Research On Prediction Of Oxygen Content In Boiler Flue Gas And Optimization Of Combustion System

The energy utilization efficiency of thermal power generation is relatively low;one of the main influence factors is that lots of important technical parameters and economic parameters are difficult to be real-time measured online, such as the temperature and oxygen content in flue gas and other thermal parameters related to boiler efficiency. Soft sensor technology is one of the effective ways to solve the problem of measuring these parameters,utilizing some parameters liable to be measured through on-line analysis to estimate these variables unable or difficult to be measured.The main study works of this dissertation focusing on the combustion system of 600MW unit in Tuoketuo Power Plant are summarized as follows:1.The characteristics of monitoring objects and thermal parameters in power plant are analyzed comprehensively. Meanwhile, the main factors which affect the accuracy of soft sensor for thermal parameters are discussed. The main factors are data preprocessing, auxiliary variables selection, modeling algorithm and model structure.2. Started with the two key technical problems" data preprocessing and auxiliary variables selection", the methods of improving the accuracy and feasibility of soft sensor for oxygen content in flue gas are focused:in one aspect of data pre-processing, the error analysis and processing of field data are considered adequately; in another aspect of auxiliary variables selection, the PCA and Partial least-Squares Regression to improve the accuracy and feasibility of the models are introduced.3. The flue gas oxygenated soft sensor model is established by using artificial neural network modeling algorithm BP, meanwhile the forecasting analysis of gas oxygen levels is implemented.4. The model of the boiler efficiency and emissions with the operating parameters of the bolier is established by using artificial neural network.Adopted by genetic algorithm, the boiler combustion system is optimized, the condition that the best operation parameters provides the basis for the closed-loop control.