Research And Development Of Data Mining System For Economical Operation Pattern Of Thermal Power Units

Large amount of data was stored in real-time/historical database of SIS, that was forture to a power plant. The data contained a lot of knowledge, operation law of thermal power unit can be found behind the data. We can obtain mode rules of economic operation from data using data mining technology. These rules have great significance to guide the operation optimization of power unit. That is good to energy-saving and emission-reduction.Economic indicators of thermal power unit was researched first, factors influencing unit economy were analyzed. The concept and expression of operating mode on economy of thermal power units were proposed. Then the application of data mining technology in thermal power plant was introduced. Combined with the characteristics of power plant data, a new data mining algorithm based on similarity analysis technique was proposed. Furthermore, an improved incremental updating algorithm was put forward. On the basis of actual needs of power plant, data mining system of economy operation pattern of thermal power units was developed. Historical data can be processed with the system, and mode rule of economic operation can be obtained. Take historical data of a 600MW power unit as an example, data mining test was carried out. The test results showed that economical operation pattern of thermal power units can be extracted with the incremental updating algorithm and the data mining software, it was valuable for economic operation of the units.