The Design And Application Of OLAP System Based On Data Mining For Power Supply Company

With the continuous development of grid technology and the popularization of enterprise information, Power grid and substation information management system will have a large amount of data per day, The accumulation of a large number of these data and showed the trend of the information explosion, make the enterprise lost into a awkward position means "data rich and information poor".After nearly one year internship and investigating at the ZouCheng Power Supply Company, I gradually discovered a number of related issues, For example, each substation tripping accident occurred, should organize a special panel on the protection and safety monitoring device given a lot of data analysis and processing, is extremely time-consuming, it is easy to delay the deal with the accident. Initiation of the study will use the data mining and OLAP knowledge, simplify the manual labor, improving the power company for the accident of processing power, and thus improve power companies for emergency response capacity and to ensure the safe operation of power idea.The main work of this paper is to build a OLAP system based on data-mining for power supply company, with the analysis to a large number of business logical data of ZouCheng Power Supply Company in accordance with the adjusted CRISP-DM data mining process The data business understanding, data understanding, data, coarse-grained processing demand by topic modeling, through data ETL into the data warehousing, Then the data warehouse were constructed according to different themes Fault analysis and prediction and analysis of grid-load-balancing model,then use the application of OLAP analysis tool for multidimensional analysis, found hidden or not easily found the law or the correlation between data.In this paper, the major work process is, first of all, in accordance with the basic principles of data mining, Though the analysis of business logic database in Zoucheng Power Supply Company, do the analysis of system needs, determine the subject domain, though business understanding and data understanding to established sources of data to support decision-making, And clearly the data warehouse success criteria and key performance indicators, The use of Microsoft SQL Server 2005 in accordance with the difference of its back-machine software platforms and database management software between 20 substations under the jurisdiction of Zoucheng Power Supply Company, take different measures, By coarse-grained data handling, Complete the preparation of data warehouse construction.Then build corresponding star models in accordance with the different requirements.Through the data ETL, build the data warehousing. For the needs of fault analysis and prediction and analysis of network-load-balancing, by dividing Measures and the dimensions to create multidimensional data analysis model.Used MDX of database components of Microsoft Analysis Services to do the intelligent OLAP processing to the data warehouse. Finally, sum up, points out the shortcomings and deficiencies, and suggest improvements for future work.