Application Of Data Mining Technology To The Fuzzy Control System Of CFBB

Fuzzy control theory has been applied to the control system of circulating fluidized bed boiler (CFBB) successfully. The main problem is how to acquire the rules base.Based on the mass history data of CFBB,modern data mining technology can solve the problem effectively. It provides a series of methods to extract practicable rules from the experience of the skilled operators of CFBB. For the first time this dissertation discusses the feasibility of data mining for the fields data of CFBB,analyses the preprocessing technology of sample data,introduces the application of clustering analysis,and finally completes the rule extraction by applying knowledge-based artificial neural network.Considering distributed parameters,nonlinear,and long time lag system of CFBB,the theory of self-organizing neural fuzzy inference system is applied to the control system of CFBB,and enables control rules extraction on-line. The models of data mining and self-organizing (FBNC-PNN controller) are programmed,embedded to the control system of a 35 t/h CFBB in Tsingtao,and finally improve the performance of CFBB effectively.