Autonomous Decision Control Of Variable Load Pulverizing System Of Unit

In recent years,with the proposal of the concept of intelligent power generation,higher requirements have been put forward for coal-fired power units.It requires coal-fired power units to integrate technologies such as intelligent sensing and execution,intelligent control and optimization,intelligent management and decisionmaking.It should be achieved a mode of power generation control and management with self-learning,self adaptation,self optimization,self recovery and selforganization.To realize the "few people even no people working" of coal-fired power units.However,in the environment of new energy power system,the external environment faced by coal-fired power units is even worse.In order to stabilize the random volatility of new energy power such as wind power and photovoltaic power,coal-fired power units,which still occupy a dominant position in China’s energy structure,still need to frequently participate in power frequency regulation and peak shaving.In this external environment,in order to realize the "few people even no people working " of coal-fired power units,it is necessary to greatly improve the automation degree of the current control system,that is,to improve the autonomous decision and control ability of the control system.In the process of frequent load rise and fall of power units,the focus of its regulation is to realize the coordination among air,coal and water.The regulation of coal quantity is related to the combustion state of the whole boiler and the balance of energy supply and demand with the steam turbine.Due to the limitation of the output of the coal mill,the pulverizing system needs to be started and stopped frequently in the process of load rise and fall,which not only brings a great burden to the operators,but also the different start and stop times of the coal mill and the different operation combination of the pulverizing system will directly change the combustion state in the furnace.Besides,it also might affect the heat absorption ratio of each heating surface of the boiler and the economy of boiler operation.Therefore,how to realize the automatic start and stop control of the pulverizing system,the autonomous decision of coal mill switching and the optimal scheduling between coal mills are of great significance to improve the safe and economic operation of boiler combustion system and realize the "few people even no people working" of coalfired power units.Based on the analysis of the start-up and stop process of the pulverizing system,in order to realize the rapid start-up and stop,this paper designs the APS control logic strategy of the pulverizing system,which lays the foundation for the independent decision-making control of the pulverizing system;Then,according to the each layer output of the pulverizing system and relevant input variables of boiler,the boiler combustion system model which can reflect the heat absorption of each heating surface,the exhaust gas temperature and the NOx at the outlet of the boiler is constructed through the sequential neural network.Combined with the daily scheduling curve of unit load and the standby capacity of coal mill,the best time to start and stop the mill is determined.Based on the neural network model of boiler combustion system,aiming at the balanced heat distribution of boiler heating surface,particle swarm optimization scheduling algorithm is adopted to obtain the mill combination scheme and optimal scheduling strategy of each layer under different working conditions.The results show that the Automatic start-up and shut down control strategy of the pulverizing system designed in this paper can realize the rapid start of the pulverizer in 6 minutes and 30 seconds,which lays a foundation for the independent decision-making control of the pulverizing system;According to the analysis of input/output variables of boiler combustion system,the boiler combustion system model constructed by LSTM neural network algorithm has high simulation accuracy and can support the optimization of mill group and coal mill output optimization based on the model;Taking the heat absorption ratio of each heating surface of the boiler,the exhaust temperature of the boiler and the NOx emission at the outlet as the objective function and the minimum adjustment amount of the coal output change as the constraint,the optimal mill group mode and the optimal distribution scheme of the coal mill output under different load conditions are obtained through the particle swarm optimization algorithm.From the simulation results,it can be seen that the optimization strategy in this paper can ensure that the heat absorption ratio of each heating surface of the boiler is more reasonable to a certain extent,avoid overheating of partial heating surfaces inside the boiler,and better improve the safety of the whole boiler combustion system without affecting the boiler combustion economy and emission indicators.