摘要
本文提出了一种鲁棒设计的方法。这种鲁棒设计是通过使闭环系统矩阵A+BF的Frobenius范数最小化来实现的。基本数学工具是一个R^m×n到R^m×m的映射。该映射将R^m×n空间的一个稠的开集转化为一个反馈矩阵F的集合,这个F的集合正好就是所有使得A+BF与某-A相似的F集。 本文将A+BF的Frobenius范数直接作为性能指标提出,并用梯度法,利用上述映射使之最小,从而使某一鲁棒性判据的下限最大。
A robust design method by minimizing the Frobenius norm of the closed-loop matrix A + BF is presented in this paper. The fundamental tool for this mothed is a Rm×n→Rm×n mapping function which maps a dense open set in Rm×n onto the set of all the F's that make A+BF similar to a certain A*,and the Frobenius norm of A+BF is directly proposed as a performance index. This index is minimized by gradient method, thus the lower bound of a robustness measure is maxized.
出处
《控制与决策》
EI
CSCD
北大核心
1989年第6期24-29,共6页
Control and Decision
