Network Loss Allocation Method Based on Improved K-means Clustering And Correlation Analysis

The allocation of power system network loss is related to the economic interests of multiple subjects, especially the rationality of the allocation of network loss is more important when large-scale new energy is connected to the power grid.In order to make a contribution to more accurately reflect the characteristics of grid operating, this paper put forward a loss allocation method based on improved K-Means of clustering analysis. First of all, the rationality of various allocation methods and existing problems are analyzed, and then introduces improved K Means clustering and correlation analysis principle, this paper proposes an improved k-means clustering and correlation analysis network loss allocation method that the network loss data are classified according to the characteristics, and the correlation between loss and allocation body is analyzed, and the allocation proportion is determined according to the correlation coefficient. Different proportion can be adopted under different flow section, which makes the result more realistic and greatly reduces the complexity of calculation. Finally, an example is used to analyze and calculate the proposed allocation method, and the feasibility and superiority of the proposed allocation method are verified by comparing it with the traditional average network loss coefficient method.