Research And Realization Of Power Marketing Assist DSS Based On Data Warehouse

On the basis of in-depth studying of the theory of Decision Support,by using the Data Warehouse,Data Mining and On-Line Analytical Processing,the paper designs and implements the Power Marketing Decision Support System.Decision-Making of Marketing belongs to semi-structure category lacking of normal form to be followed,so building Decision-Making Support System is very important.After carefully analysis of the Power Marketing Business,the paper determines the marketing analysis theme and designs the structure of the Data Warehouse and System Overall Framework.The system bases on J2EE standard of three-tier architecture and B/S development mode,using Mondrian which is the Open-Source server for OLAP to achieve the analysis of electric target,and displaying the result by JPivot.In addition,the system implements Unary Line Regression Predict Algorithm,Neural network forecasting model based on PSO Algorithm and k-center Cluster Algorithm as the Model storehouse.By using these algorithms,the system implements forecasting of electricity sales amount,power load and dividing of customer to supply the support for decision.