ISSN 1000-1026

CN 32-1180/TP

+Advanced Search 中文版
Data-driven Decoupling Evaluation Method of Wind Power Prediction Error

1.North China Power Dispatching and Control Branch Center of State Grid, Beijing 100053, China;2.North China Electric Power Research Institute Co., Ltd., Beijing 100045, China


For the complexity of multi-link interaction in wind power prediction, this paper proposes a refined evaluation method for wind power prediction error. It aims to quantitatively analyze the effect of each key link of power prediction on the total prediction error by using numerical weather prediction, meteorological observation data, wind power operation data and other multi-source heterogeneous information. Firstly, the operation mechanism of wind power prediction is analyzed, the prediction process is divided into three key links, i.e., numerical weather prediction, wind energy-power conversion modeling and prediction result correction. Secondly, based on the kernel density estimation, the segmented identification method is designed for wind power anomaly data, and a simplified model of wind resource-power characteristics is established. Finally, based on the multi-source operation data-driven in meteorology and electric power, a method for measuring equivalent error in each link of power prediction is proposed. The case results show that the proposed method can quantitatively evaluate the effects of each link on the power prediction error.



Get Citation
[1]JIANG Changming, YANG Jian, LIU Yu, et al. Data-driven Decoupling Evaluation Method of Wind Power Prediction Error[J]. Automation of Electric Power Systems,2021,45(1):105-113. DOI:10.7500/AEPS20200318005
  • Received:March 18,2020
  • Revised:June 05,2020
  • Adopted:
  • Online: January 05,2021
  • Published: