http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
Jian Pan,Sunde Liu,Jun Shu,Xiangkui Wan 제어·로봇·시스템학회 2022 International Journal of Control, Automation, and Vol.20 No.12
This paper considers the parameter identification problems of a Volterra nonlinear system. In order to overcome the excessive calculation amount of the Volterra systems, a hierarchical least squares algorithm is proposed through combining the hierarchical identification principle. The key is to decompose the Volterra systems into three subsystems with a smaller number of parameters and to estimates the parameters of each subsystem, respectively. The calculation analysis indicates that the proposed algorithm has less computational cost than the recursive least squares algorithm. Finally, the simulation results indicate that the proposed algorithm are effective for identifying Volterra systems.
Jian Pan,Huijian Zhang,Hongzhan Guo,Sunde Liu,Yuqing Liu 제어·로봇·시스템학회 2023 International Journal of Control, Automation, and Vol.21 No.5
This paper focuses on the identification of a multivariable controlled autoregressive-like (CAR-like) system. A joint identification algorithm of stochastic gradient and least squares is deduced for estimating the system parameters by decomposing the multivariable CAR-like system into two subsystems, which avoids the calculation of the matrix inversion. To further improve the parameter estimation accuracy, a joint identification algorithm of hierarchical multi-innovation stochastic gradient and least squares is proposed by using the multi-innovation identification theory. The simulation results confirm that these proposed algorithms are effective.