Regional Wind Power Forecasting Based on Hierarchical Clustering and Upscaling Method

For a long time, the world has been committed to optimizing the energy structure to reduce carbon emissions, so that renewable energies, for example wind power, have been widely integrated into power system. A large number of random and fluctuating wind power makes the system bear more and more risks, which has caused the dispatching department to pay increasing attention to regional wind power output. Although the upscaling method is widely used to predict regional wind power output, it still has shortcomings. This paper proposes a regional wind power prediction based on hierarchical clustering and upscaling method. This approach uses a greedy algorithm to search for the optimal number of sub-regions. Finally, the effectiveness of the proposed forecasting approach has been verified on real-world data.