The Boiler Tube-slagging Prediction Model And Its Application

Power plant boiler slagging has not yet been satisfactorily resolved which has been plagued by the power industry, as the thermal power installed capacity is more and more big, the slagging problem will become more severe as a potential risk of the safe and economical operation of the unit. So accurate judgment of the boiler furnace slagging, make corresponding accident plans and advance prevention have the extremely vital significance to prevent the accident of serious economic operation,In recent years, many experts and scholars at home and abroad have researched on the characteristics of coal-fired boiler slagging and put forward a prediction method of thermal power units of the degree of slagging, In general, all kinds of existing prediction evaluation methods and the actual situation are not a perfect match. Therefore, this subject is to judge the accuracy of various methods from the study, analyses the actual slagging reality and describeds the field of thermal power units of research in recent years and new research progress.which made the non professional personnel have a clear understanding to know the slagging judgement of each index. At the same time, we analyzed the advantages and disadvantages of various research methods and pointed out its accuracy and its causes. Then, we analyzed the influence of the power station boiler slagging factors to determine evaluation boiler slagging tendentiousness indicators. We combine the fuzzy mathematics with BP neural network to constitute a fuzzy neural network model. Membership function uses nonlinear function and choses the actual power plant boiler coal as the training sample data, combines with the characteristics of coal quality of conventional discriminant index in the fuzzy clustering method, and integrates into the furnace aerodynamic field in the furnace,internal temperature, and local climate effects on furnace slag.The paper proposes that used softening temperature t2, alkaline acid ratio B/A, silicon aluminium ratio SiO2/Al2O3, silicon ratio G, furnace air coefficient aj and two dynamic index:dimensionless furnace to the average temperature (?), dimensionless actual diameter of the inscribed circle (?)d as evaluation index, which through considing the conventional discriminant index of the coal characteristics, and the aerodynamic field inside the furnace, the furnace internal temperature and the local atmosphere to the influenced of slagging, in the end, used the fuzzy clustering and fuzzy neural network as two ways to establish the evaluation model for slagging forecast Shanxi Datang Yungang Power limited liability company burning coal units.