In operation control of thermal power plant boilers

there are many problems

such as various controllable operation parameters

difficult mechanism modeling and deviation between operation parameters and design values

which make it difficult to optimize the operation of boilers driven by data. Thus

a pattern mining method is applied to analyze the historical operation data of the distributed control system(DCS) of the related boiler monitoring points. Reasonable fuzzy partition matrices are constructed and matrix operations are performed to mine the fuzzy association patterns that match the boiler's high-efficiency operation under load changes

thereby to guide the operation optimization of boiler. By taking the 130 t/h circulating fluidized bed boiler of a thermal power plant in Dongguan as the research object

the historical data of its DCS are extracted. The data are analyzed by the method of this paper and the control optimization strategy implied in operation is mined. The optimization results show that

this method can provide rapid and reasonable support for the economic operation of the boiler.