基于趋势分类与空间信息集成模型的日前风电场群功率预测

Abstract With the rapid development of the wind power industry and the increase of installed wind power capacity, the factors affecting the power generation show a highly coupled relationship in time and space, which brings extremely challenges to wind farm cluster power prediction (WFCPP). To solve this problem, this study proposes a method to improve the accuracy of regional wind power prediction (WPP) method which takes into account the trend aggregation of wind power clusters and spatial information integration (SII). Firstly, A trend clustering method considering spatial characteristics is introduced to realize cluster division. This method uses static partition method to deal with the dynamic environment of continuous random changes, and weakens the influence of wind speed spatial dispersion on cluster division. Then, the multidimensional spatiotemporal coupling characteristics between multiple wind farm clusters (WFC) are deeply explored, and the input mode of incorporating spatiotemporal information is constructed. Finally, the proposed method is applied to a wind farm cluster in Jilin, China. The results show that the RMSE , MAE , and SMAPE of the proposed method are on average 2.07 %, 2.53 %, and 8.34 lower, and the R 2 and R are on average 17.91 % and 24.2 % higher, compared to other WFCPP methods. This will further boost the uptake of wind power (WP), while reducing the impact of large-scale wind power grid connection on the safe and stable operation of the power system .