在线优化参数的多变量无模型预测线性自抗扰控制
Multivariable model-free predictive active disturbance rejection control of on-line parameter optimization
侯小秋 1李丽华1
作者信息
- 1. 黑龙江科技大学 电气与控制工程学院,哈尔滨 150022
- 折叠
摘要
针对时变参数的多变量线性模型上叠加非线性的系统,采用具有辅助向量的多变量偏格式动态线性化方法逼近该系统,在多变量线性状态观测器中加入辨识的线性模型,提高状态观测精度.在多变量线性跟踪微分器和状态观测器分别引入超前参考输入与输出预测值,使多变量线性自抗扰控制器具有预测控制性能.运用非线性递推最小二乘法在线优化多变量线性自抗扰控制器的参数,给出在线优化参数的多变量无模型预测线性自抗扰控制算法.结果表明:无参数优化时的响应曲线,超调大,调节时间约1s,无稳偏;有参数优化时的响应曲线,超调较小,调节时间约0.65 s,无稳偏,控制算法的响应优良.
Abstract
This paper is focused on the solution to the overlying nonlinear system in multivariable lin-ear model of time-varying parameter.The study consists of approaching to the system by using the multi-variable partial format dynamic linear method with auxiliary vector;adding the identifiable linear model to the multivariable linear state observer for improving observation accuracy;introducing the input and out-put predicative values for reference ahead into the multivariable linear tracking-differentiator and state ob-server to enable the multivariable onlinear active disturbance rejection controller to have the predicative control function;and by using nonlinear recursive least squares method,optimizing the parameters of the multivariable linear active disturbance rejection controller,and providong the algorithm of multivariable model-free predictive active disturbance rejection control of parameters optimization on-line The results show that the response curve without parameter optimization presents big overshoot within setting time of 1 s,and withno steady deviation;and the response curve with parameter optimization presents smaller o-vershoot within the setting time of 0.65 s,thus,this control algorithm has excellent control response.
关键词
多变量线性自抗扰控制/无模型自适应控制/预测控制/非线性系统/在线优化参数Key words
multivariable linear active disturbance rejection control/model-free adaptive control/predicative-control/nonlinear system/parameters optimization on-line引用本文复制引用
出版年
2024
国家科技期刊平台
NSTL
万方数据