论文部分内容阅读
针对非均匀周期采样系统,通过状态空间模型离散化方法得到其输入输出表达形式.鉴于参数化后得到的辨识模型同时包含1个参数向量和1个参数矩阵,利用递阶辨识原理,将辨识模型分解为分别含有参数向量和参数矩阵的2个虚拟子系统;考虑到系统的因果约束问题,将包含参数矩阵的子系统分解为子子系统进行辨识,从而提出这类非均匀采样系统的递阶最小二乘辨识方法.仿真例子表明该算法是有效的.
For the nonuniform periodic sampling system, the input and output expressions are obtained by discretization of the state space model. In view of the parameterized identification model contains a parameter vector and a parameter matrix, using the hierarchical identification principle, the identification model And decompose them into two virtual subsystems respectively containing parameter vectors and parameter matrices. Considering the problem of the system’s causal constraints, the subsystems containing the parameter matrix are decomposed into sub-subsystems for identification, and then the hierarchy of such non-uniform sampling systems Least squares identification method. Simulation examples show that the algorithm is effective.