Anti-windup TS Fuzzy PI-like Control for Discrete-Time Nonlinear Systems with Saturated Actuators
Autor: | Kevin Guelton, Luis F. P. Silva, Adriano N. D. Lopes, Valter J. S. Leite |
---|---|
Přispěvatelé: | Centre de Recherche en Sciences et Technologies de l'Information et de la Communication - EA 3804 (CRESTIC), Université de Reims Champagne-Ardenne (URCA), Centro Federal de Educação Tecnológica de Minas Gerais (CEFET-MG) |
Jazyk: | angličtina |
Rok vydání: | 2020 |
Předmět: |
Anti-windup action
Computer science [MATH.MATH-DS]Mathematics [math]/Dynamical Systems [math.DS] Computational intelligence 02 engineering and technology Fuzzy logic Theoretical Computer Science [SPI.AUTO]Engineering Sciences [physics]/Automatic [INFO.INFO-AI]Computer Science [cs]/Artificial Intelligence [cs.AI] Tracking error Artificial Intelligence Control theory [INFO.INFO-AU]Computer Science [cs]/Automatic Control Engineering 0202 electrical engineering electronic engineering information engineering PI-like controller design Actuators saturation Local stabilization Nonlinear system Computational Theory and Mathematics Bounded function TS fuzzy discrete-time systems Convex optimization Piecewise 020201 artificial intelligence & image processing Real-time experiments Actuator Software |
Zdroj: | International Journal of Fuzzy Systems International Journal of Fuzzy Systems, Taiwan Fuzzy Systems Association, 2020, 22 (1), pp.46-61. ⟨10.1007/s40815-019-00781-0⟩ |
ISSN: | 1562-2479 2199-3211 |
DOI: | 10.1007/s40815-019-00781-0⟩ |
Popis: | International audience; This paper deals with piecewise constant set-points tracking control of nonlinear discrete-time systems represented by Takagi-Sugeno models under actuators’ saturation. To this end, a fuzzy Proportional Integral-like (PI-like) discrete-time control scheme is considered, which consists of a proportional state feedback, an integral action over the tracking error, and an anti-windup action. All the control gains are obtained through a convex optimization procedure formulated in term of Linear Matrix Inequalities (LMIs). The proposed method yields a Parameter Distributed Compensation (PDC) PI-like control and a non-PDC anti-windup action structure. Due to the actuators’ saturation, a local approach is considered with a fuzzy Lyapunov function to ensure the local closed-loop stability, to provide an estimate of the region of attraction, and to compute the amplitude bounds of set-points changes. This latter issue allows delivering operational security by providing a bounded range for the set-points variation. To validate and illustrate the performance of the proposed tracking control approach, real-time experiments has been performed on an industrial oriented process consisting on the nonlinear level control of two interactive tanks. |
Databáze: | OpenAIRE |
Externí odkaz: |