摘要
提出一种基于进化神经网络的PID控制器设计方法.该控制器主要由3部分组成,第1部分应用神经网络根据控制对象的输入、输出在线调整PID控制器参数.第2部分利用进化算法根据性能指标对神经网络控制器参数进行优化,找出最优的神经网络初始权系数和比例系数.第3部分是传统PID控制器.把该控制器温度控制的仿真对照结果表明,这种控制算法具有结构简单、鲁棒适应性强、进化性能良好的特点.同时还提出一种以快速响应为目标的改进方案.
A PID controller design based on an evolutionary neural network is presented. It consists of three parts. In the first part, a neural network is used to optimize and adjust PID parameters in real time. In the second part, the parameters of the neural network are optimized by an evolutionary algorithm. The third part is a traditional PID controller. A simulation was made of a temperature control system which showed that this controller is characterized by a simple structure, robust adaptation, and good evolutionary performance. An improved scheme with more rapid response is also presented.
出处
《智能系统学报》
2008年第3期245-249,共5页
CAAI Transactions on Intelligent Systems
基金
福建省教育厅科研资助项目(K03008)
关键词
进化神经网络
进化算法
神经网络
分段控制
PID
evolutionary neural networks
evolutionary algorithm
neural networks
piecewise control
PID