A Two-Stage Power System Partitioning Method Based on Improved K-Means Clustering

Grid zoning can simplify the process of grid analysis, and zonal control management is conducive to grid stability. This paper considers the homogeneity of power system unit operation, describes the principle of two-stage automatic partitioning method, and proposes the absolute spatial distance and frequency response trajectory deviation indexes according to the principle. This paper also proposes the two-stage partitioning method of provincial and local resources. The controllable regulation resources are partitioned by the improved K-means clustering method. A provincial power grid is analyzed as an example, and the results show that the unit division obtained by the proposed method has better uniformity.