Development Of SCR Denitrification System For Waste Gas Incineration And Its Predictive Control Researc

With the gradual improvement of industrial waste gas emission standards,industrial enterprises urgently need to introduce mature and stable flue gas treatment measures to dispose of waste gas in the production process.Selective catalytic reduction(SCR)is a widely used method in the field of flue gas treatment,which uses the key ammonia injection control link to achieve removal of NO_x in flue gas.However,complex chemical reaction inside SCR system and severe nonlinearity and hysteresis make it difficult for traditional ammonia injection control to achieve ideal control effect,causing flue gas emission to exceed the standard.Based on the actual industrial project,the thesis studies the ammonia injection control strategy of SCR denitration system to improves control accuracy of flue gas NO_x concentration.Based on the introduction of the reaction principle and process flow of SCR denitration technology,the thesis firstly analyzes the key parameters that affect denitration efficiency,and summarizes the problems existing in traditional ammonia injection control.In order to improve control effect of NO_x concentration at outlet of SCR system,a predictive control strategy based on radial basis function neural network is introduced to adjust the amount of ammonia injection.With the help of neural network’s nonlinear expression ability,a prediction model of the NO_xconcentration at outlet of SCR denitration is established,and based on the prediction model establishes an ammonia injection model predictive controller.By predicting NO_x concentration of the flue gas emitted by the system in the future,it calculates the optimal amount of ammonia injection at the next moment,and finally acts on the ammonia injection control valve to achieve precise adjustment of the ammonia injection.The simulation results show that the predictive control strategy of ammonia injection based on neural network is superior to conventional control methods in terms of accuracy and anti-interference performance.According to the actual SCR flue gas denitration control system requirements,this thesis has carried out the overall system plan formulation,hardware selection,communication network design,control program development,and host computer monitoring configuration.At the same time,real-time data communication between MATLAB and host computer is realized,so that the predictive control strategy participates in the actual ammonia injection adjustment.The field operation results show that the predictive control strategy of ammonia injection based on neural network can quickly and accurately complete the ammonia injection control,and it runs stably in a complex system environment,ensuring the efficiency of denitration.