Research On Modeling And Optimal Control Of SCR Denitrification System In Coal-fired Power Plants

Selective Catalytic Reduction(SCR)has become a widely used NOx treatment method in coal-fired power plants in China because of its high denitration rate,reliable technology and simple structure.Therefore,establishing an accurate SCR denitration system model and realizing optimal control is of great significance to denitration treatment in coal-fired power plants.In this paper,a coal-fired power plant in Xinjiang is taken as the research object,combined with the actual working condition data,the SCR denitration system is modeled by mechanism and data modeling methods,and on this basis,the optimal control strategy is put forward.Specific research contents are as follows:1.According to the operation characteristics of SCR denitration system,the SCR denitration system model is established by mechanism modeling method.Firstly,the collected historical production data of coal-fired power plants are preprocessed;Secondly,the mechanism model of SCR denitration system is established according to kinetic differential equation.Finally,Particle Swarm Optimization(PSO)is used to identify the parameters of the mechanism model,and the stability and generalization of the mechanism model are verified.The results show that the mechanism model can reflect the operation characteristics of SCR denitration system.2.Aiming at the large amount of historical data produced in the production process,a softsensing modeling method of NOx emission based on adaptive fuzzy neural network(AFNN)is proposed by using data modeling method and guided by soft-sensing technology.Firstly,soft sensor modeling of NOx emission is analyzed,and greyrelational analysis(GRA)method is introduced to select auxiliary variables;Secondly,the soft-sensing model of NOx emission based on AFNN is established.Finally,the effectiveness of the model is verified by comparative simulation experiments.The results show that the soft sensing method based on AFNN has higher precision and better generalization ability.3.Aiming at the optimal control problem of SCR denitration system,based on the previous modeling,a model predictive control(MPC)strategy based on adaptive fuzzy neural network is proposed.Firstly,AFNN-MPC control framework is built by using MPC theory system.Secondly,numerical simulation shows that AFNN prediction model meets the needs of MPC.Finally,using the field measured data,the effectiveness of the optimal control is verified by simulation experiments.The results show that the NOx emission at the outlet of SCR denitration system is kept in a stable range,and the denitration efficiency is improved,which basically realizes the optimal control of SCR denitration system.The research on modeling and optimization control of SCR denitration system has important practical significance.The modeling and optimization control method proposed in this paper can meet the requirements of SCR denitration system operation characteristics monitoring and can assist the actual production process.