PID控制器参数优化的仿真研究

Simulation Study of the PID Controller of Optimized GA

本文对烘干炉的PID控制器进行了参数优化,并提出用遗传算法优化神经网络PID的控制方案。PID控制的控制效果取决于比例、积分和微分三种控制作用参数的合理选择,没有一定的工程经验很难在短时间内选定合适的参数,即使选定了也不一定是最优的。为此,提出用遗传算法优化控制器参数的方案。同时对于非线性复杂系统和参数可变的时变系统,PID控制器不能获得良好的控制效果,因此,考虑采用神经元(网)的PID控制器。本文针对传统PID、神经元(网)PID实现了基于二进制及实数编码的GA参数优化。仿真结果说明了所提算法的有效性。

In this paper,the PID controller parameters of the sdrying kiln are optimized,and the neural network PID controller scheme optimized by Genetic Algorithm is proposed.It is need that three control functions of the PID controller,that is proportion,integral and differential,is suitable adjusted,so that controlled system achieve better control effect.It is very difficult for us to find suitable parameters within short time as we have less experience,and even the parameters found by project experience can't guarantee that these are optimum parameters.So we propose the scheme that Genetic Algorithm is used to optimize controller parameters.For the variable parameters system and non-linear system,traditional PID controller can't get the ideal control effect.So we consider the PID controller based on the single neuron(neuron network).For the problem of parameters optimizations of traditional PID,neuron(neural network)PID controller,Genetic Algorithms(GA)based on the binary scale and real number code are realized in this paper.The simulation results show the validity of the proposed algorithms.