基于输出信息的空间互联系统迭代学习控制

Iterative Learning Control for Spatially Interconnected Systems Based on Output Information

针对一类具有周期性互联特性的状态不可测空间互联系统,提出了一种基于输出信息的有限频域迭代学习控制算法.通过对多个子系统耦合的离散空间模型变量的提升,将二维空间互联系统转化为一维等效模型.进而利用输出信息构建迭代学习控制律,将被控系统转换为等价的重复过程模型,然后基于广义KYP引理,将保证离散重复过程在不同频率区间内沿批次稳定和跟踪误差单调渐近收敛的性能指标条件转化为线性矩阵不等式.最后通过有源梯形电路的控制仿真验证了本文所提算法的有效性.

We propose finite-frequency iterative learning control algorithm based on output information herein for a class of state-unmeasurable spatially interconnected systems with periodic interconnection characteristics.We transform the two-dimensional spatially interconnected systems into one-dimensional equivalent model by lifting the variable of the discrete space model coupled with multiple subsystems.Then,we use the output information to construct an iterative learning control law,which converts the controlled system into an equivalent repetitive process model.Subsequently,on the basis of the generalized KYP lemma,the performance index conditions that guarantee the stability of discrete repetitive process along the trail and the monotone asymptotic convergence of tracking errors in different frequency intervals are transformed into linear matrix inequality.Finally,the effectiveness of the proposed algorithm is demonstrated through the control simulation of the active ladder circuts.