A Study On The On-line Monitoring Of Slagging,Fouling And Intelligent Sootblowing On Utility Boiler Heat Surfaces

In this paper, a method for the on-line monitoring of fouling and slagging on utility boiler heat surfaces is put forward, and the basic principle of which is stated. On the basis of the method, a suit of on-line monitoring system of fouling and slagging is developed to guide and optimize the process of sootblowing.The principle of monitoring is on the knowledge of the NN(Neural Networks). Using the data from the DAS of the plant and the variations of slagging for a long time, we can train the NN to predict the parameters needed in different cases. At last, we will acquire the coefficient which signifies the degree of fouling and slagging, but needn't increase or just increase a few new monitoring points. Therefore, the feasibility of the monitoring of fouling and slagging based on NN is mainly studied, and the arithmetic of NN is greatly improved by improved RPROP networks instead of BP networks.During the discussion of the coefficient which signifies the degree of fouling and slagging, traditional coefficient is abandoned. To the heat surfaces of convection, the coefficient of enthalpy difference is advanced, and to the water-cooled walls, the coefficient of temperature difference is adopted .