论文部分内容阅读
A novel subspace identification method based on correlation function which estimates a state-space system dynamics of unknown plant operating in closed-loop experimental condition is proposed in this paper.It is shown that the cross-correlation function of the output and external input signals are equal to the crosscorrelation function of the input and external signals filtered through the system dynamics since noise signal has no correlation with the external input.The proposed algorithm is developed to obtain unbiased estimates of system matrices based on time-shifted invariance of the correlation function estimates.Later the algorithm is compared to other popular subspace methods in the simulation study and the results show the effectiveness of our method in the presence of colored noise and low signal-to-noise ratios.
A novel subspace identification method based on correlation function which estimates a state-space system dynamics of unknown plant operating in closed-loop experimental condition is proposed in this paper. It is shown that the cross-correlation function of the output and external input signals are equal to the crosscorrelation function of the input and external signals filtered through the system dynamics since noise signal has no correlation with the external input. The proposed algorithm is developed to obtain unbiased estimates of system matrices based on time-shifted invariance of the correlation function estimates .Later the algorithm is compared to other popular subspace methods in the simulation study and the results show the effectiveness of our method in the presence of colored noise and low signal-to-noise ratios.