H_∞ optimal design of robust observer against disturbances

Abstract

This paper considers the robust observer design problem for linear dynamic systems subject to the interference of external disturbances. For such systems, the state estimate from the conventional Luenberger is normally biased with respect to the true system state. To remedy this situation, this paper proposes a new structure for robust observers. With this new structure, the robust observer design problem is skillfully transformed into the well-known disturbance rejection control problem. The H ∞ optimal control design technique can then be applied to shape the proposed robust observer in the frequency domain. The proposed robust observer is a joint state and disturbance observer, which simultaneously estimates both the system state and unknown disturbances, and can be applied to non-minimum-phase systems.

Event‐triggered composite anti‐disturbance dissipative control for nonlinear systems with multiple noises and quantized inputs via nonlinear disturbance observer

“…then, there exists an ECADC law (17) based on NLDO ( 6)- (7) such that the augmented system (19) is asymptotically mean-square bounded stable when w k ≡ 0.…”

“…It is worth noting that the DOBC technique has been proven to be effective in practical applications, such as trajectory tracking control of surface vehicles 5 and spacecraft formation flying. 6 To deal with multiple noises in nonlinear systems, the DOBC can be combined with the existing disturbance attenuation control technique, such as H ∞ control, 7 fuzzy control, 8 back-stepping control, 9 stochastic control, 10 and adaptive control 11 to form a composite anti-disturbance control scheme for multi-source interference systems. It should be pointed out that the above results are developed to deal with disturbances modeled with the exosystems, and thus are not suitable to deal with the nonlinear ones extensively studied in many fields.…”

Robust active disturbance rejection controller design to improve low‐voltage ride‐through capability of doubly fed induction generator wind farms

“…A traditional proportional-differential (PD) controller is used that can help to reach the goal of minimising the effect of disturbance in output. Therefore the control law U 0 (s) is chosen as (36), where R (s) is a reference input [29][30][31][32]…”

“…This process can be demonstrated by (35), where U 0 ( s ) is the control law for regulating Y ( s )A traditional proportional‐differential (PD) controller is used that can help to reach the goal of minimising the effect of disturbance in output. Therefore the control law U 0 ( s ) is chosen as (36), where R ( s ) is a reference input [29–32]where k P is the proportional gain and k D is the derivative gain of the PD controller.…”

Improved disturbance observer-based control for multivariable processes with time delays based on modified inverse model

“…Moreover, many control strategies designed in conjunction with DOB further prove the effectiveness of DOB in disturbance rejection (Cao and Chen, 2014; Huba, 2013; Kadam and Waghmare, 2013; Jin and Jia, 2017a). However, it is noteworthy that, in the existing literature about DOB control, most of them focus on the single-input-single-output (SISO) processes (Chen and Chen, 2014; Sariyildiz and Ohnishi, 2013, 2014; Wang and Su, 2015; Zhu et al, 2005). Two critical problems in designing DOB are the computation of the inverse model and design of the filter Q ( s ) .…”