Fuzzy Neural Network In Circulating Fluidized Bed Boiler Combustion System

Circulating fluidized bed boiler (Circulating Fuidized Bed Boiler, referred to as CFBB) originated in the 70s of last century, is derived from an ancient fire-bed combustion technology, it is compared with conventional pulverized coal boiler is a cost efficient and less polluting clean combustion technology But the industry on the circulating fluidized bed boiler combustion control system uses a traditional PID control, the circulating fluidized bed combustion system of nonlinear, multivariable coupling, large delay so that PID control can not achieve good Control effect. So this paper a fuzzy neural network to circulating fluidized bed boiler combustion control system, which is the control of circulating fluidized bed boiler trendsIn this thesis, firstly, the structure of 300MW circulating fluidized bed boiler and process were described in detail and analyzes its dynamic characteristics of the combustion system. Combustion system model in the previous studies conducted on the basis of a simplified model.Secondly, introduction the neural network and fuzzy control, leads to the fuzzy neural network. And fuzzy neural network controller structure and implementation of algorithms, due to their lack of BP algorithm, FNN controller optimized by off-line BP algorithm, it is difficult to improve the accuracy of optimization. Therefore, using an improved particle swarm optimization algorithm FNN controller off-line optimization, to improve the optimization accuracy.Finally, the fuzzy neural network control used in circulating fluidized bed boiler combustion control system, on the bed temperature control and main steam pressure control of MATLAB simulation and were compared with PID control, PID control of the control performance significantly.