基于子空间的多率控制系统闭环辨识

Subspace-based Closed-loop Identification of Multirate Control System

  • 摘要: 针对一类多率控制系统,在控制器信息未知且反馈存在的情形下,采用子空间辨识方法建立闭环系统中被控对象的离散状态空间模型.在模型参数化的过程中,为了避免由于提升技术导致的非因果项列入辨识模型中,通过因果约束分析得到广义因果模型.基于状态辅助变量关系,构造最小二乘问题并利用奇异值分解求解状态序列估计,同时给出了系统矩阵的计算方法.最后,通过数值仿真实验验证了提出方法的有效性.

     

    Abstract: In accordance with a class of multi-rate control systems, a subspace identification method is used to identify the discrete state-space model of the controlled object in closed-loop systems under feedback while there is no information about the controller. In order to avoid the inclusion of non-causal terms in the identified model, the generalized causal model is obtained through the analysis of causality constraints during the process of model parameterization. Then, based on the relationship of the instrumental variable, the least squares problem is developed and the state sequence estimation is solved by singular values decomposition. Meanwhile, the computation of the system matrices is given. Finally, the numerical simulation illustrates the effectiveness of the proposed method.

     

/

返回文章
返回