RBF (Radial Basis Function) neural network based supercritical boiler nitric oxide discharging dynamic predication method

The invention discloses an RBF (Radial Basis Function) neural network based supercritical boiler nitric oxide discharging dynamic predication method and belongs to the technical field of environmental protection discharging parameter measurement. According to the technical scheme, the RBF neural network based supercritical boiler nitric oxide discharging dynamic predication method comprises the following steps of SS1 selecting a static auxiliary variable and a dynamic auxiliary variable; SS2 performing RBF neural network structure fitting on the static auxiliary variable and the dynamic auxiliary variable to obtain an RBF neural network structure based training boiler nitric oxide discharging dynamic predication model; adjusting RBF neural network parameters and obtaining an RBF neural network structure based supercritical boiler nitric oxide discharging dynamic predication model. According to the RBF neural network based supercritical boiler nitric oxide discharging dynamic predication method, under conditions that a same training error is set and the like, the number of interior nerve cells of the dynamic model is obviously smaller than that of a static model, the model structure is simple, the training time is short, and the generalization ability is strong.