东北大学学报(自然科学版) ›› 2021, Vol. 42 ›› Issue (2): 174-179.DOI: 10.12068/j.issn.1005-3026.2021.02.004

• 信息与控制 • 上一篇    下一篇

基于mRMR-ESN的单变量光伏功率预测

韩鹏, 郭天, 汪晋宽, 史泽伟   

  1. (东北大学秦皇岛分校 计算机与通信工程学院, 河北 秦皇岛066000)
  • 收稿日期:2020-06-23 修回日期:2020-06-23 接受日期:2020-06-23 发布日期:2021-03-05
  • 通讯作者: 韩鹏
  • 作者简介:韩鹏(1988-),男,山东东营人,东北大学秦皇岛分校讲师,博士; 汪晋宽(1957-),男,辽宁沈阳人,东北大学教授,博士生导师.
  • 基金资助:
    国家自然科学基金资助项目(61603083); 河北省自然科学基金资助项目(F2017501014); 中央高校基本科研业务费专项资金资助项目(N172304028).

PV Power Forecasting with Univariate Input Based on mRMR-ESN

HAN Peng, GUO Tian, WANG Jin-kuan, SHI Ze-wei   

  1. School of Computer and Communication Engineering, Northeastern University at Qinhuangdao, Qinhuangdao 066000, China.
  • Received:2020-06-23 Revised:2020-06-23 Accepted:2020-06-23 Published:2021-03-05
  • Contact: HAN Peng
  • About author:-
  • Supported by:
    -

摘要: 设计了两种预测模型:利用天气变量作为输入的传统多变量预测模型;利用历史功率数据作为输入的新型单变量预测模型.采用最小冗余最大相关(min-redundancy and max-relevance,mRMR) 方法分别对两种模型进行特征提取,并选用在时间序列预测方面具有优势的回声状态网络( echo state network,ESN)对未来5min的光伏功率进行仿真预测.仿真结果表明,采用mRMR方法对历史光伏功率数据进行特征提取,确定能够使预测模型达到最优效果的特征子集,并将其作为单变量预测模型的输入,可以得到更准确的预测效果.所构建的新型单变量预测模型能够为光伏电站提供新的光伏预测思路.

关键词: 回声状态网络;特征提取;特征子集;单变量输入;光伏功率预测

Abstract: Two kinds of prediction model are designed, ie, the traditional multivariable forecasting model using weather variables as input and the new univariate forecasting model using historical power data as input. The mRMR(minimum-redundancy and maximum-relevance)method is used to extract the features of the two models respectively, and ESN(echo state network), which has advantages in time series prediction, is used to simulate and predict the photovoltaic (PV) power in the next 5minutes.The simulation results show that the mRMR method is used to extract the features of the historical PV power data, then, the feature subset that can make the prediction model reach the optimal effect is determined. The feature subset is used as the input of the univariate prediction model to obtain more accurate prediction effect. The new univariate prediction model can provide a new PV prediction idea for PV power station.

Key words: echo state network; feature extraction; character subset; univariate input; photovoltaic power forecasting

中图分类号: