Application Research Of Swarm Intelligence Algorithm In Modeling And Control Of Pulverizing System

As an important part of the combustion system of coal-fired units,the safe and stable operation of pulverizing system under the optimal parameters is very important to improve the stability and economy of the unit.Therefore,the research on modeling and control of pulverizing system has practical significance.The modeling and control problems of pulverizing system can be transformed into optimization problems.In solving optimization problems,swarm intelligence algorithm is widely used because of its distributed,robust,good scalability and extensive applicability.In this paper,genetic algorithm,which simulates biological evolution,and Particle Swarm Optimization,which simulate bird foraging,are applied to the modeling and control of pulverizing system of coal mill.Firstly,the dynamic characteristics of pulverizing system are analyzed,and the pulverizing system is regarded as a multi-variable control system with two inputs and two outputs.Then,the basic genetic algorithm,the improved genetic algorithmmulti-population genetic algorithm and the standard particle swarm optimization are introduced.The standard particle swarm optimization is easy to fall into the local optimal solution.An improved particle swarm optimization algorithm is proposed.Five common functions are selected from the classical test functions to test the improved particle swarm optimization algorithm,which verifies the feasibility and superiority of the improved particle swarm optimization algorithm.In order to obtain the mathematical model of pulverizing system,the multivariable system of pulverizing system is modeled by using multi-population genetic algorithm and improved particle swarm optimization algorithm according to the historical data of the normal operation of a power plant.The results show that the swarm intelligence algorithm can get the more accurate mathematical model of the pulverizing system.Because of the coupling between the variables of the pulverizing system,in order to eliminate the coupling between the variables,the feedforward compensation method is used to decouple the pulverizing system.For decoupled compensated systems,the controller parameters are optimized by using swarm intelligence algorithm.The simulation results show that the controller optimized by the swarm intelligence algorithm has a good control effect on the pulverizing system.