The invention relates to a modeling method of a circulating fluidized bed boiler combustion system model. The method is characterized by comprising the following steps of: 1, collecting original data of a circulating fluidized bed boiler combustion system; 2, performing the following preprocessing on the original data; 3, performing dimension reduction processing on preprocessed sample data by a deep belief network to obtain a data set X; 4, performing fuzzy C-means clustering on the data set X obtained in the third step to obtain the clustering center and the clustering radius; and 5, using the clustering center and the clustering radius obtained in the fourth step as the center of a radial basis function of a radial basis function neural network and the initial valve of an expansion constant, using a gradient descent algorithm to train the radial basis function neural network, and forming the deep belief network and the radial basis function neural network obtained through training into the circulating fluidized bed boiler combustion system model. The modeling method provided by the invention is effective and reasonable.