Multi-dimensional Park Portrait Model Based On Clustering

Abstract

With the rapid growth of the social economy, the number of various types of parks in China has increased significantly. Faced with the increasing difficulty of man aging the power consumption of parks, qualitative analysis labels such as industrial parks and technology parks are no longer applicable, and utility companies need to develop differentiated and personalized power supply strategies for parks with different characteristics. Therefore, it is necessary to study the power consumption characteristics of t he park more detailed. A multi-dimensional park portrait model is proposed in this paper. Firstly, based on biclustering algorithm, the user's power consumption data are analyzed. Then the combined forecasting model is stablished to analyze the future power consumption of the park. Besides, an index evaluation system is established to complete the power configuration and demand response analysis of the users’. Finally, K-means cluster analysis is performed on the analysis results to obtain the park portrait.90, 000 pieces of data and related electricity business records from Henan Province are used to verify the effectiveness of the model. The results show that the park portrait obtained by the model and algorithm can quantitatively analyze the relevant attributes of the park.