基于CARIMA模型的多变量极点配置自校正直接算法及其鲁棒性

CARIMA MODEL BASED MULTIVARIABLE POLE ASSIGNMENT SELF-TUNING IMPLICIT ALGORITHM AND ITS ROBUSTNESS

本文根据 CARIMA 模型,采用双线性一步松驰算法和多项式矩阵展开交换法,给出一种适用于非最小相位系统的多变量极点配置自校正直接算法,并研究了其鲁棒性.此法计算量小,对初值选择不敏感,系统跟踪调节性能好,鲁棒性强,是一种有应用前景的极点配置直接算法。

Based on CARIMA model,a multivariable pole assignment self-tuning control implicit algorithm fornonminimum phase system is proposed,in which bilinear parameter estimation method with one step relaxationand polynomial matrices exchange based on matrix expansion is introduced.In order to increase the robustness ofthe algorithm,two strategies are also introduced.Simulation results show that the proposed algorithm is robust toparameter and structure variations of the controlled plant.It also has good tracking and regulation properties withless computing time and insensitivity to initial set values.Hence it is a new algorithm of prospects for industrial ap-plication,especially for nonminimum phase controlled system.