与应用, refType=null, unstructuredReference=蒋金良, 林广明. 基于ARIMA模型的自动站风速预测[J]. 控制理论与应用, 2008(2): 374-376., articleTitle=基于ARIMA模型的自动站风速预测, refAbstract=null), Reference(id=1261377074644829028, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, doi=null, pmid=null, pmcid=null, year=2008, volume=null, issue=2, pageStart=374, pageEnd=376, url=null, language=null, rfNumber=[5], rfOrder=8, authorNames=Jiang Jinliang, Lin Guangming, journalName=Control Theory & Applications, refType=null, unstructuredReference=Jiang Jinliang, Lin Guangming. Automatic station wind speed forecasting based on ARIMA model[J]. Control Theory & Applications, 2008(2): 374-376., articleTitle=Automatic station wind speed forecasting based on ARIMA model, refAbstract=null), Reference(id=1261377074883904364, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, doi=null, pmid=null, pmcid=null, year=2022, volume=44, issue=2, pageStart=200, pageEnd=204, url=null, language=null, rfNumber=[6], rfOrder=9, authorNames=余泳涛, 张扬, journalName=南昌大学学报(工科版), refType=null, unstructuredReference=余泳涛, 张扬. 基于改进Logistic模型法的配电网负荷预测[J]. 南昌大学学报(工科版), 2022, 44 (2): 200-204., articleTitle=基于改进Logistic模型法的配电网负荷预测, refAbstract=null), Reference(id=1261377075039093617, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, doi=null, pmid=null, pmcid=null, year=2022, volume=44, issue=2, pageStart=200, pageEnd=204, url=null, language=null, rfNumber=[6], rfOrder=10, authorNames=Yu Yongtao, Zhang Yang, journalName=Journal of Nanchang University(Engineering & Technology), refType=null, unstructuredReference=Yu Yongtao, Zhang Yang. Distribution network load forecasting based on PSO-Logistic model method[J]. Journal of Nanchang University(Engineering & Technology), 2022, 44(2): 200-204., articleTitle=Distribution network load forecasting based on PSO-Logistic model method, refAbstract=null), Reference(id=1261377075215254393, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, doi=null, pmid=null, pmcid=null, year=2020, volume=146, issue=null, pageStart=149, pageEnd=165, url=null, language=null, rfNumber=[7], rfOrder=11, authorNames=Wu C Y, Wang J Z, Chen X J, journalName=Renewable Energy, refType=null, unstructuredReference=Wu C Y, Wang J Z, Chen X J, et al. A novel hybrid system based on multi-objective optimization for wind speed forecasting[J]. Renewable Energy, 2020, 146: 149-165., articleTitle=A novel hybrid system based on multi-objective optimization for wind speed forecasting, refAbstract=null), Reference(id=1261377075383026561, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, doi=null, pmid=null, pmcid=null, year=2019, volume=16, issue=1, pageStart=202, pageEnd=214, url=null, language=null, rfNumber=[8], rfOrder=12, authorNames=Ibrahim I A, HossainM J, Duck B C, journalName=IEEE Transactions on Industrial Informatics, refType=null, unstructuredReference=Ibrahim I A, HossainM J, Duck B C. An optimized offline random forests based model for ultra-short-term prediction of PV characteristics[J]. IEEE Transactions on Industrial Informatics, 2019, 16(1): 202-214., articleTitle=An optimized offline random forests based model for ultra-short-term prediction of PV characteristics, refAbstract=null), Reference(id=1261377075626296203, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, doi=null, pmid=null, pmcid=null, year=2022, volume=22, issue=36, pageStart=16067, pageEnd=16074, url=null, language=null, rfNumber=[9], rfOrder=13, authorNames=王愈轩, 刘尔佳, 黄永章, journalName=科学技术与工程, refType=null, unstructuredReference=王愈轩, 刘尔佳, 黄永章. 基于CNN-LSTM-lightGBM组合的超短期风电功率预测方法[J]. 科学技术与工程, 2022, 22(36): 16067-16074., articleTitle=基于CNN-LSTM-lightGBM组合的超短期风电功率预测方法, refAbstract=null), Reference(id=1261377075869565842, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, doi=null, pmid=null, pmcid=null, year=2022, volume=22, issue=36, pageStart=16067, pageEnd=16074, url=null, language=null, rfNumber=[9], rfOrder=14, authorNames=Wang Yuxuan, Liu Erjia, Huang Yongzhang, journalName=Science Technology and Engineering, refType=null, unstructuredReference=Wang Yuxuan, Liu Erjia, Huang Yongzhang. An ultra-short-term wind power prediction method based on CNN-LSTM-lightGBM combination[J]. Science Technology and Engineering, 2022, 22(36): 16067-16074., articleTitle=An ultra-short-term wind power prediction method based on CNN-LSTM-lightGBM combination, refAbstract=null), Reference(id=1261377076205110172, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, doi=null, pmid=null, pmcid=null, year=2021, volume=215, issue=null, pageStart=119180, pageEnd=null, url=null, language=null, rfNumber=[10], rfOrder=15, authorNames=Dong Y X, Wang J, Xiao L, journalName=Energy, refType=null, unstructuredReference=Dong Y X, Wang J, Xiao L, et al. Short-term wind speed time series forecasting based on a hybridmethod with multiple objective optimization for non-convex target[J]. Energy, 2021, 215:119180., articleTitle=Short-term wind speed time series forecasting based on a hybridmethod with multiple objective optimization for non-convex target, refAbstract=null), Reference(id=1261377076473545635, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, doi=null, pmid=null, pmcid=null, year=2021, volume=42, issue=10, pageStart=237, pageEnd=242, url=null, language=null, rfNumber=[11], rfOrder=16, authorNames=向玲, 李京蓄, 王朋鹤, journalName=太阳能学报, refType=null, unstructuredReference=向玲, 李京蓄, 王朋鹤, . 基于VMD-FIG和参数优化GRU的风速多步区间预测[J]. 太阳能学报, 2021, 42(10): 237-242., articleTitle=基于VMD-FIG和参数优化GRU的风速多步区间预测, refAbstract=null), Reference(id=1261377076607763370, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, doi=null, pmid=null, pmcid=null, year=2021, volume=42, issue=10, pageStart=237, pageEnd=242, url=null, language=null, rfNumber=[11], rfOrder=17, authorNames=Xiang Ling, Li Jingxu, Wang Penghe, journalName=Acta Energiae Solaris Sinica, refType=null, unstructuredReference=Xiang Ling, Li Jingxu, Wang Penghe, et al. Wind speed multistep interval forecasting based on VMD-FIG and parameter optimized GRU[J]. Acta Energiae Solaris Sinica, 2021, 42(10): 237-242., articleTitle=Wind speed multistep interval forecasting based on VMD-FIG and parameter optimized GRU, refAbstract=null), Reference(id=1261377076741981105, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, doi=null, pmid=null, pmcid=null, year=2024, volume=24, issue=25, pageStart=10790, pageEnd=10797, url=null, language=null, rfNumber=[12], rfOrder=18, authorNames=朱彦民, 李忠虎, 杨立清, journalName=科学技术与工程, refType=null, unstructuredReference=朱彦民, 李忠虎, 杨立清, . 基于卷积双向长短时记忆模型的风电机组故障预测[J]. 科学技术与工程, 2024, 24(25): 10790-10797., articleTitle=基于卷积双向长短时记忆模型的风电机组故障预测, refAbstract=null), Reference(id=1261377078570697654, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, doi=null, pmid=null, pmcid=null, year=2024, volume=24, issue=25, pageStart=10790, pageEnd=10797, url=null, language=null, rfNumber=[12], rfOrder=19, authorNames=Zhu Yanmin, Li Zhonghu, Yang Liqing, journalName=Science Technology and Engineering, refType=null, unstructuredReference=Zhu Yanmin, Li Zhonghu, Yang Liqing, et al. Fault prediction of wind turbine based on convolutional bidirectional long short term memory model[J]. Science Technology and Engineering, 2024, 24(25): 10790-10797., articleTitle=Fault prediction of wind turbine based on convolutional bidirectional long short term memory model, refAbstract=null), Reference(id=1261377078847521723, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, doi=null, pmid=null, pmcid=null, year=2024, volume=45, issue=1, pageStart=159, pageEnd=170, url=null, language=null, rfNumber=[13], rfOrder=20, authorNames=毕贵红, 黄泽, 赵四洪, journalName=太阳能学报, refType=null, unstructuredReference=毕贵红, 黄泽, 赵四洪, . 基于混合分解和PCG-BiLSTM的风速短期预测[J]. 太阳能学报, 2024, 45(1): 159-170., articleTitle=基于混合分解和PCG-BiLSTM的风速短期预测, refAbstract=null), Reference(id=1261377079023682495, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, doi=null, pmid=null, pmcid=null, year=2024, volume=45, issue=1, pageStart=159, pageEnd=170, url=null, language=null, rfNumber=[13], rfOrder=21, authorNames=Bi Guihong, Huang Ze, Zhao Sihong, journalName=Acta Energiae Solaris Sinica, refType=null, unstructuredReference=Bi Guihong, Huang Ze, Zhao Sihong, et al. Short-term prediction of wind speed based on hybird decomposition and PCG-BiLSTM[J]. Acta Energiae Solaris Sinica, 2024, 45(1): 159-170., articleTitle=Short-term prediction of wind speed based on hybird decomposition and PCG-BiLSTM, refAbstract=null), Reference(id=1261377079162094534, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, doi=null, pmid=null, pmcid=null, year=2023, volume=42, issue=6, pageStart=70, pageEnd=77, url=null, language=null, rfNumber=[14], rfOrder=22, authorNames=沈建良, 来骏, 张翼, journalName=浙江电力, refType=null, unstructuredReference=沈建良, 来骏, 张翼, . 基于ACMD和BiGRU-Attention的负荷预测模型研究[J]. 浙江电力, 2023, 42(6): 70-77., articleTitle=基于ACMD和BiGRU-Attention的负荷预测模型研究, refAbstract=null), Reference(id=1261377079447307212, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, doi=null, pmid=null, pmcid=null, year=2023, volume=42, issue=6, pageStart=70, pageEnd=77, url=null, language=null, rfNumber=[14], rfOrder=23, authorNames=Shen Jianliang, Lai Jun, Zhang Yi, journalName=Zhejiang Electric Power, refType=null, unstructuredReference=Shen Jianliang, Lai Jun, Zhang Yi, et al. Research on a load forecasting model based on ACMD and BiGRU-Attention[J]. Zhejiang Electric Power, 2023, 42(6): 70-77., articleTitle=Research on a load forecasting model based on ACMD and BiGRU-Attention, refAbstract=null), Reference(id=1261377079719936978, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, doi=null, pmid=null, pmcid=null, year=2022, volume=41, issue=5, pageStart=8, pageEnd=17, url=null, language=null, rfNumber=[15], rfOrder=24, authorNames=段雪滢, 李小腾, 陈文洁, journalName=电工电能新技术, refType=null, unstructuredReference=段雪滢, 李小腾, 陈文洁. 基于改进粒子群优化算法的VMD-GRU短期电力负荷预测[J]. 电工电能新技术, 2022, 41(5): 8-17., articleTitle=基于改进粒子群优化算法的VMD-GRU短期电力负荷预测, refAbstract=null), Reference(id=1261377079942235093, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, doi=null, pmid=null, pmcid=null, year=2022, volume=41, issue=5, pageStart=8, pageEnd=17, url=null, language=null, rfNumber=[15], rfOrder=25, authorNames=Duan Xueying, Li Xiaoteng, Chen Wenjie, journalName=Advanced Technology of Electrical Engineering and Energy, refType=null, unstructuredReference=Duan Xueying, Li Xiaoteng, Chen Wenjie. Improved particles swarm optimization algorithm-based VMD-GRU method for short-term load forecasting[J]. Advanced Technology of Electrical Engineering and Energy, 2022, 41(5): 8-17., articleTitle=Improved particles swarm optimization algorithm-based VMD-GRU method for short-term load forecasting, refAbstract=null), Reference(id=1261377080424580057, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, doi=null, pmid=null, pmcid=null, year=2022, volume=36, issue=4, pageStart=17, pageEnd=23, url=null, language=null, rfNumber=[16], rfOrder=26, authorNames=王贺, 陈蕻峰, 熊敏, journalName=电子测量与仪器学报, refType=null, unstructuredReference=王贺, 陈蕻峰, 熊敏, . 融合CEEMDAN和ICS-LSTM的短期风速预测建模[J]. 电子测量与仪器学报, 2022, 36(4): 17-23., articleTitle=融合CEEMDAN和ICS-LSTM的短期风速预测建模, refAbstract=null), Reference(id=1261377080802067422, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, doi=null, pmid=null, pmcid=null, year=2022, volume=36, issue=4, pageStart=17, pageEnd=23, url=null, language=null, rfNumber=[16], rfOrder=27, authorNames=Wang He, Chen Honngfeng, Xiong Min, journalName=Journal of Electronic Measurement and Instrumentation, refType=null, unstructuredReference=Wang He, Chen Honngfeng, Xiong Min, et al. Short-term wind speed forecasting modeling integrating CEEMDAN and ICS-LSTM[J]. Journal of Electronic Measurement and Instrumentation, 2022, 36(4): 17-23., articleTitle=Short-term wind speed forecasting modeling integrating CEEMDAN and ICS-LSTM, refAbstract=null), Reference(id=1261377081011782625, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, doi=null, pmid=null, pmcid=null, year=2022, volume=34, issue=4, pageStart=99, pageEnd=106, url=null, language=null, rfNumber=[17], rfOrder=28, authorNames=刘可真, 阮俊枭, 赵现平, journalName=电力系统及其自动化学报, refType=null, unstructuredReference=刘可真, 阮俊枭, 赵现平, . 基于麻雀搜索优化的Attention-GRU短期负荷预测方法[J]. 电力系统及其自动化学报, 2022, 34(4): 99-106., articleTitle=基于麻雀搜索优化的Attention-GRU短期负荷预测方法, refAbstract=null), Reference(id=1261377081309578217, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, doi=null, pmid=null, pmcid=null, year=2022, volume=34, issue=4, pageStart=99, pageEnd=106, url=null, language=null, rfNumber=[17], rfOrder=29, authorNames=Liu Kezhen, Ruan Junxiao, Zhao Xianping, journalName=Proceedings of the CSU-EPSA, refType=null, unstructuredReference=Liu Kezhen, Ruan Junxiao, Zhao Xianping, et al. Short-term load forecasting method based on sparrow search optimized Attention-GRU[J]. Proceedings of the CSU-EPSA, 2022, 34(4): 99-106., articleTitle=Short-term load forecasting method based on sparrow search optimized Attention-GRU, refAbstract=null), Reference(id=1261377083243152369, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, doi=null, pmid=null, pmcid=null, year=2023, volume=23, issue=35, pageStart=15091, pageEnd=15099, url=null, language=null, rfNumber=[18], rfOrder=30, authorNames=程杰, 陈鼎, 李春, journalName=科学技术与工程, refType=null, unstructuredReference=程杰, 陈鼎, 李春, . 基于GWO-CNN-BiLSTM的超短期风电预测[J]. 科学技术与工程, 2023, 23(35): 15091-15099., articleTitle=基于GWO-CNN-BiLSTM的超短期风电预测, refAbstract=null), Reference(id=1261377083507393523, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, doi=null, pmid=null, pmcid=null, year=2023, volume=23, issue=35, pageStart=15091, pageEnd=15099, url=null, language=null, rfNumber=[18], rfOrder=31, authorNames=Cheng Jie, Chen Ding, Li Chun, journalName=Science Technology and Engineering, refType=null, unstructuredReference=Cheng Jie, Chen Ding, Li Chun, et al. Ultra-short-term wind power prediction based on GWO-CNN-BiLSTM[J]. Science Technology and Engineering, 2023, 23(35): 15091-15099., articleTitle=Ultra-short-term wind power prediction based on GWO-CNN-BiLSTM, refAbstract=null), Reference(id=1261377083666777077, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, doi=null, pmid=null, pmcid=null, year=2020, volume=48, issue=12, pageStart=90, pageEnd=96, url=null, language=null, rfNumber=[19], rfOrder=32, authorNames=韩宏志, 唐振浩, journalName=电力系统保护与控制, refType=null, unstructuredReference=韩宏志, 唐振浩. 基于CEEMDAN与回声状态网络的风速预测方法[J]. 电力系统保护与控制, 2020, 48(12): 90-96., articleTitle=基于CEEMDAN与回声状态网络的风速预测方法, refAbstract=null), Reference(id=1261377084082013181, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, doi=null, pmid=null, pmcid=null, year=2020, volume=48, issue=12, pageStart=90, pageEnd=96, url=null, language=null, rfNumber=[19], rfOrder=33, authorNames=Han Hongzhi, Tang Zhenhao, journalName=Power System Protection and Control, refType=null, unstructuredReference=Han Hongzhi, Tang Zhenhao. Wind speed prediction method based on CEEMDAN and echo state network[J]. Power System Protection and Control, 2020, 48(12): 90-96., articleTitle=Wind speed prediction method based on CEEMDAN and echo state network, refAbstract=null), Reference(id=1261377084509831169, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, doi=null, pmid=null, pmcid=null, year=2023, volume=532, issue=null, pageStart=183, pageEnd=214, url=null, language=null, rfNumber=[20], rfOrder=34, authorNames=Su H, Zhao D, Heidari A A, journalName=Neurocomputing, refType=null, unstructuredReference=Su H, Zhao D, Heidari A A, et al. RIME: A physics-based optimization[J]. Neurocomputing, 2023, 532: 183-214., articleTitle=RIME: A physics-based optimization, refAbstract=null), Reference(id=1261377084639854600, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, doi=null, pmid=null, pmcid=null, year=2023, volume=351, issue=null, pageStart=121823, pageEnd=null, url=null, language=null, rfNumber=[21], rfOrder=35, authorNames=Li K, Mu Y, Yang F, journalName=Applied Energy, refType=null, unstructuredReference=Li K, Mu Y, Yang F, et al. A novel short-term multi-energy load forecasting method for integrated energy system based on feature separation-fusion technology and improved CNN[J]. Applied Energy, 2023, 351: 121823., articleTitle=A novel short-term multi-energy load forecasting method for integrated energy system based on feature separation-fusion technology and improved CNN, refAbstract=null), Reference(id=1261377084895707147, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, doi=null, pmid=null, pmcid=null, year=2023, volume=285, issue=null, pageStart=128762, pageEnd=null, url=null, language=null, rfNumber=[22], rfOrder=36, authorNames=Zhang D, Chen B, Zhu H, journalName=Energy, refType=null, unstructuredReference=Zhang D, Chen B, Zhu H, et al. Short-term wind power prediction based on two-layer decomposition and BiTCN-BiLSTM-attention model[J]. Energy, 2023, 285: 128762., articleTitle=Short-term wind power prediction based on two-layer decomposition and BiTCN-BiLSTM-attention model, refAbstract=null), Reference(id=1261377085201891344, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, doi=null, pmid=null, pmcid=null, year=2021, volume=49, issue=16, pageStart=60, pageEnd=66, url=null, language=null, rfNumber=[23], rfOrder=37, authorNames=彭华, 王文超, 朱永利, journalName=电力系统保护与控制, refType=null, unstructuredReference=彭华, 王文超, 朱永利, . 基于LSTM神经网络的风电场集电线路单相接地智能测距[J]. 电力系统保护与控制, 2021, 49(16): 60-66., articleTitle=基于LSTM神经网络的风电场集电线路单相接地智能测距, refAbstract=null), Reference(id=1261377087391318038, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, doi=null, pmid=null, pmcid=null, year=2021, volume=49, issue=16, pageStart=60, pageEnd=66, url=null, language=null, rfNumber=[23], rfOrder=38, authorNames=Peng Hua, Wang Wenchao, Zhu Yongli, journalName=Power System Protection and Control, refType=null, unstructuredReference=Peng Hua, Wang Wenchao, Zhu Yongli, et al. An intelligent single-phase grounding fault location for a wind farm collection line based on an LSTM neural network[J]. Power System Protection and Control, 2021, 49(16): 60-66., articleTitle=An intelligent single-phase grounding fault location for a wind farm collection line based on an LSTM neural network, refAbstract=null), Reference(id=1261377087777194011, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, doi=null, pmid=null, pmcid=null, year=2014, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[24], rfOrder=39, authorNames=Mnih V, Heess N, Graves A, journalName=Advances in Neural Information Processing Systems, refType=null, unstructuredReference=Mnih V, Heess N, Graves A. Recurrent models of visual attention[J]. Advances in Neural Information Processing Systems, 2014, 27: DOI: 10.48550/arXiv.1406.6247., articleTitle=Recurrent models of visual attention, refAbstract=null), Reference(id=1261377088200818721, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, doi=null, pmid=null, pmcid=null, year=2022, volume=43, issue=12, pageStart=273, pageEnd=281, url=null, language=null, rfNumber=[25], rfOrder=40, authorNames=杨国清, 刘世林, 王德意, journalName=太阳能学报, refType=null, unstructuredReference=杨国清, 刘世林, 王德意, . 基于Attention-GRU风速修正和Stacking的短期风电功率预测[J]. 太阳能学报, 2022, 43(12): 273-281., articleTitle=基于Attention-GRU风速修正和Stacking的短期风电功率预测, refAbstract=null), Reference(id=1261377088540557352, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, doi=null, pmid=null, pmcid=null, year=2022, volume=43, issue=12, pageStart=273, pageEnd=281, url=null, language=null, rfNumber=[25], rfOrder=41, authorNames=Yang Guoqing, Liu Shilin, Wang Deyi et al, journalName=Acta Energiae Solaris Sinica, refType=null, unstructuredReference=Yang Guoqing, Liu Shilin, Wang Deyi et al. Short-term wind power forecasting based on Attention-GRU wind speed correction and Stacking[J]. Acta Energiae Solaris Sinica, 2022, 43(12): 273-281., articleTitle=Short-term wind power forecasting based on Attention-GRU wind speed correction and Stacking, refAbstract=null)], funds=[Fund(id=1261377069762659105, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, awardId=71961001, language=CN, fundingSource=国家自然科学基金(71961001), fundOrder=null, country=null)], companyList=[AuthorCompany(id=1261377020966125876, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, xref=1, ext=[AuthorCompanyExt(id=1261377020987097397, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, companyId=1261377020966125876, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 College of Economic and Management, East China University of Technology, Nanchang 330000, China), AuthorCompanyExt(id=1261377021066789176, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, companyId=1261377020966125876, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 东华理工大学经济与管理学院, 南昌 330000)]), AuthorCompany(id=1261377021431693627, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, xref=2, ext=[AuthorCompanyExt(id=1261377021809180990, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, companyId=1261377021431693627, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 College of Science, East China University of Technology, Nanchang 330013, China), AuthorCompanyExt(id=1261377021939204417, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, companyId=1261377021431693627, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 东华理工大学理学院, 南昌 330013)])], figs=[ArticleFig(id=1261377039039382002, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, language=EN, label=Fig.1, caption=Flow chart of RIME algorithm, figureFileSmall=0TFtSOdRI+pN1hMMfHTPOQ==, figureFileBig=/5sDwk6uhxuNYgLLXhx0VA==, tableContent=null), ArticleFig(id=1261377039567864318, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, language=CN, label=图1, caption=RIME算法流程图, figureFileSmall=0TFtSOdRI+pN1hMMfHTPOQ==, figureFileBig=/5sDwk6uhxuNYgLLXhx0VA==, tableContent=null), ArticleFig(id=1261377040675160588, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, language=EN, label=Fig.2, caption=CNN structure diagram, figureFileSmall=a5IDj7xTgjBNZpv1lHD1/A==, figureFileBig=9B2dWt8j/X8nZF7xkT6XPw==, tableContent=null), ArticleFig(id=1261377041270751766, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, language=CN, label=图2, caption=CNN结构图, figureFileSmall=a5IDj7xTgjBNZpv1lHD1/A==, figureFileBig=9B2dWt8j/X8nZF7xkT6XPw==, tableContent=null), ArticleFig(id=1261377042730369565, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, language=EN, label=Fig.3, caption=BiLSTM structure diagram, figureFileSmall=wtksXvhiFbJy/OKXmKWGmQ==, figureFileBig=wW/cK/SJ8ycwcL+3+cZX9w==, tableContent=null), ArticleFig(id=1261377042902336037, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, language=CN, label=图3, caption=BiLSTM结构图, figureFileSmall=wtksXvhiFbJy/OKXmKWGmQ==, figureFileBig=wW/cK/SJ8ycwcL+3+cZX9w==, tableContent=null), ArticleFig(id=1261377043304989228, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, language=EN, label=Fig.4, caption=Structure of the attention mechanism, figureFileSmall=SQiP7nPVemsPp/E5nrG/YA==, figureFileBig=1X2lEXeerwfDY2pJYW6v6w==, tableContent=null), ArticleFig(id=1261377043728613943, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, language=CN, label=图4, caption=注意力机制结构图, figureFileSmall=SQiP7nPVemsPp/E5nrG/YA==, figureFileBig=1X2lEXeerwfDY2pJYW6v6w==, tableContent=null), ArticleFig(id=1261377044877853254, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, language=EN, label=Fig.5, caption=Model prediction framework, figureFileSmall=EEjgEKw060Jr4bp1l86lNQ==, figureFileBig=7n2YGemXwh2NTb65BuUMSg==, tableContent=null), ArticleFig(id=1261377045871903315, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, language=CN, label=图5, caption=模型预测框架, figureFileSmall=EEjgEKw060Jr4bp1l86lNQ==, figureFileBig=7n2YGemXwh2NTb65BuUMSg==, tableContent=null), ArticleFig(id=1261377047797088860, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, language=EN, label=Fig.6, caption=Plot of raw wind speed data, figureFileSmall=Tc+uFWtwfgcL7C4eRgi71w==, figureFileBig=SxymYLAeCdjO3fJnzA6HSQ==, tableContent=null), ArticleFig(id=1261377048329765471, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, language=CN, label=图6, caption=原始风速数据图, figureFileSmall=Tc+uFWtwfgcL7C4eRgi71w==, figureFileBig=SxymYLAeCdjO3fJnzA6HSQ==, tableContent=null), ArticleFig(id=1261377048736612964, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, language=EN, label=Fig.7, caption=CEEMDAN decomposition chart, figureFileSmall=iwXTZI+5gqFdfRjZ0HqlBg==, figureFileBig=Wz3MLZBM3Gr+iW1AtYU2Iw==, tableContent=null), ArticleFig(id=1261377049558696555, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, language=CN, label=图7, caption=CEEMDAN分解图, figureFileSmall=iwXTZI+5gqFdfRjZ0HqlBg==, figureFileBig=Wz3MLZBM3Gr+iW1AtYU2Iw==, tableContent=null), ArticleFig(id=1261377051693597298, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, language=EN, label=Fig.8, caption=Comparison of the prediction results of different decomposition algorithms, figureFileSmall=bM2Aw0UxZI7gYx5zn2KKBQ==, figureFileBig=ucmN08UtGqYPtjObDDoBnw==, tableContent=null), ArticleFig(id=1261377052486320766, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, language=CN, label=图8, caption=不同分解算法预测结果对比, figureFileSmall=bM2Aw0UxZI7gYx5zn2KKBQ==, figureFileBig=ucmN08UtGqYPtjObDDoBnw==, tableContent=null), ArticleFig(id=1261377053891412624, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, language=EN, label=Fig.9, caption=Comparison of prediction results of different optimization algorithms, figureFileSmall=i4VAijPa+jNcFnN/Gm5hvQ==, figureFileBig=0GrImCuGowu3DFnzYUUG9Q==, tableContent=null), ArticleFig(id=1261377056277971613, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, language=CN, label=图9, caption=不同优化算法预测结果对比, figureFileSmall=i4VAijPa+jNcFnN/Gm5hvQ==, figureFileBig=0GrImCuGowu3DFnzYUUG9Q==, tableContent=null), ArticleFig(id=1261377057028752039, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, language=EN, label=Fig.10, caption=Comparison with other model predictions, figureFileSmall=h4+5saz1R0OzmjMtzZt7cg==, figureFileBig=JKj6d+P7GT7oA9ykJSXvkw==, tableContent=null), ArticleFig(id=1261377057582400172, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, language=CN, label=图10, caption=与其他模型预测结果对比, figureFileSmall=h4+5saz1R0OzmjMtzZt7cg==, figureFileBig=JKj6d+P7GT7oA9ykJSXvkw==, tableContent=null), ArticleFig(id=1261377057985053367, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, language=EN, label=Fig.11, caption=Comparison of predicted results from ablation experiments, figureFileSmall=3TZmYCeGeljy0Fk8iHslMQ==, figureFileBig=RAsxdp9/iL831JWM9Srp4w==, tableContent=null), ArticleFig(id=1261377059138486976, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, language=CN, label=图11, caption=消融实验预测结果对比, figureFileSmall=3TZmYCeGeljy0Fk8iHslMQ==, figureFileBig=RAsxdp9/iL831JWM9Srp4w==, tableContent=null), ArticleFig(id=1261377060946232009, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, language=EN, label=Fig.12, caption=Model prediction error plot, figureFileSmall=3WlY0AbHj0JqcD3AU1DmGg==, figureFileBig=L58xUJrEeRiWBgwMwh6ilQ==, tableContent=null), ArticleFig(id=1261377061734761166, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, language=CN, label=图12, caption=模型预测误差图, figureFileSmall=3WlY0AbHj0JqcD3AU1DmGg==, figureFileBig=L58xUJrEeRiWBgwMwh6ilQ==, tableContent=null), ArticleFig(id=1261377062162580178, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, language=EN, label=Fig.13, caption=Model multi-step prediction error plot, figureFileSmall=kyigtSQyd6yS1ir9DbW7fg==, figureFileBig=AHjHx/6ECLcDr/YX6vJMIQ==, tableContent=null), ArticleFig(id=1261377063492174560, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, language=CN, label=图13, caption=模型多步预测误差图, figureFileSmall=kyigtSQyd6yS1ir9DbW7fg==, figureFileBig=AHjHx/6ECLcDr/YX6vJMIQ==, tableContent=null), ArticleFig(id=1261377065207644902, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, language=EN, label=Table 1, caption=

Raw wind speed series statistics

, figureFileSmall=null, figureFileBig=null, tableContent=
统计指标 原始风速
最大值/
(m·s-1)
最小值/
(m·s-1)
均值/
(m·s-1)
方差 标准差
数值 27.385 0.189 4.274 10.896 3.301
), ArticleFig(id=1261377065463497448, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, language=CN, label=表1, caption=

原始风速序列统计值

, figureFileSmall=null, figureFileBig=null, tableContent=
统计指标 原始风速
最大值/
(m·s-1)
最小值/
(m·s-1)
均值/
(m·s-1)
方差 标准差
数值 27.385 0.189 4.274 10.896 3.301
), ArticleFig(id=1261377065899705073, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, language=EN, label=Table 2, caption=

RIME-CNN hyperparameters network setup

, figureFileSmall=null, figureFileBig=null, tableContent=
超参数 取值
卷积和数量 3
卷积核大小 10
批处理大小 142
学习率 0.05
), ArticleFig(id=1261377066407215867, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, language=CN, label=表2, caption=

RIME-CNN超参数网络设置

, figureFileSmall=null, figureFileBig=null, tableContent=
超参数 取值
卷积和数量 3
卷积核大小 10
批处理大小 142
学习率 0.05
), ArticleFig(id=1261377066805674753, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, language=EN, label=Table 3, caption=

Values of evaluation indicators for each model

, figureFileSmall=null, figureFileBig=null, tableContent=
模型 MSE RMSE MAE R2
Model1:ELM 4.726 2.174 2.428 0.774
Model2:BPNN 4.425 2.104 2.357 0.795
Model3:LSTM 3.956 1.989 2.185 0.816
Model4:BiLSTM 3.639 1.907 2.037 0.859
Model5:BiLSTM-AM 3.472 1.863 1.839 0.905
Model6:CNN-BiLSTM-AM 3.425 1.851 1.769 0.913
Model7:RIME-CNN-BiLSTM-AM 2.957 1.719 1.539 0.929
Model8:EMD-RIME-CNN-BiLSTM-AM 2.472 1.572 1.409 0.936
Model9:EEMD-RIME-CNN-BiLSTM-AM 2.057 1.434 1.339 0.948
Model10:CEEMDAN-WOA-CNN-
BiLSTM-AM
1.992 1.412 1.013 0.962
Model11:CEEMDAN-GWO-CNN-
BiLSTM-AM
1.726 1.314 0.968 0.969
Model12:CEEMDAN-RIME-CNN-
BiLSTM-AM
1.384 1.176 0.888 0.981
), ArticleFig(id=1261377067262853898, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, language=CN, label=表3, caption=

各模型评价指标值

, figureFileSmall=null, figureFileBig=null, tableContent=
模型 MSE RMSE MAE R2
Model1:ELM 4.726 2.174 2.428 0.774
Model2:BPNN 4.425 2.104 2.357 0.795
Model3:LSTM 3.956 1.989 2.185 0.816
Model4:BiLSTM 3.639 1.907 2.037 0.859
Model5:BiLSTM-AM 3.472 1.863 1.839 0.905
Model6:CNN-BiLSTM-AM 3.425 1.851 1.769 0.913
Model7:RIME-CNN-BiLSTM-AM 2.957 1.719 1.539 0.929
Model8:EMD-RIME-CNN-BiLSTM-AM 2.472 1.572 1.409 0.936
Model9:EEMD-RIME-CNN-BiLSTM-AM 2.057 1.434 1.339 0.948
Model10:CEEMDAN-WOA-CNN-
BiLSTM-AM
1.992 1.412 1.013 0.962
Model11:CEEMDAN-GWO-CNN-
BiLSTM-AM
1.726 1.314 0.968 0.969
Model12:CEEMDAN-RIME-CNN-
BiLSTM-AM
1.384 1.176 0.888 0.981
), ArticleFig(id=1261377067531289359, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, language=EN, label=Table 4, caption=

Multi-step prediction evaluation indicator values

, figureFileSmall=null, figureFileBig=null, tableContent=
步长 模型 MSE RMSE MAE R2
Model1:ELM 5.839 2.416 3.426 0.725
Model2:BPNN 5.025 2.242 3.272 0.756
Model3:LSTM 4.792 2.189 2.845 0.774
Model4:BiLSTM 4.425 2.104 2.579 0.784
Model5:BiLSTM-AM 3.886 1.971 2.451 0.793
Model6:CNN-BiLSTM-AM 3.834 1.958 2.280 0.802
2步 Model7:RIME-CNN-BiLSTM-AM 3.431 1.853 1.918 0.826
Model8:EMD-RIME-CNN-
BiLSTM-AM
3.410 1.847 1.703 0.841
Model9:EEMD-RIME-CNN-
BiLSTM-AM
3.267 1.808 1.566 0.873
Model10:CEEMDAN-WOA-
CNN-BiLSTM-AM
3.202 1.789 1.439 0.896
Model11:CEEMDAN-GWO-
CNN-BiLSTM-AM
3.024 1.739 1.380 0.917
Model12:CEEMDAN-RIME-
CNN-BiLSTM-AM
2.305 1.518 1.328 0.935
Model1:ELM 7.629 2.762 4.357 0.611
Model2:BPNN 7.285 2.699 4.026 0.694
Model3:LSTM 6.926 2.632 3.878 0.717
Model4:BiLSTM 6.602 2.569 3.731 0.732
Model5:BiLSTM-AM 6.343 2.519 3.638 0.756
Model6:CNN-BiLSTM-AM 6.174 2.485 3.343 0.762
3步 Model7:RIME-CNN-
BiLSTM-AM
5.782 2.405 2.781 0.799
Model8:EMD-RIME-CNN-
BiLSTM-AM
5.508 2.347 2.697 0.818
Model9:EEMD-RIME-CNN-
BiLSTM-AM
4.526 2.127 2.523 0.827
Model10:CEEMDAN-WOA-
CNN-BiLSTM-AM
3.472 1.863 2.462 0.845
Model11:CEEMDAN-GWO-
CNN-BiLSTM-AM
3.239 1.799 2.409 0.853
Model12:CEEMDAN-RIME-
CNN-BiLSTM-AM
3.116 1.765 2.301 0.874
), ArticleFig(id=1261377067967496981, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1261262693726544554, language=CN, label=表4, caption=

多步预测评价指标值

, figureFileSmall=null, figureFileBig=null, tableContent=
步长 模型 MSE RMSE MAE R2
Model1:ELM 5.839 2.416 3.426 0.725
Model2:BPNN 5.025 2.242 3.272 0.756
Model3:LSTM 4.792 2.189 2.845 0.774
Model4:BiLSTM 4.425 2.104 2.579 0.784
Model5:BiLSTM-AM 3.886 1.971 2.451 0.793
Model6:CNN-BiLSTM-AM 3.834 1.958 2.280 0.802
2步 Model7:RIME-CNN-BiLSTM-AM 3.431 1.853 1.918 0.826
Model8:EMD-RIME-CNN-
BiLSTM-AM
3.410 1.847 1.703 0.841
Model9:EEMD-RIME-CNN-
BiLSTM-AM
3.267 1.808 1.566 0.873
Model10:CEEMDAN-WOA-
CNN-BiLSTM-AM
3.202 1.789 1.439 0.896
Model11:CEEMDAN-GWO-
CNN-BiLSTM-AM
3.024 1.739 1.380 0.917
Model12:CEEMDAN-RIME-
CNN-BiLSTM-AM
2.305 1.518 1.328 0.935
Model1:ELM 7.629 2.762 4.357 0.611
Model2:BPNN 7.285 2.699 4.026 0.694
Model3:LSTM 6.926 2.632 3.878 0.717
Model4:BiLSTM 6.602 2.569 3.731 0.732
Model5:BiLSTM-AM 6.343 2.519 3.638 0.756
Model6:CNN-BiLSTM-AM 6.174 2.485 3.343 0.762
3步 Model7:RIME-CNN-
BiLSTM-AM
5.782 2.405 2.781 0.799
Model8:EMD-RIME-CNN-
BiLSTM-AM
5.508 2.347 2.697 0.818
Model9:EEMD-RIME-CNN-
BiLSTM-AM
4.526 2.127 2.523 0.827
Model10:CEEMDAN-WOA-
CNN-BiLSTM-AM
3.472 1.863 2.462 0.845
Model11:CEEMDAN-GWO-
CNN-BiLSTM-AM
3.239 1.799 2.409 0.853
Model12:CEEMDAN-RIME-
CNN-BiLSTM-AM
3.116 1.765 2.301 0.874
)], 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.1671-1815.2407479, detailUrlEn=https://castjournals.cast.org.cn/joweb/kxjsygc/EN/10.12404/j.issn.1671-1815.2407479, pdfUrlCn=https://castjournals.cast.org.cn/joweb/kxjsygc/CN/PDF/10.12404/j.issn.1671-1815.2407479, pdfUrlEn=https://castjournals.cast.org.cn/joweb/kxjsygc/EN/PDF/10.12404/j.issn.1671-1815.2407479, aliStartDate=null, aliEndDate=null, collectionFlag=false, citedCount=null, citedUrl=null, reference=null)
收藏切换
基于模态分解和RIME-CNN-BiLSTM-AM的风速预测方法
收藏切换
PDF下载
朱婷 1 , 颜七笙 2, *
科学技术与工程 | 论文·电工技术 2025,25(20): 8514-8525
收起
收藏切换
科学技术与工程 | 论文·电工技术 2025, 25(20): 8514-8525
基于模态分解和RIME-CNN-BiLSTM-AM的风速预测方法
全屏
朱婷1 , 颜七笙2, *
作者信息
  • 1 东华理工大学经济与管理学院, 南昌 330000
  • 2 东华理工大学理学院, 南昌 330013
  • 朱婷(2001—),女,汉族,安徽阜阳人,硕士研究生。研究方向:智能计算及应用。E-mail:

通讯作者:

* 颜七笙(1976—),男,汉族,江西临川人,博士,教授。研究方向:智能计算及应用。E-mail:
Wind Speed Prediction Based on Modal Decomposition and RIME-CNN-BiLSTM-AM
Ting ZHU1 , Qi-sheng YAN2, *
Affiliations
  • 1 College of Economic and Management, East China University of Technology, Nanchang 330000, China
  • 2 College of Science, East China University of Technology, Nanchang 330013, China
出版时间: 2025-07-18 doi: 10.12404/j.issn.1671-1815.2407479
文章导航
收藏切换

作为一种清洁的可再生能源,风能在缓解日益严重的能源危机方面充当着重要作用。然而,风速的波动性和随机性给电力系统的稳定运行带来了严峻的挑战。针对该问题,提出一种基于自适应噪声完备集合经验模态分解(complete ensemble empirical mode decomposition with adaptive noise,CEEMDAN)与霜冰优化算法(rime optimization algorithm,RIME)-卷积神经网络(convolutional neural network,CNN)-双向长短期记忆网络(bidirectional long short-term memory network,BiLSTM)-注意力机制(attention mechanism,AM)的短期风速预测组合模型CEEMDAN-RIME-CNN-BiLSTM-AM。首先,对初始风速序列采用CEEMDAN算法,得到一系列较平稳的子模态,以降低风速序列的波动性;然后,采用RIME霜冰优化算法优化CNN超参数,建立CNN-RIME模型,对风速数据进行自适应提取和挖掘;接着,采用BiLSTM-AM模型对处理后的数据进行预测;最后,将各子序列的预测结果叠加,得到最终预测结果。以某地实际风速数据集进行对比试验,该模型在单步与多步预测中均展现出良好的预测性能,可以为制定调度计划提供参考,以最大程度地提高能源利用率和供电。

风速预测  /  自适应噪声完备集合经验模态分解(CEEMDAN)  /  霜冰优化算法(RIME)  /  卷积神经网络(CNN)  /  双向长短期记忆网络(BiLSTM)  /  注意力机制(AM)

Serving as a clean and renewable energy source, wind energy plays a significant role in mitigating the increasingly severe energy crisis. However, the fluctuation and randomness of wind speed pose severe challenges to the stable operation of power systems. To address this issue, a combined short-term wind speed forecasting model named CEEMDAN-RIME-CNN-BiLSTM-AM was proposed, which was based on complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN), rime optimization algorithm (RIME), convolutional neural network (CNN), bidirectional long short-term memory network (BiLSTM), and attention mechanism (AM). Initially, the CEEMDAN algorithm was applied to the original wind speed series to obtain a series of relatively stable sub-modes, thereby reducing the volatility of the wind speed series. Subsequently, the CNN hyperparameters were optimized using the RIME algorithm to establish the CNN-RIME model for adaptive extraction and mining of wind speed data. Then, the BiLSTM-AM model was employed to forecast the processed data. Finally, the forecasting results of each sub-series were superimposed to obtain the final forecasting result. A comparative experiment was conducted using an actual wind speed dataset from a certain location. The proposed model demonstrates good forecasting performance in both single-step and multi-step forecasting, providing a reference for scheduling plans to maximize energy utilization and power supply.

wind speed prediction  /  complete ensemble empirical mode decomposition with adaptive noise(CEEMDAN)  /  rime optimization algorithm (RIME)  /  convolutional neural network (CNN)  /  bidirectional long short-term memory network (BiLSTM)  /  attention mechanism(AM)
朱婷, 颜七笙. 基于模态分解和RIME-CNN-BiLSTM-AM的风速预测方法. 科学技术与工程, 2025 , 25 (20) : 8514 -8525 . DOI: 10.12404/j.issn.1671-1815.2407479
Ting ZHU, Qi-sheng YAN. Wind Speed Prediction Based on Modal Decomposition and RIME-CNN-BiLSTM-AM[J]. Science Technology and Engineering, 2025 , 25 (20) : 8514 -8525 . DOI: 10.12404/j.issn.1671-1815.2407479
近年来,能源短缺、环境恶化等一系列问题让绿色可再生能源成为人们关注的焦点。其中,风能以其高效、纯净、资源丰富等优势得到了快速发展。随着风电在电网中所占比例的不断增加,风电输出的不稳定性给电网的稳定运行带来了挑战,其中风速的随机性和不稳定性是其中的主要因素[1]。因此,准确、稳定的风速预测对高效的电网规划和确保可再生能源的充分接纳起着十分重要的作用[2-3]
中外学者在风速预测领域开展了广泛研究。目前,风速预测的主要方法可归纳为三大类别:统计学模型、人工智能模型以及组合预测模型。
统计学模型包括ARMA(auto regressive moving average)[4]、ARIMA(auto regressive integrated moving average)[5]、Logistic模型[6]等。尽管统计学方法具备结构简洁、建模便利的特点,然而,输入数据的分布特征对模型输出结果产生影响显著。
人工智能模型包括支持向量机(support vector machine,SVM)[7]、随机森林(random forest,RF)[8]等。其中, SVM算法既适用于线性问题也适用于非线性问题,其泛化误差率低,能有效应对传统算法面临的高维数据处理难题,但在面对大规模时间序列数据时,存在收敛速度慢和预测精度下降的问题。随机森林算法通过同时训练多个决策树来提高预测性能,但随机森林算法对噪声数据比较敏感。近年来,深度学习备受瞩目,并被应用于风速预测领域,如长短期记忆网络(long short-term memory,LSTM)[9]、卷积神经网络(convolutional neural network,CNN)[10]和门控循环单元(gated recurrent unit,GRU)[11]等。文献[12]通过将CNN融入预测模型,实现了对数据内部变化规律的空间特征的有效提取。文献[13]选用双向长短期记忆网络(bidirectional long short-term memory,BiLSTM)作为预测框架,该网络凭借其双向信息流结构,能够更全面地捕捉各分量的动态变化,进而提升了预测模型的精确度。文献[14]在预测模型中引入注意力机制(attention mechanism,AM),为双向门控循环单元(bidirectional recurrent neural network,BiGRU)的输出状态分配了不同权重,从而增强了关键特征的输出,实验结果显示,其预测精度相较于BiGRU模型有明显提升。
组合预测模型即将数据分解算法或优化算法与预测模型进行组合,将数据分解为若干个简单分量或对模型重涉及的多个超参数进行优化。文献[15]采用变分模态分解(variational mode decomposition,VMD)对风速序列进行分解并与GRU组成组合预测模型对负荷进行预测,VMD展现出较好的抗噪声能力,然而其分解效果却高度依赖于核心参数的设定,而这些参数的确定往往带有较大的主观判断成分。文献[16]采用自适应噪声完备集合经验模态分解(complete ensemble empirical mode decomposition with adaptive noise,CEEMDAN)算法,将风速序列分解为一系列具有不同特征的模态分量结合相关预测算法,展现出优异的预测效果。文献[17]采用麻雀搜索算法(sparrow search algorithm,SSA)对预测模型隐藏层神经元个数、学习率等参数进行优化,确定模型的最优参数组合。文献[18]采用灰狼优化算法(grey wolf optimizer,GWO)对组合模型的权重参数进行优化,经过实验表明,经过优化后的模型预测性能显著提高。
上述方法均采用信号分解、人工智能模型和优化算法去研究风速的变化特征。但在利用这些方法时均没有深入探究风速变化规律与预测模型之间的深层联系,算法参数设定后固定不变,使得风速预测的准确度存在一定局限性。现采用智能算法主动适应风速变化规律,采用信号分解和深度学习方法深入探究风速的变化规律。
针对风速序列的特点,根据“分解-预测-重构”的研究思路,提出CEEMDAN和霜冰优化算法(rime optimization algorithm,RIME)优化CNN的BiLSTM-AM的组合预测模型。将风速序列采用CEEMDAN算法分解为若干分量,以降低风速的波动性;把影响因素进行归一化处理,消除数据量纲差异。采用RIME算法对CNN的卷积层和的数量N、卷积核的大小K,批处理大小和学习率进行优化以构建RIME-CNN模型,极大优化CNN的特征处理能力。建立BiLSTM-AM模型,将经过特征处理后的的数据输入至BiLSTM-AM模型中训练。BiLSTM可以同时处理过去与未来的时序信息,AM可以区分来自BiLSTM的时间序列信息的重要性,模型可以聚焦数据的关键部分,提取最深层的时间相关性。最后,将各子序列的预测结果进行叠加,得出最终的预测值。将所提模型分别进行不同分解算法的对比、不同化算法的对比、其他模型的对比、消融实验及多步预测实验。实验结果表明,所提模型具有较好的预测效果,可为风电并网稳定运行提供可靠信息。
CEEMDAN是一种改进的数据分解算法[19],通过自适应的添加高斯白噪声,使得在实验次数较少的情况下对信号进行分解重构,得到若干个IMF分量,CEEMDAN不仅在很大程度上消除了EMD和EEMD的模态混叠和噪声残差大等问题,而且对于处理非线性和非平稳信号分解效果很好。具体步骤如下。
步骤1 在第M次实验中,运用式(5)对原始信号xt进行分解,取平均后得到第一个IMF分量和第一个残差信号r1(t)。
$IM{\mathrm{F}}_{1}=\frac{1}{M}\stackrel{M}{\sum _{i=1}}\mathrm{I}\mathrm{M}{\mathrm{F}}_{1i}$
式(1)中:$\mathrm{I}\mathrm{M}{\mathrm{F}}_{1}$M次实验取平均得到的第一个分量;M为实验次数;$\mathrm{I}\mathrm{M}{\mathrm{F}}_{1i}$i次实验得到的IMF分量。
${r}_{1}\left(t\right)=x\left(t\right)-\mathrm{I}\mathrm{M}{\mathrm{F}}_{1}\left(t\right)$
步骤2 在残差信号中加入自适应噪声${\epsilon }_{1}{E}_{1}\left[{\omega }_{i}\right(t\left)\right],\mathrm{其}\mathrm{中},{\epsilon }_{1}$为高斯白噪声${\omega }_{i}\left(t\right)$的信噪比,E1(·)为分解后第i个IMF分量的算子,则下一个IMF分量可表示为
$IM{\mathrm{F}}_{2}=\frac{1}{M}\stackrel{M}{\sum _{i=1}}{E}_{1}\left\{{r}_{1}\right(t)+{\epsilon }_{1}{E}_{1}[{\omega }_{i}\left(t\right)\left]\right\}$
步骤3 对剩下的残差信号重复以上步骤,则第n个IMF分量可表示为
$IM{\mathrm{F}}_{n}=\frac{1}{M}\stackrel{M}{\sum _{i=1}}{E}_{1}\left\{{r}_{n}\right(t)+{\epsilon }_{n}{E}_{n}[{\omega }_{i}\left(t\right)\left]\right\}$
式(4)中:rn(t)为第n个分量的残差;${\epsilon }_{n}$为第n个分量中的信噪比;En(·)为分解后第n个IMF分量的算子。
步骤4 重复上述步骤,直到残差信号不能再被分解为止。假设最后分解出N个IMF分量,则最终原始信号可表示为
$\mathrm{x}\left(\mathrm{t}\right)=\stackrel{N}{\sum _{i=1}}\mathrm{I}\mathrm{M}{\mathrm{F}}_{n}+\stackrel{M}{\sum _{i=1}}{r}_{i}\left(t\right)$
RIME是Su等[20]于2023年提出的算法,该算法通过模拟软霜冰粒子的运动,提出一种用于算法搜索的霜冰搜索策略。RIME算法基于物理的优化器,利用模拟霜冰中软冰针和硬冰针生长过程,来实现优化方法的探索和开发行为,并通过改进的正向贪婪选择机制进行全局寻优。Su等[20]将该算法划分为初始化霜冰结晶点、引入软霜冰搜索策略和硬霜冰穿刺机制,并改进贪婪选择策略来建立数学模型。图1为RIME算法流程图。
RIME霜冰粒子种群R位置可表示为
$R=\left[\begin{array}{llll}{x}_{11}& {x}_{12}& \dots & {x}_{1j}\\ {x}_{21}& {x}_{22}& \dots & {x}_{2j}\\ ︙& ︙& \mathrm{ }& ︙\\ {x}_{i1}& {x}_{i2}& \dots & {x}_{ij}\end{array}\right]$
式(6)中:xij为霜冰粒子种群,其中,i为霜冰晶体的序号,j为霜冰颗粒的序号。
所有霜冰粒子的适应度值F(Si)可表示为
$F\left({S}_{i}\right)=\left[\begin{array}{llll}\mathrm{f}\left(\right[{x}_{11}& {x}_{12}& \dots & {x}_{1j}\left]\right)\\ \mathrm{f}\left(\right[{x}_{21}& {x}_{22}& \dots & {x}_{2j}\left]\right)\\ \mathrm{ }& \mathrm{ }& ︙ & \\ \mathrm{f}\left(\right[{x}_{i1}& {x}_{i2}& \dots & {x}_{ij}\left]\right)\end{array}\right]$
式(7)中:f为每个霜冰粒子的适应度。
每个霜冰粒子xij在凝结之前会按照一定规律运动,自由状态的粒子则会移动到软冰附近凝结,但当超过一定的范围则无法凝结,在软冰形成的过程中每个粒子的随机性会导致凝结面积增大,受环境影响最终达到一个稳定状态。
针对霜冰粒子的运动性质,霜冰粒子的位置可表示为
${R}_{ij}^{\mathrm{n}\mathrm{e}\mathrm{w}}$=Rbest,j+r1cos(θ)β[ψ(Ubij-Lbij)+Lbij)]

r2<E

式中:${R}_{ij}^{\mathrm{n}\mathrm{e}\mathrm{w}}$为粒子更新后的位置;${R}_{\mathrm{b}\mathrm{e}\mathrm{s}\mathrm{t},j}$为霜冰种群R中最优晶体的第j个粒子;r1为(-1,1)的随机数,控制粒子的移动方向,并且会随着粒子移动方向$\mathrm{c}\mathrm{o}\mathrm{s}\theta $迭代发生变化;$\beta $为环境因子,表示外部的影响因素;粒子附着力$\psi $为(0,1)的随机数,用于控制粒子间的距离;${U}_{\mathrm{b}ij}、{L}_{\mathrm{b}ij}$分别为粒子运动空间的上界和下界;E为附着系数,会随迭代的次数增大,影响粒子的凝结概率;r2为(0,1)的随机数,它与E控制粒子是否凝结,即粒子位置是否更新。
为了确保收敛需设定可表示为
$\theta =\mathrm{\pi }\frac{h}{10H}$
式(10)中:h为当前迭代次数;H为最大迭代次数。
$\beta =1-\frac{\left(\frac{wh}{H}\right)}{w}$
式(11)中:$\beta $为阶跃函数;w为用于控制步进函数的段数;
E的表达式为
$E=\sqrt{\frac{h}{H}}$
随软冰凝结面积增大,产生强随机性和覆盖范围广的特点,让算法能够快速覆盖全空间的搜索,从而在优化过程中兼顾了全局性和局部性。硬冰受外界因素影响会向着同一方向凝结,由于硬冰的增长方向一致,导致硬冰之间很容易交叉,实现普通粒子和最优粒子的维度交换,有利于提高解的精度。由于在迭代的过程中,粒子之间的交叉互换,避免了局部最优,提高了收敛性。随着硬冰的生长穿刺,粒子之间的交叉替换可表示为
${R}_{ij}^{\mathrm{n}\mathrm{e}\mathrm{w}}={R}_{\mathrm{b}\mathrm{e}\mathrm{s}\mathrm{t},j},{r}_{3}<{F}^{\mathrm{n}\mathrm{o}\mathrm{r}\mathrm{m}\mathrm{r}}\left({S}_{i}\right)$
式(13)中:${F}^{\mathrm{n}\mathrm{o}\mathrm{r}\mathrm{m}\mathrm{r}}\left({S}_{i}\right)$为当前适应度值归一化的值,表示第i个粒子被选中的概率;r3为(-1,1)的随机数。
作为深度学习领域最广泛的算法,CNN因其卓越的特征提取性能而被广泛应用于能源系统[21]。具体来说,CNN主要由卷积层和池化层组成,卷积层通过应用卷积核对风速数据中的非线性特征进行精确的局部提取,有效捕捉数据中的核信息。池化层对提取的特征进行压缩和过滤,以提炼出更重要的信息,这不仅提高了模型处理数据的效率,也显著增强了其泛化能力,为风电系统研究提供了强大的技术支持。CNN的基本结构如图2所示。
BiLSTM通过集成前向LSTM层和后向LSTM层,对传统单向LSTM进行增强,前向LSTM层和后向LSTM层都有助于整体输出[22]。虽然单一单向LSTM利用了历史风速数据,但值得注意的是,它在预测中缺乏对未来信息的考虑[23]。BiLSTM不仅考虑了序列的前向输入,还考虑了后向输入,即同时利用了序列的过去和未来信息。这种双向的处理方式有助于模型更全面地利用数据信息,从而进一步提高预测精度。BiLSTM结构如图3所示。
${\overrightarrow{h}}_{t}=\mathrm{L}\mathrm{S}\mathrm{T}\mathrm{M}({x}_{t},{\overrightarrow{h}}_{t-1})$
${\overleftarrow{h}}_{t}=\mathrm{L}\mathrm{S}\mathrm{T}\mathrm{M}({x}_{t},{\overleftarrow{h}}_{t-1})$
${y}_{t}=\sigma \left({W}_{y}\right[{\overrightarrow{h}}_{t},{\overleftarrow{h}}_{t}]+{b}_{y})$
式中:LSTM单元为传统LSTM网络的计算过程;σ为Sigmoid函数;${\overrightarrow{h}}_{t}$t时刻的正向LSTM隐藏层状态;${\overleftarrow{h}}_{t}$t时刻的向后LSTM隐藏层状态;Wyby分别为权重矩阵和偏差项;${x}_{1},{x}_{2},\dots,{x}_{t}$${t}_{1},{t}_{2},\dots,{t}_{i}$各时刻对应的输入数据,$i\in \left\{\mathrm{1,2},\dots,t\right\};{y}_{1},{y}_{2},\dots,{y}_{t}$${t}_{1},{t}_{2},\dots,{t}_{i}$各时刻对应的输出数据,$i\in \{\mathrm{1,2},\dots,t\}。$
注意力机制源于对人脑注意力特性的模拟,最早用于图像处理领域[24]。在深度学习中,注意力机制将对预测起关键作用的信息分配较大的权重,不重要的信息赋予较小的权重,通过该机制来优化信息处理的效率。注意单元的结构如图4所示。注意状态转换过程[25]如式(17)~式(20)所示。在深度学习中,注意力机制能够自适应地调整不同信息的权重,从而增强模型在特征挖掘方面的能力。通过将注意力机制融入BiLSTM网络,能够更精准地聚焦于关键特征,同时降低或忽略次要特征的影响。这不仅提高了模型的预测效率和准确性,还增强了其应对信息过载的能力。

Sti=Vtanh(WHt+Uhi+b), i=1,2,…,t-1

${a}_{ti}=\frac{exp{S}_{ti}}{\stackrel{t-1}{\sum _{i=1}}\mathrm{e}\mathrm{x}\mathrm{p}{S}_{ti}}, i=\mathrm{1,2},\dots,t-1$
$\mathrm{F}=\stackrel{t}{\sum _{i=1}}{a}_{ti}{h}_{i}, i=\mathrm{1,2},\dots,t-1$
$h\text{'}{\mathrm{ }}_{t}=f(F,{H}_{t})$
式中:Sti为利用tanh双曲正切激活函数计算时间步的隐藏层状态对当前时刻输出的重要性得分;${y}_{1},{y}_{2},\dots,{y}_{t}$为输入序列;${h}_{1},{h}_{2},\dots,{h}_{t}$为输入序列对应的BiLSTM隐藏层输出的状态值;Htt时刻CNN-BiLSTM输出权重矩阵;ati为BiLSTM隐藏层输出的状态值对应的当前注意权值;F为模型输出结果;h' t为最终的特征向量;f(·)为h' t的计算过程;VWUb为模型的学习参数。
该模型由CEEMDANRIME-CNNBiLSTM-AM模型组成。首先,采用CEEMDAN分解算法对原始风速数据进行分解;其次,为了提高CNN模型的性能,使用RIME寻找CNN模型的最优超参数,超参数包括卷积核数量N、卷积核大小K、批量大小和学习率;然后,使用BiLSTM-AM模型训练从CNN提取的特征,BiLSTM的隐藏层可以学习并捕捉从CNN提取的局部特征的内部变化规律,并通过迭代过程从局部特征中提取出更复杂、更完整的全局特征,以提高信息的有效利用率。AM可以区分来自BiLSTM隐藏层的时间序列信息的重要性,模型可以聚焦数据的关键部分,提取最深层的时间相关性。最后,全连接层提供最终预测结果。CEEMDAN-RIME-CNN-BiLSTM-AM模型的框架如图5所示。具体步骤如下。
步骤1 将海平面气压、温度、相对湿度、平均风等气象因素进行归一化处理。
步骤2 采用CEEMDAN分解方法对风速数据进行分解,得到若干个分量,在保留风速原始特征的同时,最大限度地降低预测精度。
步骤3 采用RIME优化算法对CNN的超参数进行优化,有效提取风速数据中的局部特征;构建BiLSTM-AM模型对处理后的数据进行预测。
步骤4 将各子序列的预测结果进行叠加,得到最终的预测结果,并将此模型与其他模型进行对比分析,采用误差评估指标来衡量模型的预测精确度。
为验证模型的有效性和科学性,风速数据集来源于国家可再生能源实验室(National Renewable Energy Labortory,NREL),采用美国科罗拉多州戈尔登地区2023年10月1日—12月31日的风速数据集,采样间隔为1h,数据包括温度(单位:℃)、相对湿度(单位:%)、海平面气压(单位:Pa)、平均风向[单位:(°)]、平均风速(单位:m/s)5维数据,共2 161个数据。取80%的数据作为训练集,20%的数据作为测试集,预测方法为单步滚动预测。
图6表1可以看出,风速变化无明显规律,最大风速为27.385 m/s,最小风速仅为0.189 m/s,风速变化范围较大,其无规律的变化使得对风速的预测变得较为困难。
表1中原始风速序列的统计指标值可知,平均风速为4.274 m/s,从风速数据的方差与标准差分析来看,其离散程度较高,表明风速具有较强的波动性和不稳定性的特点。
首先需要对原始风速数据进行CEEMDAN分解,将非平稳、非线性的风速数据分解成多个较为平稳的分量,提高风速预测精度。CEEMDAN分解图如图7所示。
图7可以看出,原始风速序列分被分解为8个IMF模态分量和1个残差分量,将模态分量作为输入变量输入到预测模型中,可以有效地提取风速信号的重要信息,减小风速序列的随机波动性。然后,需要对气象因素进行归一化处理,归一化公式为
${x}^{\mathrm{*}}=\frac{x-{x}_{\mathrm{m}\mathrm{i}\mathrm{n}}}{{x}_{\mathrm{m}\mathrm{a}\mathrm{x}}-{x}_{\mathrm{m}\mathrm{i}\mathrm{n}}}$
式(21)中:x*为归一化后的数据;x为原始数据;xmax为原始数据中的最大值;xmin为原始数据中的最小值。
本实验基于Pytorch框架实现,并且在Python3.9、CPU为Intel(R) Core(TM) i5 13500H,显卡为Intel Iris Xe Graphics,内存为16 G 5200 MHz的环境中进行训练和预测。
在模型训练过程中,各网络的参数对模型训练结果有很大影响结果。通过多次训练对比,BiLSTM网络参数设置如下:隐藏层数目为1,隐藏层神经元数目为64,学习率为0.003,最大迭代次数为50。经过RIME优化之后的CNN结构为:卷积核的数量N为3、卷积核的大小K为10,批处理大小为142,学习率为0.05。具体RIME-CNN的超参数设置信息如表2所示。
为了有效比较和分析的预测模型,选取均方误差(mean squared error,MSE)、均方根误差(root mean square error,RMSE)、平均绝对误差(mean absolute error,MAE)、决定系数R2作为模型的评价标准,评估本文模型及其他对比模型的预测性能。其中,MSE、RMSE、MAE的结果越小,表示模型的预测误差越小,R2越大,表示模型的精确度越高。这4项指标的计算公式分别为
$MSE=\frac{1}{n}\stackrel{n}{\sum _{i=1}}|{Y}_{i}-Y\text{'}{\mathrm{ }}_{i}{|}^{2}$
$\mathrm{R}\mathrm{M}\mathrm{S}\mathrm{E}=\sqrt{\frac{1}{n}\stackrel{n}{\sum _{i=1}}|{Y}_{i}-Y\text{'}{\mathrm{ }}_{i}{|}^{2}}$
$MAE=\frac{1}{n}\stackrel{n}{\sum _{i=1}}|{Y}_{i}-Y\text{'}{\mathrm{ }}_{i}|$
$R^{2}=\frac{\left[\sum_{i=1}^{n}\left(Y_{i}-\bar{Y}_{i}\right)\left(Y_{i}^{\prime}-\bar{Y}^{\prime}\right)\right]^{2}}{\left[\sum_{i=1}^{n}\left(Y_{i}-\bar{Y}_{i}\right)^{2}\right]\left[\sum_{i=1}^{n}\left(Y_{i}^{\prime}-\bar{Y}^{\prime}\right)^{2}\right]}$
式中:n为样本数;Yi为实际值;Y' i为预测值,${\stackrel{—}{Y}}_{i}$为实际平均值;$\stackrel{—}{Y}\text{'}$为预测平均值。
为了验证CEEMDAN在数据分解任务中的效能,构建CEEMDAN-RIME-CNN-BiLSTM-AM模型,并将其预测性能与EMD-RIME-CNN-BiLSTM-AM及EEMD-RIME-CNN-BiLSTM-AM这两种不同的分解算法进行对比分析。图8展示了这3种分解模型所得预测结果的对比情况。
图8可以看出,与EMD-RIME-CNN-BiLSTM-AM和EEMD-RIME-CNN-BiLSTM-AM模型相比,本文模型预测性能最好,预测误差最小,说明CEEMDAN分解算法通过解决EMD和EEMD中存在的模态混叠现象以及它们残留的噪声问题,有效提升了预测的准确性。
为了评估RIME冰霜优化算法优异性,分别运用GWO和WOA对CNN的4个关键超参数进行了优化。随后,将这两种优化算法与所提出的RIME冰霜优化算法的预测性能进行了对比分析。图9展示了这3种不同优化算法所得模型的预测结果。
图9可以看出,经过GWO、WOA、RIME算法优化后的预测模型都能够很好地拟合出风速的大致趋向,但相较于GWO和WOA算法,RIME算法展现出更加优秀的预测效果。
为进一步验证所提模型CEEMDAN-CNN-BiLSTM-AM模型对风速预测的准确性,将BiLSTM、LSTM、BPNN和ELM模型与其进行对比,具体预测结果如图10所示。
图10可以看出,仅采用单一模型进行预测会受到数据观测视角局限性的影响。相比之下,CEEMDAN-RIME-CNN-BiLSTM-AM组合模型在追踪风速变化趋势上表现更佳,且细节上更接近实际值。通过对比部分样本点(70~100)的预测结果,可以观察到样本点70~75属于风速平稳阶段,在该阶段模型的预测性能普遍要优于风速剧烈变化阶段。与单一模型相比,CEEMDAN-RIME-CNN-BiLSTM-AM组合模型在风速波动条件下仍保持了良好的预测效果,这表明组合模型相较于单一模型更能适应复杂多变的气象条件,不仅提升了预测精度,还增强了模型的普遍适用性。
为验证所提出组合模型的有效性,将所提出模型CEEMDAN-RIME-CNN-BiLSTM-AM与RIME-CNN-BiLSTM-AM、CNN-BiLSTM-AM、BiLSTM-AM和BiLSTM模型进行对比,预测结果如图11所示。
图11为CEEMDAN-RIME-CNN-BiLSTM-AM、RIME-CNN-BiLSTM-AM、CNN-BiLSTM-AM、BiLSTM-AM和BiLSTM模型的预测图,可以看出,5种模型都能够大致拟合出风速的大致趋向,但RIME-CNN-BiLSTM-AM、CNN-BiLSTM-AM、BiLSTM-AM和BiLSTM模型在波峰与波谷区域的预测效果不是很理想,CEEMDAN-RIME-CNN-BiLSTM-AM模型无论在波峰还是波谷区域都能很好地捕捉到风速的变化规律,能更好地拟合出真实值,预测误差更小。
为了更准确地展示各模型的预测性能,采用MSE、RMSE、MAE和R2评价指标来表示各模型误差值,各模型预测误差评价指标值如表3图12所示。
表3图12可知,相较于BiLSTM模型,BiLSTM-AM模型通过引入注意力机制,实现了对输入信息权重的差异化分配,使关键信息获得更大权重,减少了冗余信息的干扰,使得MSE下降了4.589%,RMSE降低了2.307%,MAE减少了9.720%,R2提升了5.355%,这表明引入注意力机制的有效性。与BiLSTM-AM模型相比,CNN-BiLSTM-AM模有效捕捉了风速数据中的局部特征,进一步降低了预测误差,MSE下降了1.354%,RMSE减少了0.644%,MAE下降了3.801%,R2增加了0.884%。相较于CNN-BiLSTM-AM模型,采用RIME算法优化CNN超参数的RIME-CNN-BiLSTM-AM模型展现出了更好的预测性能,MSE降低了13.664%,RMSE减少了7.131%,MAE下降了13.001%,R2增加了1.752%,这充分说明了RIME算法在提高预测精度方面的优势。
相较于EMD-RIME-CNN-BiLSTM-AM与EEMD-RIME-CNN-BiLSTM-AM模型,CEEMDAN-RIME-CNN-BiLSTM-AM模型展现出更优的预测性能,这证明了CEEMDAN在数据分解方面中的显著优势。
CEEMDAM-RIME-CNN-BiLSTM-AM模型的MSE为 1.384,RMSE为1.176,MAE为0.888,R2为0.981,相较于BiLSTM、LSTM、BPNN、ELM模型,MSE降低了61.968%、65.014%、68.724%和70.715%;RMSE分别下降了38.332%、40.873%、44.093%和45.906%;MAE分别下降56.406%、59.359%、62.325%和63.427%;R2分别增长14.203%、20.221%、23.396%和26.744%。这说明所提模型相较于其他单一模型有更高的预测精度和模型通用性。
为了验证本文方法的稳健性和普遍适用性,在单步预测的基础上拓展了多步预测实验,多部预测实验能够更好地反映出模型对时序信息更深层次的据挖掘能力。该实验重点实施了两步及三步预测,相关的评价指标结果如表4图13所示。
表4图13的对比结果可知,随着预测步数的增加,预测误差呈现逐步上升的趋势,原因在于随时间推进,风速分量与历史数据间的关联性减弱,导致有效的时序特征难以被有效捕捉,从而使得多步预测相较于单步预测更具挑战性。在多步预测实验中,所提出的模型依然展现出卓越的性能,各项评估指标均优于其他模型,这表明CEEMDAN-RIME-CNN-BiLSTM-AM模型具备高度的稳健性。
综上所述,无论是进行单步预测还是多步预测,将信号分解技术与深度学习相结合的方法都能高效地模拟风速的波动特性,从而弥补单一预测算法的局限性,进而提升模型的预测精确度。
提出一种基于CEEMDAN-RIME-CNN-BiLSTM-AM的短期风速预测方法。通过仿真实验分析,得出以下结论。
(1)针对波动性强且具有随机性风速数据预测的难题,采用了CEEMDAN算法对原始风速数据进行分解处理,以获得更为平稳的分量,这一方法能够高效地提取风速信号中的关键信息,显著降低风速序列的随机波动性。相较于传统的EMD和EEMD算法,CEEMDAN算法在解决模态混叠现象及减少噪声残差方面展现出了更优的性能。
(2)通过霜冰优化算法RIME自动寻找 CNN模型超参数的最佳组合,减少手动调参的繁琐过程,从而提高了模型的预测精度,使模型的性能达到最优。
(3)采用单步以及多步实验对不同模型展开多组实验,实验结果表明,所提出的组合模型均呈现出优异的预测性能。
  • 国家自然科学基金(71961001)
参考文献 引证文献
排序方式:
[1]
刘向龙, 蔡文斌, 于源, . 基于系统调度成本和储能经济收益规划的“源-网-储”协同互动优化调度策略[J]. 科学技术与工程, 2024, 24(35): 15065-15073.
Liu Xianglong, Cai Wenbin, Yu Yuan, et al. “Source-network-storage” collaborative and interactive optimal dispatch strategy based onsys tem dispatch cost and energy storage economic benefit objectives[J]. Science Technology and Engineering, 2024, 24(35): 15065-15073.
[2]
郑晓亮, 杨晓亮, 来文豪. 新能源发电预测方法分类及研究进展[J]. 科学技术与工程, 2024, 24 (34): 14503-14521.
Zheng Xiaoliang, Yang Xiaoliang, Lai Wenhao. Classification and research progress of new energy power generation forecasting methods[J]. Science Technology and Engineering, 2024, 24(34): 14503-14521.
[3]
Wang H, Han S, Liu Y, et al. Sequence transfer correction algorithm for numerical weather prediction wind speed and its application in a wind power forecasting system[J]. Applied Energy, 2019, 237: 1-10.
[4]
张金良, 刘子毅. 基于混合模型的超短期风速区间预测[J]. 电力系统保护与控制, 2022, 50(22): 49-58.
Zhang Jinliang, Liu Ziyi. Ultra short term wind speed interval prediction based on a hybrid model[J]. Power System Protection and Control, 2022, 50(22): 49-58.
[5]
蒋金良, 林广明. 基于ARIMA模型的自动站风速预测[J]. 控制理论与应用, 2008(2): 374-376.
Jiang Jinliang, Lin Guangming. Automatic station wind speed forecasting based on ARIMA model[J]. Control Theory & Applications, 2008(2): 374-376.
[6]
余泳涛, 张扬. 基于改进Logistic模型法的配电网负荷预测[J]. 南昌大学学报(工科版), 2022, 44 (2): 200-204.
Yu Yongtao, Zhang Yang. Distribution network load forecasting based on PSO-Logistic model method[J]. Journal of Nanchang University(Engineering & Technology), 2022, 44(2): 200-204.
[7]
Wu C Y, Wang J Z, Chen X J, et al. A novel hybrid system based on multi-objective optimization for wind speed forecasting[J]. Renewable Energy, 2020, 146: 149-165.
[8]
Ibrahim I A, HossainM J, Duck B C. An optimized offline random forests based model for ultra-short-term prediction of PV characteristics[J]. IEEE Transactions on Industrial Informatics, 2019, 16(1): 202-214.
[9]
王愈轩, 刘尔佳, 黄永章. 基于CNN-LSTM-lightGBM组合的超短期风电功率预测方法[J]. 科学技术与工程, 2022, 22(36): 16067-16074.
Wang Yuxuan, Liu Erjia, Huang Yongzhang. An ultra-short-term wind power prediction method based on CNN-LSTM-lightGBM combination[J]. Science Technology and Engineering, 2022, 22(36): 16067-16074.
[10]
Dong Y X, Wang J, Xiao L, et al. Short-term wind speed time series forecasting based on a hybridmethod with multiple objective optimization for non-convex target[J]. Energy, 2021, 215:119180.
[11]
向玲, 李京蓄, 王朋鹤, . 基于VMD-FIG和参数优化GRU的风速多步区间预测[J]. 太阳能学报, 2021, 42(10): 237-242.
Xiang Ling, Li Jingxu, Wang Penghe, et al. Wind speed multistep interval forecasting based on VMD-FIG and parameter optimized GRU[J]. Acta Energiae Solaris Sinica, 2021, 42(10): 237-242.
[12]
朱彦民, 李忠虎, 杨立清, . 基于卷积双向长短时记忆模型的风电机组故障预测[J]. 科学技术与工程, 2024, 24(25): 10790-10797.
Zhu Yanmin, Li Zhonghu, Yang Liqing, et al. Fault prediction of wind turbine based on convolutional bidirectional long short term memory model[J]. Science Technology and Engineering, 2024, 24(25): 10790-10797.
[13]
毕贵红, 黄泽, 赵四洪, . 基于混合分解和PCG-BiLSTM的风速短期预测[J]. 太阳能学报, 2024, 45(1): 159-170.
Bi Guihong, Huang Ze, Zhao Sihong, et al. Short-term prediction of wind speed based on hybird decomposition and PCG-BiLSTM[J]. Acta Energiae Solaris Sinica, 2024, 45(1): 159-170.
[14]
沈建良, 来骏, 张翼, . 基于ACMD和BiGRU-Attention的负荷预测模型研究[J]. 浙江电力, 2023, 42(6): 70-77.
Shen Jianliang, Lai Jun, Zhang Yi, et al. Research on a load forecasting model based on ACMD and BiGRU-Attention[J]. Zhejiang Electric Power, 2023, 42(6): 70-77.
[15]
段雪滢, 李小腾, 陈文洁. 基于改进粒子群优化算法的VMD-GRU短期电力负荷预测[J]. 电工电能新技术, 2022, 41(5): 8-17.
Duan Xueying, Li Xiaoteng, Chen Wenjie. Improved particles swarm optimization algorithm-based VMD-GRU method for short-term load forecasting[J]. Advanced Technology of Electrical Engineering and Energy, 2022, 41(5): 8-17.
[16]
王贺, 陈蕻峰, 熊敏, . 融合CEEMDAN和ICS-LSTM的短期风速预测建模[J]. 电子测量与仪器学报, 2022, 36(4): 17-23.
Wang He, Chen Honngfeng, Xiong Min, et al. Short-term wind speed forecasting modeling integrating CEEMDAN and ICS-LSTM[J]. Journal of Electronic Measurement and Instrumentation, 2022, 36(4): 17-23.
[17]
刘可真, 阮俊枭, 赵现平, . 基于麻雀搜索优化的Attention-GRU短期负荷预测方法[J]. 电力系统及其自动化学报, 2022, 34(4): 99-106.
Liu Kezhen, Ruan Junxiao, Zhao Xianping, et al. Short-term load forecasting method based on sparrow search optimized Attention-GRU[J]. Proceedings of the CSU-EPSA, 2022, 34(4): 99-106.
[18]
程杰, 陈鼎, 李春, . 基于GWO-CNN-BiLSTM的超短期风电预测[J]. 科学技术与工程, 2023, 23(35): 15091-15099.
Cheng Jie, Chen Ding, Li Chun, et al. Ultra-short-term wind power prediction based on GWO-CNN-BiLSTM[J]. Science Technology and Engineering, 2023, 23(35): 15091-15099.
[19]
韩宏志, 唐振浩. 基于CEEMDAN与回声状态网络的风速预测方法[J]. 电力系统保护与控制, 2020, 48(12): 90-96.
Han Hongzhi, Tang Zhenhao. Wind speed prediction method based on CEEMDAN and echo state network[J]. Power System Protection and Control, 2020, 48(12): 90-96.
[20]
Su H, Zhao D, Heidari A A, et al. RIME: A physics-based optimization[J]. Neurocomputing, 2023, 532: 183-214.
[21]
Li K, Mu Y, Yang F, et al. A novel short-term multi-energy load forecasting method for integrated energy system based on feature separation-fusion technology and improved CNN[J]. Applied Energy, 2023, 351: 121823.
[22]
Zhang D, Chen B, Zhu H, et al. Short-term wind power prediction based on two-layer decomposition and BiTCN-BiLSTM-attention model[J]. Energy, 2023, 285: 128762.
[23]
彭华, 王文超, 朱永利, . 基于LSTM神经网络的风电场集电线路单相接地智能测距[J]. 电力系统保护与控制, 2021, 49(16): 60-66.
Peng Hua, Wang Wenchao, Zhu Yongli, et al. An intelligent single-phase grounding fault location for a wind farm collection line based on an LSTM neural network[J]. Power System Protection and Control, 2021, 49(16): 60-66.
[24]
Mnih V, Heess N, Graves A. Recurrent models of visual attention[J]. Advances in Neural Information Processing Systems, 2014, 27: DOI: 10.48550/arXiv.1406.6247.
[25]
杨国清, 刘世林, 王德意, . 基于Attention-GRU风速修正和Stacking的短期风电功率预测[J]. 太阳能学报, 2022, 43(12): 273-281.
Yang Guoqing, Liu Shilin, Wang Deyi et al. Short-term wind power forecasting based on Attention-GRU wind speed correction and Stacking[J]. Acta Energiae Solaris Sinica, 2022, 43(12): 273-281.
2025年第25卷第20期
PDF下载
51
20
引用本文
BibTeX
文章信息
doi: 10.12404/j.issn.1671-1815.2407479
  • 接收时间:2024-10-10
  • 首发时间:2026-05-13
  • 出版时间:2025-07-18
补充材料
相关文章
文章信息
作者
出版历史
  • 收稿日期:2024-10-10
  • 修回日期:2025-04-28
基金
国家自然科学基金(71961001)
作者信息
    1 东华理工大学经济与管理学院, 南昌 330000
    2 东华理工大学理学院, 南昌 330013

通讯作者:

* 颜七笙(1976—),男,汉族,江西临川人,博士,教授。研究方向:智能计算及应用。E-mail:
参考文献
分享链接
https://castjournals.cast.org.cn/joweb/kxjsygc/CN/10.12404/j.issn.1671-1815.2407479
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