基于WGAN-GP和CNN-LSTM-Attention的短期光伏功率预测
针对非晴天天气类型历史数据量匮乏导致光伏功率预测精度低的问题,提出了一种含有梯度惩罚的改进生成对抗网络(Wasserstein generative adversarial network with gradient penalty,WGAN-GP)和CNN-LSTM-Attention光伏功率短期预测模型.首先,利用K-means++聚类算法将历史光伏数据划分为若干天气类型,使用WGAN-GP生成符合各天气类型数据分布规律的高质量新样本,实现训练集数据增强.其次,结合卷积神经网络(convolutional neural network,CNN)在特征提取上的优势和长短期记忆网络(long...
Saved in:
| Published in | 电力系统保护与控制 Vol. 51; no. 9; pp. 108 - 118 |
|---|---|
| Main Authors | , , , , , |
| Format | Journal Article |
| Language | Chinese |
| Published |
新疆大学电气工程学院,新疆 乌鲁木齐 830049%国网综合能源服务集团有限公司,北京 100053
01.05.2023
|
| Subjects | |
| Online Access | Get full text |
| ISSN | 1674-3415 |
| DOI | 10.19783/j.cnki.pspc.221287 |
Cover
| Abstract | 针对非晴天天气类型历史数据量匮乏导致光伏功率预测精度低的问题,提出了一种含有梯度惩罚的改进生成对抗网络(Wasserstein generative adversarial network with gradient penalty,WGAN-GP)和CNN-LSTM-Attention光伏功率短期预测模型.首先,利用K-means++聚类算法将历史光伏数据划分为若干天气类型,使用WGAN-GP生成符合各天气类型数据分布规律的高质量新样本,实现训练集数据增强.其次,结合卷积神经网络(convolutional neural network,CNN)在特征提取上的优势和长短期记忆网络(long short-term memory,LSTM)在时间序列预测上的优势,提升预测模型学习光伏功率与气象数据间长期映射关系的能力.此外,引入注意力机制(Attention)弥补输入序列长时LSTM难以保留关键信息的不足.实验结果表明:基于WGAN-GP对各类型天气样本扩充能有效提高预测精度;与3种经典预测模型相比,所提出的CNN-LSTM-Attention模型具有更高的预测精度. |
|---|---|
| AbstractList | 针对非晴天天气类型历史数据量匮乏导致光伏功率预测精度低的问题,提出了一种含有梯度惩罚的改进生成对抗网络(Wasserstein generative adversarial network with gradient penalty,WGAN-GP)和CNN-LSTM-Attention光伏功率短期预测模型.首先,利用K-means++聚类算法将历史光伏数据划分为若干天气类型,使用WGAN-GP生成符合各天气类型数据分布规律的高质量新样本,实现训练集数据增强.其次,结合卷积神经网络(convolutional neural network,CNN)在特征提取上的优势和长短期记忆网络(long short-term memory,LSTM)在时间序列预测上的优势,提升预测模型学习光伏功率与气象数据间长期映射关系的能力.此外,引入注意力机制(Attention)弥补输入序列长时LSTM难以保留关键信息的不足.实验结果表明:基于WGAN-GP对各类型天气样本扩充能有效提高预测精度;与3种经典预测模型相比,所提出的CNN-LSTM-Attention模型具有更高的预测精度. |
| Author | 苏宁 吐松江·卡日 雷柯松 崔传世 吴现 伊力哈木·亚尔买买提 |
| AuthorAffiliation | 新疆大学电气工程学院,新疆 乌鲁木齐 830049%国网综合能源服务集团有限公司,北京 100053 |
| AuthorAffiliation_xml | – name: 新疆大学电气工程学院,新疆 乌鲁木齐 830049%国网综合能源服务集团有限公司,北京 100053 |
| Author_FL | YILIHAMU·Yaermaimaiti LEI Kesong SU Ning TUSONGJIANG·Kari CUI Chuanshi WU Xian |
| Author_FL_xml | – sequence: 1 fullname: LEI Kesong – sequence: 2 fullname: TUSONGJIANG·Kari – sequence: 3 fullname: YILIHAMU·Yaermaimaiti – sequence: 4 fullname: SU Ning – sequence: 5 fullname: WU Xian – sequence: 6 fullname: CUI Chuanshi |
| Author_xml | – sequence: 1 fullname: 雷柯松 – sequence: 2 fullname: 吐松江·卡日 – sequence: 3 fullname: 伊力哈木·亚尔买买提 – sequence: 4 fullname: 苏宁 – sequence: 5 fullname: 吴现 – sequence: 6 fullname: 崔传世 |
| BookMark | eNotjT9Lw0AcQG-oYK39BK6uF-93l8vlBocQtAqxChYcS3K5k0S5VBPxC1TEP1RwFCHg7uhS_DhNg9_Cgk5veby3gTq2sBqhLSAOSOGzndxR9iJzJuVEOZQC9UUHdcETLmYu8HXUL8ssIYQB554vu2i3qeeL-exsEAzx4KR5fQ6HQxydjo5wUFXaVllh27dpW38u3-vm7mHx_dI81u3s_udjuvx62kRrJr4sdf-fPTTa3xuFBzg6HhyGQYRLKQGnkkqlXMYNB2O4Nq5nYs19zWUsE0iMKxQVJkk1kwmlgnlqJRnBQIKiKWE9tP2XvY2tie35OC9uru1qOM7TK0ooI5IAsF_VKFhd |
| ContentType | Journal Article |
| Copyright | Copyright © Wanfang Data Co. Ltd. All Rights Reserved. |
| Copyright_xml | – notice: Copyright © Wanfang Data Co. Ltd. All Rights Reserved. |
| DBID | 2B. 4A8 92I 93N PSX TCJ |
| DOI | 10.19783/j.cnki.pspc.221287 |
| DatabaseName | Wanfang Data Journals - Hong Kong WANFANG Data Centre Wanfang Data Journals 万方数据期刊 - 香港版 China Online Journals (COJ) China Online Journals (COJ) |
| DatabaseTitleList | |
| DeliveryMethod | fulltext_linktorsrc |
| DocumentTitle_FL | Prediction of short-term photovoltaic power based on WGAN-GP and CNN-LSTM-Attention |
| EndPage | 118 |
| ExternalDocumentID | jdq202309011 |
| GrantInformation_xml | – fundername: (国家自然科学基金); (新疆维吾尔自治区自然科学基金面上项目); (新疆维吾尔自治区优秀青年科技人才培养项目) funderid: (国家自然科学基金); (新疆维吾尔自治区自然科学基金面上项目); (新疆维吾尔自治区优秀青年科技人才培养项目) |
| GroupedDBID | -03 2B. 4A8 92I 93N ALMA_UNASSIGNED_HOLDINGS CCEZO CEKLB GROUPED_DOAJ PSX TCJ |
| ID | FETCH-LOGICAL-s991-d929cc435f51ff5ef46fae58e59a9b1bf47c27fbde39b22736c5eff73191c2d03 |
| ISSN | 1674-3415 |
| IngestDate | Thu May 29 04:03:04 EDT 2025 |
| IsPeerReviewed | true |
| IsScholarly | true |
| Issue | 9 |
| Keywords | 长短期记忆网络 注意力机制 生成对抗网络 卷积神经网络 光伏功率预测 |
| Language | Chinese |
| LinkModel | OpenURL |
| MergedId | FETCHMERGED-LOGICAL-s991-d929cc435f51ff5ef46fae58e59a9b1bf47c27fbde39b22736c5eff73191c2d03 |
| PageCount | 11 |
| ParticipantIDs | wanfang_journals_jdq202309011 |
| PublicationCentury | 2000 |
| PublicationDate | 2023-05-01 |
| PublicationDateYYYYMMDD | 2023-05-01 |
| PublicationDate_xml | – month: 05 year: 2023 text: 2023-05-01 day: 01 |
| PublicationDecade | 2020 |
| PublicationTitle | 电力系统保护与控制 |
| PublicationTitle_FL | Power System Protection and Control |
| PublicationYear | 2023 |
| Publisher | 新疆大学电气工程学院,新疆 乌鲁木齐 830049%国网综合能源服务集团有限公司,北京 100053 |
| Publisher_xml | – name: 新疆大学电气工程学院,新疆 乌鲁木齐 830049%国网综合能源服务集团有限公司,北京 100053 |
| SSID | ssib003155689 ssib023166999 ssib002424069 ssj0002912115 ssib051374514 ssib036435463 |
| Score | 2.5349107 |
| Snippet | 针对非晴天天气类型历史数据量匮乏导致光伏功率预测精度低的问题,提出了一种含有梯度惩罚的改进生成对抗网络(Wasserstein generative adversarial network with gradient... |
| SourceID | wanfang |
| SourceType | Aggregation Database |
| StartPage | 108 |
| Title | 基于WGAN-GP和CNN-LSTM-Attention的短期光伏功率预测 |
| URI | https://d.wanfangdata.com.cn/periodical/jdq202309011 |
| Volume | 51 |
| hasFullText | 1 |
| inHoldings | 1 |
| isFullTextHit | |
| isPrint | |
| link | http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwnV1Lb9QwEI5KuXBBIEC8inrApyolcWInPnBwstlWiK6QWERv1ebFS1oKu730XIR4qEgcEdJK3Dlyqfg53Vb8C2Ym7sZokXhIq2h2PB7PzLexx1nbcZybXIqCi6J2xaCGCQqk_G5c557rlYEKgxpwL_Af3Y2eXH8Q3tkUmwuLI2vV0s44Xy12f7uv5H9QBR7girtk_wHZmVJgAA34whUQhutfYcwywVSXJZplIV7j7OGa7rlr96iAszhNez337v3-hqvH42ZhI8sipkA0JKLLdIdlkqkUaagVw0eRupTFDUdTUQTaWRyxTDHNqbpkCZQmdnpLOkPkm4oJcpKAJQ2RkCoQ6DJF7YKMDokTo37kZExHVD1myeyxITYLypKIjAWru0R0WJK1IuAzLtuYlZCJPoqnnKqC0g7TPglETIu2auOvtswGZQHa0ARHxyc6KNCKJBMPnUWOQrolJIZOea16cC4x4dQQRd9-1sKtlY10d1CTkhRFSMSS6oUmLgCYllag5YkdAu3TAos0NdgKQ_AINp7OK18xhscpyiWArd86jZwOuLIS45lpinFCFpHoEKBQ5htkIYANAhg0aD8gGUnh8oxGCH8TYQSBEG2cAwKsMzLK2Iu16AepSTOED34kvKFTBLDBQqcrPm32tsZXGYUuJE7CHoDNicNNR6Os0dT3Yisx85uBem7Mx4eXNOgXw2dPVrdH28Uqh4TM5HG_Hqb-tHyBuHq44fqUc5pHkHBbz2FMzoq7wa1BDw_pm32H-ZCUqp3DB5DS26-UEH4QhcIsTcB0kCs8RRGXUs-cN6efoeG35s2m7YPDejB8ZGW6_XPOWTNFXdZNf3PeWdh9fMG5PZ0cHB7sm_5l-vH9fM9y_GnvePL16PNk-urN4fcP07eT4_3XP77sHX17d9Hpd7N-uu6aV6-4I1wKWcKkqSjAsVr4dS2qOpT1oBJxJdRA5X5eh1HBozovq0DlHGZAsgChOoLx3C946QWXnMXh82F12VmuFcdTM-GiZFgFhapKKYs8KkKRB0EZXnGWjLdbpmcdbdkoXf1D-TXnTHunXncWxy93qiWYJ4zzG4TrT44XuqQ |
| linkProvider | Directory of Open Access Journals |
| openUrl | ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=%E5%9F%BA%E4%BA%8EWGAN-GP%E5%92%8CCNN-LSTM-Attention%E7%9A%84%E7%9F%AD%E6%9C%9F%E5%85%89%E4%BC%8F%E5%8A%9F%E7%8E%87%E9%A2%84%E6%B5%8B&rft.jtitle=%E7%94%B5%E5%8A%9B%E7%B3%BB%E7%BB%9F%E4%BF%9D%E6%8A%A4%E4%B8%8E%E6%8E%A7%E5%88%B6&rft.au=%E9%9B%B7%E6%9F%AF%E6%9D%BE&rft.au=%E5%90%90%E6%9D%BE%E6%B1%9F%C2%B7%E5%8D%A1%E6%97%A5&rft.au=%E4%BC%8A%E5%8A%9B%E5%93%88%E6%9C%A8%C2%B7%E4%BA%9A%E5%B0%94%E4%B9%B0%E4%B9%B0%E6%8F%90&rft.au=%E8%8B%8F%E5%AE%81&rft.date=2023-05-01&rft.pub=%E6%96%B0%E7%96%86%E5%A4%A7%E5%AD%A6%E7%94%B5%E6%B0%94%E5%B7%A5%E7%A8%8B%E5%AD%A6%E9%99%A2%2C%E6%96%B0%E7%96%86+%E4%B9%8C%E9%B2%81%E6%9C%A8%E9%BD%90+830049%25%E5%9B%BD%E7%BD%91%E7%BB%BC%E5%90%88%E8%83%BD%E6%BA%90%E6%9C%8D%E5%8A%A1%E9%9B%86%E5%9B%A2%E6%9C%89%E9%99%90%E5%85%AC%E5%8F%B8%2C%E5%8C%97%E4%BA%AC+100053&rft.issn=1674-3415&rft.volume=51&rft.issue=9&rft.spage=108&rft.epage=118&rft_id=info:doi/10.19783%2Fj.cnki.pspc.221287&rft.externalDocID=jdq202309011 |
| thumbnail_s | http://utb.summon.serialssolutions.com/2.0.0/image/custom?url=http%3A%2F%2Fwww.wanfangdata.com.cn%2Fimages%2FPeriodicalImages%2Fjdq%2Fjdq.jpg |