Analysis Of Electric Power User Consumption Pattern Based On Big Data Method

At present,more and more attention has been paid to the role and status of electricity user in the whole power network system all over the world.The analysis of electric power user consumption pattern can help to guide the management of power companies and power user potential mining.It can also help to find efficient strategies to balance the power load and help users to achieve efficient management.With the widespread use of smart meters,we have acquired and collected a large number of electricity information of the power user.These high-frequency high-density and precision sampling data is a valuable resource for power system analysis,and therefore we use the information of electric power user consumption from smart meter and some detail user data obtained from research and Big data method to analyze the electric power user consumption patterns.In this paper,the following work is done:1)Combined smart meter measurement information,to the original user of electricity data and user information using a large data base sampling method collected data,and other data cleansing and standardization pretreatment method mechanism for large data were introduced and explained,and on applications related to this article using the large data method practical examples were expanded.2)User preprocessed information data base sampling were carried out attribute reduction based on information entropy decision table attribute reduction based on rough sets,the energy saving potential users find the closest correlation decision variables,and method and relatively large data similarities and differences between the traditional method of attribute reduction and superiority.Then use the decision tree method to establish user behavior patterns and decision rules electricity users saving potential.3)The establishment of a high-dimensional correlation based on user consumption pattern random nonparametric model based on high-dimensional random matrix theory.Analysis of user patterns of different users of electricity and the associated potential for energy saving relationship with the user's electricity consumption data.4)Actual power user electric energy data as well as the research object that contains the user's home and living conditions,electricity habits,saving electricity and other detailed research profile attitude,study habits of users of electricity in each mode associated with a user saving potential relationship,based on the results Looking for high energy-saving potential of user types.The data presented herein establish large stochastic nonparametric model,information entropy attribute reduction method based on data preprocessing methods and data used in the large users of electricity urban analysis of the behavior of a European study,the electricity consumption information to users and user electricity habits attitude information for analysis,to achieve a model entirely data-driven,information and data on the consumption behavior of users of electricity acquired from the smart meter customary relationship mining,user behavior and saving potential study to verify the validity of the analysis model for the future behavior of the targeted user guide customized management and energy efficiency to establish a good foundation.