进化神经网络PID控制器的研究与应用

PDF下载HTML

Title:

An improved PID controller based on an evolutionary neural network

文章编号:

1673-4785(2008)03-0245-05

作者:

邱建斌; 王劭伯;

福州大学电气工程与自动化学院; 福建福州;

Author(s):

QIU Jian-bin; WANG Shao-bo

College of Electrical Engineering and Automation; Fuzhou University; Fujian 350002; China

关键词:

进化神经网络;  进化算法;  神经网络;  分段控制;  PID

Keywords:

evolutionary neural networks;  evolutionary algorithm;  neural networks;  piecewise control;  PID

分类号:

TP13

文献标志码:

A

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.

Abstract:

A P ID controller design based on an evolutionary neural network is p resented. It consists of three parts. In the first part, a neural network is used to op timize and adjust P ID parameters in real time. In the second part, the parameters of the neural network are op timized by an evolutionary algorithm. The third part is a traditional P ID controller. A simulation wasmade of a temperature control system which showed that this controller is characterized by a simp le structure, robust adap tation, and good evolutionary performance. An imp roved scheme with more rap id response is also p resented.