Design And Implementation Of Power Grid System Based On Electricity Demand Forecast And Customer Behavior Analysis

With the information updates faster and faster, the amount of data is also growing. Apower grid corporation data center has established data warehouse covering multiple topics.The power grid needs to advance and implement data mining model based on data analysis,and then design and implement a visible power grid system based on data mining to decideand analyze.The paper studies the data warehouse and the application requirements of the power girdto determine electricity demand forecast, customer segmentation and arrears forecast as themining target. Then design and implement a power grid system based on electricity demandforecast and customer behavior analysis according to the software engineering process. Thesystem functions run through the whole process of business object determination, datapreparation, data mining, result analysis and knowledge assimilation.At the stage of requirements analysis, make detailed research to understand the specificapplication needs, and then divide the system into five modules of data acquisition, dataexploration, data preprocessing, forecast and analysis, system management. At the stage ofoutline design, introduce the three-tier architecture, and then use sequence diagram and otherUML modeling language to summarize the outline design, and describe the design of thedatabase. At the stage of detailed design, use class diagram to summarize the design of eachmodule, and use flow chart and pseudo code to emphasize the detailed process of machinelearning, statistical analysis and other data mining method used in electricity demand forecast,customer segmentation and arrears forecast. At the stage of implementation and application,use J2EE architecture as the development platform and provide a description of thedeployment and application, and then introduce the specific operation and display the mainachievement with the instance to analyze the result of the model evaluation.The power grid system has been designed and implemented. In the practical application, the user only needs to maintain a connection to the data warehouse, and then he can use thevisible interface realizing the functions of electricity demand forecast, customer segmentationand arrears forecast to make targeted marketing programs.