基于改进空间资源匹配法的风电集群功率预测技术

大规模风电集群的功率预测,有利于调度部门制定科学合理的发电计划,提升电网的健壮性。基于空间资源匹配法(spatial resources matching algorithm,SRMA)的风电集群功率预测方法,比广泛采用的统计升尺度法具有更高的精度,而且需要的计算资源较少。但是现有的空间资源匹配法,匹配参数单一,不利于预测精度的进一步提升。文章在详细介绍空间资源匹配法的基础上,提出了一种考虑风电功率测量数据的改进空间资源匹配法,并通过52个风电场组成的风电集群开展了0~12 h的风电功率预测。结果表明,改进的空间资源匹配法前4 h的预测精度比传统的匹配法有较大幅度的提升,具有较强的工业应用推广价值。

A Lifting Spatial Resources Matching Approach Based Wind Power Prediction of Regions

Wind power prediction of large scale wind farm clusters will contribute to the scientific and reasonable power generation schedule establishment and enhance the robustness of the power grid.Spatial resources matching algorithm (SRMA) based wind power prediction of regions is with higher prediction accuracy and less computing time than the method of up-scaling approach, which is widely adopted by industrial companies.However, there is only one matching parameter of the SRMA method, which restricts the further improvement of the prediction accuracy.This paper presents an improved SRMA method, which contains the parameter of the historical wind power output, based on the introduction of the SRMA method.Then, this paper predicts the wind power within 0~12 hours with the data derived from one wind farm cluster which contains 52 wind farms.The results show that, the prediction accuracy of the improved SRMA method within 4 hours is higher than that of the traditional SRMA method, and is applicable for industrial application.