Power-assisted Research And Application Of Decision Analysis System

Nowadays, computer system is used widely in lots of electric power enterprises. But the enterprise's leaders have found the disproportion between the high investment and low revenue of building a computer system regretfully. Corporations apply Management Information System (MIS) whose requirements are network, hardware and software maintenance. Many systems of network, client machines, workstations and high-grade computing centers are introduced to gather all sorts of data rapidly and store them largely, but just to get all kinds of bills, files, reporting forms and so on. So, it is necessary to make analysis and statistics by manpower to make important decision. Decision Support System (DSS) is the very enterprise solution to conquer the disadvantages mentioned above. Under the precondition that network is complete and data is enough, DSS can improve the availability of all resources including software, hardware, network and data. With it, decision-making information can form through effective data filtering, data clearing and data processing. It shortens the process from collecting data to making decision, replacing trivial manual work and making computer system worthy of the cost of building itself. This thesis looked into the theory and technology of building a DSS including three correlative technologies, which are data warehouse, on-line analytical processing (OLAP), and data mining, comprehensively and systemically. From the view of engineering design, with the method of systems engineering and specific requirement of electric power enterprises, the thesis discussed the whole process of building an aided power decision analysis system including constructing a distributive framework which is a combination of B/S and C/S, designing Microsoft data warehouse environment, developing OLAP application on Hyperion Performance Suite Business Intelligence Platform, introducing a methodology of data mining by statistical method and also developing the corresponding data mining tool. The system studied by the thesis can work normally and steadily through debugging and can meet the demand of electric power enterprises basically. To...