RUS  ENG
Full version
JOURNALS // Upravlenie Bol'shimi Sistemami // Archive

UBS, 2018 Issue 76, Pages 6–25 (Mi ubs973)

This article is cited in 9 papers

Analysis and Synthesis of Control Systems

Estimation of external disturbances derivatives based on virtual dynamic models

S. A. Krasnovaab

a V. A. Trapeznikov Institute of Control Sciences of Russian Academy of Sciences, Moscow
b Bauman Moscow State Technical University

Abstract: The method of synthesis of invariant systems in the presence of a dynamic model, that simulates the effect of external disturbances, and the construction on its basis of a dynamic compensator or disturbances observer, is most developed in the theory of automatic control. An alternative method for estimating external influences that does not require the extension of the state space is to use state observers with discontinuous corrective actions functioning in the sliding mode or their continuous analogues. Within the framework of this approach, the original method for estimating the derivatives of external disturbances based on virtual dynamic models composed by sequential differentiation of a part of the derivative of state variables of the control plant, to which the estimated disturbances acting, is proposed. Virtual models are not introduced into the regulator, only dynamic observers of disturbances derivatives are present in the feedback loop. Observers are built as replicas of virtual models. Procedure for cascade synthesis of observers of disturbances derivatives with piecewise linear correcting actions is developed. In this procedure, the motions separation method in virtual space of observation errors is realized. And observer's variables, and their corrective actions, serve as estimates of the derivatives of external disturbances, and they can be used to synthesize combined feedback and/or current diagnostics of the control process.

Keywords: nonlinear systems, external disturbances, disturbances observers, cascade approach, motion separation method.

UDC: 62.50
BBK: 32.817

Received: May 11, 2018
Published: November 30, 2018

DOI: 10.25728/ubs.2018.76.1



© Steklov Math. Inst. of RAS, 2024