Analysis the robustness of control systems based on disturbance observer
Analysis the robustness of control systems based on disturbance observer
复制标题
DOI：
10.1080/00207179.2013.795663
复制
发表时间：
2013-09
影响因子：
2.1
通讯作者：
E. Sariyildiz;K. Ohnishi
中科院分区：
文献类型：
--
作者：
E. Sariyildiz;K. Ohnishi
扰动观测器（DOB）通过使用标称对象模型的逆和低通滤波器（LPF）来估计系统扰动。虽然低通滤波器提供了内环的适当性，但它是基于DOB的控制系统的主要设计约束。LPF的带宽被设计得尽可能高，使得DOB可以在更宽的频率范围内估计扰动。然而，其带宽受到噪声和系统鲁棒性的限制。鲁棒性极限与鲁棒性分析方法直接相关，并且它们显著地影响基于DOB的控制系统的性能。本文将三种不同的鲁棒性分析方法应用到基于DOB的控制系统中，并阐明了系统鲁棒性与DOB带宽之间的关系。针对传统分析方法对DOB带宽的保守性，提出了一种新的基于真实的参数不确定性的分析方法。文中对所提出的方法进行了详细的比较，并给出了仿真结果.
Disturbance observer (DOB) estimates the system disturbances by using the inverse of the nominal plant model and a low pass filter (LPF). Although the LPF provides the properness in the inner-loop, it is the main design constraint of the control systems based on DOB. The bandwidth of the LPF is designed as high as possible so that the DOB can estimate the disturbances in a wider frequency range. However, its bandwidth is limited by noise and robustness of the system. The robustness limitation is directly related with the robustness analysis methods, and they significantly affect the performance of the DOB based control systems. In this paper, three different robustness analysis methods are implemented into the DOB based control systems, and the relation between the robustness of the system and bandwidth of DOB is clearly explained. The conservatism, which is the main drawback of the conventional analysis methods, on the bandwidth of DOB is removed by proposing a new real parametric uncertainty based analysis method. The proposed methods are compared in detail, and simulation results are given to show the validation.