The Study On System Of The On-line Monitoring Of Fouling And Slagging On Utility Boiler Heat Surfaces

In this paper, a suit of on-line monitoring system of fouling and slagging on utility boiler heating surfaces is improved on the basis of existing method, which can offer strong guarantee for security and economical operation of the unit. The'Oxygen'is included in operational parameters of the unit, making for practice. The dirty parameter of dust is chosen in the study. By applying the improved arithmetic called BP neural network, using the data from the DAS of the plant, the neural network trains samples to predict the heat-flux of the clean surface in all cases, while the factual one in same case is got by compute. Then the dirty parameter of dust is acquired through its definition. Newton iterative arithmetic is programmed for obtaining the factual heat-flux of the water wall, that is classical. In this paper, data is divided into Real-time and historical data. SQL Server Database and the ODBC method are applied in local database, improving the running of the system. At the same time, taking exactness and error of samples into account improves its precision.