Research On Prediction Of Boiler Oxygen And Optimal Control Of Air Supply System Based On Neural Network

At present,with the rapid development of science and technology in China,the demand for energy in major industrial fields is increasing day by day,in which the power industry is still dominated by coal-fired power generation.To improve the economic efficiency of thermal power units and meet the requirements of environmental protection is an urgent problem to be solved in thermal power plants.The oxygen content in flue gas is an important control parameter of large coal-fired units,which too high and too low will pose a threat to the safe and stable operation of the unit.Therefore,it is of great practical significance to measure the oxygen content of boiler flue gas timely and accurately and to optimize the control of oxygen content.Considering that the boiler combustion process is complex and the flue gas oxygen content has many influence variables,the traditional measuring instrument has low precision,delayed response and short service life,and the cascade PID control with fixed parameters is difficult to solve the problems of large delay and large inertia in the supply air control system.It is impossible to accurately control the oxygen.Therefore,a model predictive optimal control based on artificial neural network is proposed in this paper.This method provides a new idea for solving the problem of complex constrained optimal control.The optimal control of oxygen content in flue gas is realized by adjusting the opening of the motor blades of supply air,and finally ensure the safe,economic and stable operation of the boiler unit.In this paper,the 600 MW supercritical simulation unit developed by Huafing is taken as the research object,and the structure characteristics,operation steps and control mode of its air supply control system are analyzed in detail.In this paper,the principle and characteristics of artificial neural network and the modeling method of nonlinear system are introduced in detail,the prediction model of boiler oxygen content is established,and the off-line verification and on-line verification of the model are carried out.The experimental results show that the model has high accuracy and good dynamic characteristics.On this basis,combined with an improved particle swarm optimization algorithm,a model predictive optimal control(MPOC)scheme based on BP neural network is proposed and applied to the air supply control system.In this scheme,the opening instruction of the fan is optimized in real time in the process of variable load of the unit,and the air supply volume of the boiler is improved in order to improve the actual control effect of oxygen and optimize it.In this paper,the real-time optimal control algorithm of flue gas oxygen content is compiled based on MATLAB platform,and the simulation experiment of optimal control is carried out on 600 MW supercritical simulator.The experimental results show that the model predictive optimal control scheme proposed in this paper can greatly improve the quality of boiler oxygen control of supercritical unit and improve the response speed and regulation accuracy of dynamic process load of supercritical unit.Therefore,it can be well combined with the actual industrial production to improve the actual control effect.