Research On Wet Flue Gas Desulfurization Control System Of Thermal Power Plant

In 2020,the country proposed the slogan of"carbon neutral"for energy saving and emission reduction,under the background of"carbon neutral"concept,more and more Chinese people began to enhance the awareness of environmental protection.Coal-fired power plants lead the way in the composition of China’s electric energy,but the burning of coal produces large amounts of SO₂ and other air pollutants.In terms of current technology,in order to solve the contradiction between the growing demand for electricity and environmental protection,the use of limestone-gypsum wet flue gas desulfurization technology is a more effective technical means to reduce SO₂emissions.The desulfurization efficiency is the key parameter to measure the desulfur-ization process.Among the many determinants of the desulfurization efficiency,the absorption tower slurry PH value has the greatest influence and directly affects the safety and economy of the flue gas desulfurization system.However,in the actual operating environment,PH value measurement instrument working environment is harsh,prone to interference,often clogging,failure and other problems,when the general use of redundant equipment or sampling test and other methods for temporary detection,but these methods have high costs,time lag and other problems.In response to the above problems,this paper investigates two aspects:the prediction model and the control method of the slurry PH value of the absorption tower,in order to improve the desulfurization efficiency and thus control the desulfurization process rationally.The main work is as follows:Firstly,the composition and process flow of limestone-gypsum wet flue gas desulfurization system are analyzed,followed by determining the influencing factors of slurry PH and analyzing the desulfurization system process,so as to provide the theoretical basis for the system modeling and optimal control methods.Next,the prediction models of slurry PH value were established by using extreme learning machine（ELM）and long short-term memory network（LSTM）,respectively.For the problem that the LSTM neural network with strong generalization ability has difficulty in parameter rectification,the sparrow search algorithm（SSA）and the improved sparrow search algorithm（t SSA）were used to optimize the LSTM network parameters,and the prediction models of slurry PH value based on SSA-LSTM and t SSA-LSTM were established in turn.Once again,the validity of the slurry PH prediction model was verified.The pre-processing of field data mainly includes coarse error processing,random error processing,and then the Pearson coefficient method is used to determine the final input parameters.The established slurry PH prediction model is simulated and validated on MATLAB platform,and the simulation results show that the prediction model based on t SSA-LSTM has better prediction effect.Finally,for the shortcomings of the conventional cascade control for coal-fired power plants,the sparrow search algorithm-based cascade control and the improved sparrow search algorithm-based cascade control are designed respectively,and the parameters of the main controller are optimized.The simulation results of the slurry PH value of thermal power plant in MATLAB show that the control effect of the improved sparrow search algorithm cascade control is significantly better than that of the traditional cascade control and the sparrow search algorithm cascade control.Figure[61]Table[11]Reference[80]...