摘要
PID控制因其算法简单,被人们广泛应用到控制之中,包括可建立准确的数学模型的控制系统。但是,在电动变量施肥控制系统的运行过程中,常常会出现非线性和时变不确定性的问题,很难建立起准确的数学模型。在使用常用的PID控制器时,其参数往往整定不够好,性能欠佳,在运行过程中的适应能力也很差,不能达到理想的控制效果。为此,提出了一种基于遗传神经网络算法整定PID的方法,采用MatLab/Simulink软件建立传递函数仿真模型,之后分别采用RBF-PID和基于遗传算法优化的RBF-PID进行仿真对比。结果表明:基于遗传算法优化的RBF神经网络PID整定的电动变量施肥控制系统稳定性好,精度更高,具有更强的鲁棒性。
PID control is widely applied to control due to its simple algorithm,including the control system that can establish accurate mathematical model.But,in the process of the operation of the electric control system of variable rate fertilization,often there will be a problem of nonlinear and time-varying uncertainty,therefore,it is difficult to establish accurate mathematical model,in the use of commonly used PID controller,the parameter setting is not good enough,performance is poor,often in the process of running ability to adapt is also very poor,so always can’t achieve ideal control effect.Considering the disadvantages of conventional PID controller,a method of PID tuning based on genetic neural network algorithm is proposed.MATLAB/Simulink software was used to establish the transfer function simulation model,RBF-PID and RBF-PID based on genetic algorithm optimization were used for simulation comparison.The results show that RBF neural network PID tuning based on genetic algorithm is more stable,precise and robust.
作者
黄丽萍
梁春英
王振民
于天齐
曲云霞
Huang Liping;Liang Chunying;Wang Zhenmin;Yu Tianqi;Qu Yunxia(College of Electrical and Information,Heilongjiang August First Land Reclamation University,Daqing 163319,China)
出处
《农机化研究》
北大核心
2020年第10期32-36,共5页
Journal of Agricultural Mechanization Research
基金
“十三五”国家重点研发计划项目(2016YFD0200600).
关键词
电动变量施肥
RBF-PID
参数优化
遗传算法
variable fertilizer
RBF-PID
parameter optimization
genetic algorithm
rbf-pid
作者简介
黄丽萍(1995-),女,河北满城人,硕士研究生,(E-mail)hlp1107@foxmail.com。;通讯作者:梁春英(1971-),女,山东商河人,教授,硕士生导师,博士,(E-mail)ndliangchunying@163.com。
