非均匀周期采样多率系统的一种辨识方法

本文利用提升技术,推导了非均匀采样多率系统的提升状态空间模型.对于状态可测量的多率系统,利用最小二乘原理,给出了模型参数矩阵辨识方法;对于状态不可测的未知参数多率系统,利用递阶辨识原理,在考虑提升模型的因果约束下,将提升系统分解为子系统进行辨识,形成了状态空间模型递阶辨识方法.仿真例子表明,本文提出的递阶辨识方法是有效的.

Identification of Non-Uniformly Periodically Sampled Multirate Systems

For non-uniformly periodically sampled multirate systems,we derive the corresponding lifted state-space models by using the lifting technique.Furthermore,for multirate systems whose states are measurable,an identification method to estimate the lifted system parameter matrices is given by using the least squares principle;for multirate systems whose states are unmeasurable,the lifted multirate systems are decomposed into some subsystems according to the hierarchical identification principle,and the hierarchical state-space model identification methods are developed,which take causality constraints into consideration.The simulation results indicate that the proposed algorithms are effective.