The Research On Control System Of Wet Flue Gas Desulfurization Absorber Tower Based On Single Neuron

With the rapid development of the national economy,people's requirements for the quality of life are constantly improving,and the problem of air pollution has become the focus of people's attention.Sulfur dioxide emission from coal fired power plants is one of the major atmospheric pollutants,controlling and reducing its emission is the most important task in the field of air pollution control in China.The absorber system as the most important link in the wet flue gas desulphurization system,its controlled object has the characteristics of non-linearity,large delay and large inertia.At present,the control quality of conventional PID control in field application is difficult to reach higher requirements so it needs to be further studied.In this paper,the basic working principle of wet flue gas desulfurization system is briefly described,the factors that influences desulfiurization efficiency are analyzed,the absorber slurry PH control system is selected as the research object,and the mathematical model of absorber slurry valve and absorber slurry tank is established by means of test method modeling.By analyzing the PH control system of absorber slurry,a cascade control system that based on single neuron controller was established.Single neural control is a simple neural network control,which can realize online learning and modification of weight parameters,and realize adaptive control parameters when controlling the PH of absorber slurry.Considering that it is difficult to establish accurate mathematical model for the absorber system,in order to further improve the control quality of the system,the single-neuron controller was optimized in this paper.The concept of quadratic performance index was introduced to optimize the optimization process of neuron weight parameters,so as to further improve the response speed of the control system.The design of the control program was established a simulation control model in MATLAB and do simulation research.When a unit step signal was input by the system,the two control programs are compared with PID control,the simulation results show that the optimized single neuron control can realize the rapid optimization of weight parameters,and speed up the response time of the system when ensuring no overshoot of the system.After that,a series of programs such as disturbance input and parameter change of the controlled object are carried out,and It is verified that the optimized control program has better robustness.