Research And Application Of Time Series Data Mining In Power Plante Equipment Management

Currently, information technology plays an important role in power companies and that is unparalleled. Among information technology database technology has been more widely used. focusing on the application of database technology for large-scale equipment running historical data are becoming more and more important. With the power to run the equipment accumulated large enterprise magnitude of historical data, The way of historical relational data processing, which is difficult to meet the needs of users. Conservative existing data processing, only to meet the primary analysis of the data, but does not find out of implied knowledge in massive data and that is great practical significance and value, so, which is unable to penetrate deeper meaning research from the time and space to understand massive historical data.resulting in the utilization of the data is insufficient. However, any company which have a "treasure trove of historical data" limited by the lack of knowledge, can’t to dig out "treasure" in the valuable information and data.The paper focuses on the issues that current massive historical data can not use in the most efficient, though knowledge of data warehouse, data mining technology and the data mining algorithm, explored a solution to meet the manager of power equipment to management power plant equipment in effective and reasonable. On the premise of a deeper understanding of the operation principle of power plant equipment, in use of data warehouse knowledge theory, and time series data analysis of all aspects of the mining algorithm of massive historical data of the power plant. The main technology and method of time series data mining is ARTXP algorithm and ARIMA algorithm, in process of KDD for temporal data information in order to develop reasonable long-term or short-term prediction, and provide effective support for rational decision-making, through the analysis and research on the two kinds of mining algorithm, create a hybrid model combine the ARTXP algorithm with ARIMA algorithm are the ideal choice, which can improve the trend analysis of the accuracy and timeliness of unearthed rule, knowledge from historical data. At the end of get judgment and analysis to the condenser vacuum equipment which is the core equipment in the power plantof. by the way, found that the main factor which affecti condenser vacuumo from historical data.finally, give a scientific and reasonable method for maintenance and monitoring measures.