Application of data mining in boiler combustion optimization
Tingting Yang, Jizhen Liu, Deliang Zeng, Xie Xie · 2010
In order to improve boiler efficiency and reduce NOx emissions of a coal-fired boiler, a new solution strategy of combustion optimization is proposed is this paper. The key point of combustion optimization is the optimal setpoints of fuel and air parameters. As the development of electric industry, large amounts of history data are accumulated and data mining technique is applied to find some useful results. Fuzzy sets theory is introduced into the association mining process in order to soften the partition boundary of the domain and generalize the data. And then cluster algorithm and fuzzy association rule are employed to obtain the optimal values of important parameters. The rules mined out are combined with control system of operation parameters and can be used to provide guides for operation. Experimental results in a 600 MW power plant show that the method is useful and effective.