Recursive closed-loop subspace identification based on innovation estimation
YU Miao
WANG You-yi
GUO Ge
LIU Jian-chang
Abstract：A recursive closed-loop subspace identification method based on innovation estimation is proposed for closed-loop control systems.Firstly,the influence of past noise on future input in the closed-loop control system is eliminated by innovation estimation,and the estimation of subspace matrix parameters is obtained.Secondly,the recursive method of least squares is used to eliminate the influence of noise,which realizes the parameter estimation of the lower triangle Toeplitz matrix.Finally,based on the Kung's realization algorithm,the system matrices are extracted from Hankel matrix constructed by the estimated parameters.The simulation results show the effectiveness and superiority of the proposed method.
Keywords：subspace identificationrecursive closed-loop identificationinnovation estimationleast squares approxi-mations
Publication Date：2025-02-28
Online Publishing Date：2025-08-15(First online date of this platform, not the publication date of the document)
Pages：8( 245-252 )