Extraction and application of intrinsic predictable component in day-ahead power prediction for wind farm cluster

IF 9 1区工程技术 Q1 ENERGY & FUELS

Energy Pub Date : 2025-05-12 DOI:10.1016/j.energy.2025.136530

Mao Yang , Renxian Jiang , Xinnan Yu , Bo Wang , Xin Su , Chenglian Ma

{"title":"Extraction and application of intrinsic predictable component in day-ahead power prediction for wind farm cluster","authors":"Mao Yang , Renxian Jiang , Xinnan Yu , Bo Wang , Xin Su , Chenglian Ma","doi":"10.1016/j.energy.2025.136530","DOIUrl":null,"url":null,"abstract":"<div><div>With the continuous updating and iteration of artificial intelligence algorithms, prediction models emerge one after another, but the research and utilization of the predictability of wind power is still less. Therefore, this paper proposed a day-ahead power prediction method for Wind Farm Cluster (WFC) based on intrinsic predictable component extraction. Firstly, based on the difference of power distribution between wind farm and WFC, a method of Wind Power Curve (WPC) modeling for WFC is proposed, which provides the basis for establishing the Final Set of Wind Power Curve (FSWPC). Secondly, the Intrinsic Predictable Component (IPC) of wind power is extracted based on the FSWPC, and the Interference Component (IC) of wind power is separated to eliminate the influence of IC on IPC in the process of prediction. Thirdly, the historical similarity matching method with large threshold is used to predict the IC to make up for the numerical deficiency of the IPC. Finally, the proposed method was applied to a WFC in China to verify its effectiveness. Compared with the traditional prediction strategy, the NRMSE, NMAE and MAPE are reduced by 1.59 %, 1.15 % and 6.42 %, respectively.</div></div>","PeriodicalId":11647,"journal":{"name":"Energy","volume":"328 ","pages":"Article 136530"},"PeriodicalIF":9.0000,"publicationDate":"2025-05-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Energy","FirstCategoryId":"5","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S0360544225021723","RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"ENERGY & FUELS","Score":null,"Total":0}

引用次数: 0

Abstract

With the continuous updating and iteration of artificial intelligence algorithms, prediction models emerge one after another, but the research and utilization of the predictability of wind power is still less. Therefore, this paper proposed a day-ahead power prediction method for Wind Farm Cluster (WFC) based on intrinsic predictable component extraction. Firstly, based on the difference of power distribution between wind farm and WFC, a method of Wind Power Curve (WPC) modeling for WFC is proposed, which provides the basis for establishing the Final Set of Wind Power Curve (FSWPC). Secondly, the Intrinsic Predictable Component (IPC) of wind power is extracted based on the FSWPC, and the Interference Component (IC) of wind power is separated to eliminate the influence of IC on IPC in the process of prediction. Thirdly, the historical similarity matching method with large threshold is used to predict the IC to make up for the numerical deficiency of the IPC. Finally, the proposed method was applied to a WFC in China to verify its effectiveness. Compared with the traditional prediction strategy, the NRMSE, NMAE and MAPE are reduced by 1.59 %, 1.15 % and 6.42 %, respectively.

查看原文本刊更多论文

固有可预测分量在风电场组团日前功率预测中的提取与应用

随着人工智能算法的不断更新迭代，预测模型层出不穷，但对风电可预测性的研究和利用仍然较少。为此，本文提出了一种基于固有可预测分量提取的风电场集群日前功率预测方法。首先，根据风电场与风力发电中心的功率分布差异，提出了风力发电中心的风电功率曲线建模方法，为风电功率曲线最终集的建立提供了依据。其次，基于FSWPC提取风电的内在可预测分量（IPC），并分离风电的干扰分量（IC），消除预测过程中干扰分量对IPC的影响；第三，采用大阈值历史相似度匹配方法预测集成电路，弥补集成电路在数值上的不足。最后，将该方法应用于中国的一个WFC，验证了其有效性。与传统预测策略相比，NRMSE、NMAE和MAPE分别降低了1.59%、1.15%和6.42%。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

Energy 工程技术-能源与燃料

CiteScore

15.30

自引率

14.40%

发文量

审稿时长

14.2 weeks

期刊介绍： Energy is a multidisciplinary, international journal that publishes research and analysis in the field of energy engineering. Our aim is to become a leading peer-reviewed platform and a trusted source of information for energy-related topics. The journal covers a range of areas including mechanical engineering, thermal sciences, and energy analysis. We are particularly interested in research on energy modelling, prediction, integrated energy systems, planning, and management. Additionally, we welcome papers on energy conservation, efficiency, biomass and bioenergy, renewable energy, electricity supply and demand, energy storage, buildings, and economic and policy issues. These topics should align with our broader multidisciplinary focus.