论文部分内容阅读
本文研究多变量线性系统的结构和参数辨识.利用解不定对称系数方程组的方法,导出了搜索系统结构特征值的最小二乘递推算法.并提出了一个判断系统结构的新准则——系统特征方程系数符号检验法.把两者结合在一起,构成了完整的多变量系统结构和参数辨识的递推最小二乘法,并给出了仿真例子. 本方法也可作为最小二乘的各种改进算法(如广义最小二乘法等)的基础.
In this paper, we study the structure and parameter identification of multivariable linear systems. By using the solution of unsymmetrical equations of symmetry, the least-squares recursive algorithm for the eigenvalues of search system structures is derived and a new criterion for system structure-system Eigenvalue coefficient sign test method combines the two together to form a recursive least square method of complete multivariable system structure and parameter identification and gives the simulation example.The method can also be used as the least squares Improve the algorithm (such as generalized least squares, etc.) basis.