Power System Operation Mode Identification Method Based on Improved Clustering Algorithm

Abstract

For power system calculation and analysis, the accuracy and rationality of operation mode selection is the key to determine the calculation quality. With the access of a high proportion of renewable energy, the traditional manual selection method is not applicable. How to automatically extract the typical operation mode form the data set obtained from production simulation is an urgent scientific and technical problem to be solved. This paper firstly carries out the demand analysis of operation mode extraction of high proportion renewable energy power system. Secondly, an automatic mode extraction algorithm based on K-means++ algorithm and improved cluster validity index is proposed. Then this paper designed a mode extraction approach with joint manual processing and automatic algorithm. Finally, based on the practical data of a region power grid in China, the numerical experiments demonstrate the effectiveness and rationality of the proposed algorithm based on the comparison with the manually selected operation mode from two aspects of mode characteristics and security check. The contribution of the algorithm in improving the level of power system planning was proved.