BIRCH-AHC Two-Layer Clustering Algorithm-Based Typical Scenario Generation of Photovoltaic Output for Power Systems

The BIRCH-AHC two-layer clustering algorithm is presented to reduce a large number of photovoltaic output scenarios. In the presented algorithm, CH index is introduced to select the appropriate clustering structural parameters, which can construct a BIRCH clustering feature tree that contains different photovoltaic outputs. Then, combining the elbow method to determine the optimal number of clusters, AHC algorithm is proposed to generate a typical scenario set of photovoltaic output. Finally, the photovoltaic output data of a regional power grid is taken as an example to verify the feasibility of the proposed method, and the result shows that the clustering analysis on large-scale photovoltaic power generation data can be well performed, which means the proposed method is capable of being applied to the planning and decision-making models of power system, and thus improves the accuracy and effectiveness of power system operation.