Application Of Grid Clustering Algorithm In Electric Power Marketing

With the rapid development of Smart grid, the requirement of information degree in power enterprise becomes higher and higher. And the rapid development of information technology makes the electric power industry has accumulated a large amount of data, analysis and process of a large number of scattered data in electric power marketing system and the external environment (local climate, local economic development) and other data resources is very necessary. In order to use these data effectively, analysis methods and technical means to these massive data have been proposed, i.e., data mining.This paper first introduces the base theory knowledge and method of data mining. Base on the study of typical clustering method at home and abroad some popular grid clustering analysis methods are discussed, Grid-based Clustering Algorithm for Multi-Density (GDD) is analyzed and realized. The result shows GDD algorithm has utilized the technique of grid to improve the efficiency of clustering. GDD algorithm can avoid clustering algorithm is sensitive to parameters through the parameters of the automatic processing, can solve diversity of the data set, can effectively recognize the various isolated points and data interference and is not restrict the sequence of input data. In this paper, the GDD algorithm is applied to electricity sales analysis module and used to analyze a large number of dispersed data in electric power marketing system and the external environment data. The result shows the algorithm can not only deal with the data of the diversity and complex, and has the main advantages of general grid clustering algorithm. At the same time, the clustering algorithm based on grid distance is analyzed and realized. Aiming at the characteristics of electric power marketing, in this paper, an electric power marketing system applied in electric power enterprise is developed, the GDD algorithm and the clustering algorithm based on grid distance in research is used to electric power marketing system. The results show the grid clustering algorithm can get favorable classification in electric power marketing system.