Power Station Boiler Combustion Condition Of Feature Extraction And Analysis

China is a big country that produces coal, coal accounted for about70%of once energy structure, such that the energy structure desides China’s electric power industry in the future for a long time, still gives priority to with coal fired thermal power as the core. In recent years, the price of coal rose sharply and pollution emission index restricted context, plant as energy consumption, especially in the rapid development of the national economy on electricity demand increased dramatically circumstances improve boiler combustion efficiency and reduced the emission of pollutants has become a burning optimization goal, based on the power plant boiler combustion feature extraction and analysis of boiler combustion optimization, it is the great significance to ensure security unit operation safety, stability, economy, environmental protection.Reflect the combustion condition signal of chamber pressure, combustion efficiency, the central position of flame boiler furnace outlet, oxygen and grindability index, for boiler combustion process of flame center parameters not measurable or flue gas oxygen content measurement of high cost hard problems, this paper suggests that we should use the power plant supervisory information system (SIS) database, in the hearth structure space coordinates, through the database of radiant energy signal is combined in the heat transfer distance of the inverse square law of four corner tangentially fired boiler flame central position of structural nonlinear equation model, and Matlab software to solve the model to estimate the flame central position; flame centre height by four corner tangential flame center and steam-water heat absorption, boiler load in relation to each other, through mathematical analysis and mechanism analysis to verify the flame center, horizontal offset with furnace flue gas temperature on both sides of the temperature difference test. Oxygen content in the smoke can not only adjust the furnace combustion flame position and combustion condition, through mechanism analysis to determine the impact of flue gas oxygen content of process variables, based on the support vector machine to establish flue gas oxygen content soft measurement model, support vector machine soft measurement modeling method to satisfy the requirement of measurement result is more accurate, then need to optimize the soft measurement model. The pursuit of model prediction accuracy can be selected and output all the related factors, but reduces the model to predict the velocity, also increased the complexity of the model. In order to solve this problem, this paper joins the main element analysis method, so that the model prediction can meet the requirements of accuracy and complexity.