Fouling And Slagging On-line Monitoring Of Heat Transfer Surface In Coal-Fired Utility Boiler

In order to solve the economic and safety problems of coal-fired boiler caused by irrational soot-blowing, based on heat balance method and artificial neural network(ANN), ash on-line monitoring model for boiler heating surface was established to monitor its real pollution status for further study of soot-blowing optimization.The basic principle of heat balance method is introduced. Calculation model was different for different types of heating surface. A kind of metal and medium separate lumped-parameter dynamical correction model was adopted. Ash on-line monitoring system for a coal-fired power plant boiler is developed, which can calculate the inlet and outlet flue-gas temperature of heating surface, and obtain the ultimate pollution coefficient. Calculation results of some important parameters under several typical working conditions were recorded, and compared with the design values, which can verify the validity of the model.Based on artificial neural network(ANN), ash on-line monitoring model for boiler heating surface was established Monitoring model was different for different types of heating surface, including high-temperature convection heating surface, low-temperature convection heating surface and water wall. Application process of artificial neural network is detailedly introduced, including obtaining and screening of clean sample points, neural network structure, training and verifying. A parameter diagnosis model was established for preprocessing of sample data to remove "bad values" and guarantee the reality and continuity of thermal parameters. Ash on-line monitoring system for a300mw boiler is developed. The fouling curves of low-temperature super-heater and water wall area near a furnace soot-blower are displayed and analyzed.The design, training and predict are accomplished in Matlab because it has a strong NN function. The real time communication between Matlab and the on-line monitoring system of fouling and slagging is finished by DLL, which is written by VC. The local data-base is SQL Serve and interviewed by ODBC in order to optimize the speed of the system.