Optimal Design of Residual-Driven Dynamic Compensator Using Iterative Algorithms With Guaranteed Convergence

A stabilizing regulator designed by any technique whatsoever can be viewed as the combination of a state estimate feedback controller and additional feedback from a dynamic compensator driven by the residual signal (difference between actual and estimated system outputs). Motivated by such an observ...

Full description

Saved in:
Bibliographic Details
Published inIEEE transactions on systems, man, and cybernetics. Systems Vol. 46; no. 4; pp. 548 - 558
Main Authors Yong Zhang, Ying Yang, Ding, Steven X., Linlin Li
Format Journal Article
LanguageEnglish
Published New York IEEE 01.04.2016
The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Subjects
Adaptive control
Algorithm design and analysis
Closed loop systems
Compensators
Control system analysis
Controllers
Convergence
Design engineering
Dynamical systems
Dynamics
feedback systems
Heuristic algorithms
Iterative algorithms
least squares methods
Online
optimal control
Optimization
Performance analysis
reinforcement learning
residual-driven dynamic compensator
Systems stability
feedback systems
optimal control
Control system analysis
residual-driven dynamic compensator
least squares methods
reinforcement learning
Online AccessGet full text

Cover

More Information
Summary:A stabilizing regulator designed by any technique whatsoever can be viewed as the combination of a state estimate feedback controller and additional feedback from a dynamic compensator driven by the residual signal (difference between actual and estimated system outputs). Motivated by such an observation, this paper presents the controller design from the premise that system stability is first guaranteed. Control performance can then be enhanced by the optimal design of a residual-driven dynamic compensator subject to some quadratic performance index. The resulting compensator design methods are carried out using both offline and online iterative algorithms with guaranteed convergence. Moreover, the final iterative realization strategy can be implemented online with observed state variables and input updates in case of unknown system dynamics or parameter changes. Simulation results are presented to illustrate design procedures as well as the feasibility of our proposed scheme.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
content type line 23
ISSN:2168-2216
2168-2232
DOI:10.1109/TSMC.2015.2450203