Spraying Ammonia Intelligent Optimization Methods Of Coal-fired Boiler SCR Systems

With the development of energy and environment, the problems of Haze and acid rain have become increasingly prominent. Atmospheric pollution’s prevention and control becomes important issue of energy production. NOx is one of main factors leading to the atmospheric pollution in the flue gas. The emission standard of NOx has been stipulated strictly in China, but due to de-NOx technology started late, the problems of immature theory, poor denitration efficiency and so on exist in industrial applications. Therefore, the study of de-NOx technology is important. This paper unfolds the research aiming at the control of spraying ammonia as follows.It has summarized denitration method at present, and mainly introduced the reaction process and reaction theory of Selective Catalytic Reduction. Also it has analyzed the impact of factors affecting the efficiency of denitration and the impact of de-NOx system to the boiler. It lays a foundation for the optimal control of spraying ammonia in SCR system.By deeply analysis on the control of fixed molar ratio and the control of export value NOx, it has been modeled and simulated for the latter. Experimental results indicate, the model can better reflect the dynamic property of field equipment, but it is difficult to overcome the SCR system lags and nonlinearity with the traditional PID controller. This results in the increase of denitration cost and secondary pollution.Based on the theory of Support Vector Machine and modeling method, using it to the model of power plant. Combining the characteristics of SCR system and reaction mechanism, the input variables of SVM model were screened, while the data’s preprocessing, effectively reducing the dimension of the model, the model’s complexity and improving the accuracy and robustness of the model. Optimal control problem for the SCR system in the amount of spraying ammonia which based on the traditional PID control system of thermal power plants, introducing the model based on support vector machine spraying ammonia compensation controller to control the NOx emission to meet national standards while optimizing the amount of ammonia spray. Simulation results show that this method is correct and effective on the control of NOx emission, and decreases the amount of spraying ammonia. So it has a certain value in engineering.