摘要
数据挖掘技术能充分提取出电厂SIS存储的大量生产数据中隐藏的生产规律。目前的电厂生产数据挖掘因受数据源不足、电厂生产过程了解不够等因素制约,存在原始数据量不够、非稳态数据挖掘、挖掘结果实际指导意义不足的问题。按电厂生产特点提出了基于工况划分、以关联规则为手段的电厂数据挖掘,同时以盘山电厂600MW机组1个月的历史数据为基础,抽取出所有稳态数据,在以负荷、循环水进口温度、煤质为外部条件进行工况划分的基础上,用关联规则对电厂经济性指标——供电煤耗率进行了挖掘,抽取出了相关工况下的运行模式。实际应用表明,该方法对机组运行经济性规律的分析及优化运行具有指导意义。
There is a lot of operation data of power plant in the SIS, and data mining technology can fully extract the production rule hidden in the data. Many data mining experiments have been tried in power plant. But because of insufficient data sources and inadequate understanding of the production process of power plant, there are many disadvantages such as lacking of original data, non-steady-state data mining, unsystematic results, and results of no practical significance. According to the features of the power plant, a data mining method based on working condition division and association rule was proposed. Using a month of data of 600 MW unit in Datang Panshan Power Plant, all the steady-state data were extracted. Then based on operation condition division of load, circulating water inlet temperature and coal, economic index versus coal consumption rate were mined with association rule. Finally the different modes under diverse operation conditions were extracted. The practical application indicates that the method in this paper provide practical reference of economical analysis and the further optimization in power plant operation.
出处
《中国电力》
CSCD
北大核心
2009年第7期68-71,共4页
Electric Power
基金
国家863重点资助项目(2007AA041105)
关键词
电厂
数据挖掘
关联规则
工况划分
供电煤耗率
优化运行
power plant
data mining
association rule
operation condition division
coal consumption rate
optimized operation
