|
|
中国电机工程学报 2007, 27(35) 96-101 DOI:
ISSN: 0258-8013 CN: 11-2107/TM |
|
|
|
|
本期目录 |
下期目录 |
过刊浏览 |
高级检索
[打印本页]
[关闭] |
|
| 发电 |
|
|
协调控制系统神经网络PID优化控制与仿真研究 |
|
|
王爽心 杨辉 李亚光 |
|
|
北京交通大学机械与电子控制工程学院 北京交通大学机械与电子控制工程学院 华电煤业集团有限公司 |
|
|
摘要：
针对火电单元机组协调控制系统具有多变量、强耦合、非线性及参数时变的特点，将对角递归神经网络与PID控制方法相结合，并利用提出的改进型变尺度混沌优化策略对神经网络的权值参数和PID控制器参数进行整定，从而实现多变量系统的优化控制。此方法不仅保持了传统PID控制器结构简单、算法实用等特点，而且算法稳定性好，寻优效率高，避免了控制器参数陷入局部极小等问题。仿真结果表明，对于100%和70%不同负荷时的工况，即使对象的传递函数发生了较大的改变，系统仍具有响应速度快、鲁棒性好、自适应性好等特点。 |
|
|
关键词：
对角递归神经网络
协调控制系统
混沌优化
PID
|
|
|
Research on Neural Network PID Optimization and Simulation of Coordinated Control System |
|
|
|
|
|
|
|
|
Abstract:
Aiming at the characters such as multivariable, strong coupling, nonlinear and time-varying parameters for coordinated control system of fossil-fired power unit, diagonal recurrent neural network combined with PID control theory was researched. Meanwhile, an improved mutative scale chaotic optimization algorithm was proposed for tuning weight parameters of neural network and PID parameters to realize multivariable optimal control. The new control strategy not only retains the characters of the traditional PID, such as simple structure and practicable control, etc, but also shows the algorithm is of high stability and searching efficiency; it avoids the problem of controller parameters trapping into local minimum. Simulation results show that under different load conditions of 100% and 70%, even transfer functions change greatly, the system also has many advantages such as fast response, strong robustness and good self-adaptability. |
|
|
Keywords:
diagonal recurrent neural network
coordinated control system
chaotic optimization
PID
|
|
|
收稿日期 2007-06-22 修回日期 1900-01-01 网络版发布日期 |
|
|
DOI: |
|
|
基金项目:
|
|
|
通讯作者: 王爽心 |
|
|
作者简介: |
|
作者Email: shxwang1@bjtu.edu.cn;wsx@jdxy.njtu.edu.cn |
|
|
| 参考文献： |
|
| 本刊中的类似文章 |
| 1．胡雪峰 谭国俊.应用神经网络和重复控制的逆变器综合控制策略[J]. 中国电机工程学报, 2009,29(6): 43-47 |
|
| Copyright by 中国电机工程学报 |
