Power User Classification Based On FCM Clustering And Design & Implementation Of Demand Dide Management Platform

In recent years,the total amount of electricity load in various regions of China continues to rise,and the imbalance between supply and demand in the power market continues to escalate.Moreover,the current power demand side management work is not in place,the allocation of resources is not reasonable.Under the condition of limited power resources,orderly power consumption is the most direct and effective load management measure to alleviate the contradiction between power supply and demand in peak period.In order to improve the utilization rate of power resources,the value mining of power user side load data is the basis of analyzing the power consumption behavior of users,and the implementation of differentiated demand-side management according to different types of users is also one of the hot topics in current power grid research.The main contents of this paper are as follows:(1)Collect and process load data.Sample data in this paper were obtained from EMS system in Jiangxi Province,and Kettle tool was used to extract the data.First,load data is cleaned,and then load characteristics of typical users are extracted.In order to improve the effectiveness and accuracy of load clustering,load power data is normalized,and median filtering method is used to smooth the curve.After processing,the data can more intuitively reflect the curve shape and variation of the load curve.(2)Power User Classification Based on FCM Clustering Algorithm.FCM clustering algorithm is applied to power user classification.In order to extract power load mode categories of users.Firstly,clustering was carried out based on the processed data,and the number of clustering was determined to be 5 according to the DBI index.Then,the sample data were divided into five load curve modes by continuously determining the cluster center point and membership degree.Finally,the potential of different users to participate in orderly power consumption was analyzed according to the clustering results.It has been verified by practice that according to the fine load division mode of power users proposed in this paper,differentiated power supply demands can be provided for users,and the qualified rate of enterprises participating in orderly power consumption can be improved.(3)A power demand side management platform based on orderly electricity consumption is designed and implemented in the dispatching safety zone.Platform takes load data collection,real-time load monitoring,orderly power consumption and interruptible analysis and management,and load data analysis and statistics as the main line.The relevant technology,overall architecture design,interface and safety design,load management process,and main function realization of the platform are discussed.By establishing a two-way interaction mechanism between grid side and user side,combined with big data development and Web development technology,the platform can dynamically collect data and distribute power resources in real time and accurately.Taking total amount regulation and structure optimization as the standard,the power demand side management integrating management,monitoring and distribution is realized.