多项式非线性系统的并行分布辨识  

Parallel distributed identification of polynomial nonlinear systems

作　　者：王建宏[1] 熊朝华[1] 许莺[1] 徐欣[1] 

机构地区：[1]中国电子科技集团公司第二十八研究所信息系统工程重点实验室,南京210007

出　　处：《控制与决策》2016年第5期889-894,共6页Control and Decision

基　　金：国家自然科学基金项目(61402426)

摘　　要：在状态空间方程中引入输入和状态的多项式函数,以此多项式函数表示非线性因素.为了辨识多项式非线性系统中的各系统矩阵,对于矢量化各系统矩阵组成的未知参数矢量,分别在无约束和有约束条件下采用两并行分布算法求解.在以状态方程等式为约束条件时,将各状态瞬时刻值与由系统矩阵组成的未知参数矢量合并为一个新的优化矢量.对于优化矢量的辨识,给出了并行分布算法的求解过程和迭代式.最后,通过仿真算例验证了所提出方法的有效性.Polynomial functions about inputs or states are added in the state space model, and the polynomial functions are used to denote the nonlinear factors. To identify every system matrix in the polynomial nonlinear state space model,each system matrix is vectorized as an unknown parameter vector, and two parallel distribution algorithms are applied in unconstrained or constrained conditions respectively. When some state equation equalities are deemed as the constrained conditions, the new optimization variables include the state instants and the unknown parameter vector which is consisted by all system matrices. These optimization variables are identified by using the parallel distribution algorithm, and the whole process about the parallel distribution algorithm are given explicitly. Finally, the simulation example is used to illustrate the effectiveness of the proposed parallel distribution algorithm.

关 键 词：系统辨识 多项式非线性系统 状态空间模型 并行分布算法 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置]