Design and implementation of a fuzzy supervisor for on-line compensation of nonlinearities: An instability avoidance module

Authors:
M. Sanjuan;A. Kandel;C. A. Smith
Affiliations:
Department of Mechanical Engineering, Universidad del Norte, Km 5 Autop. Pto. Colombia, Barranquilla, Colombia;Computer Science and Engineering Department, University of South Florida, 4202 E. Fowler Av., Tampa, FL 33620, USA;Chemical Engineering Department, University of South Florida, 4202 E. Fowler Av., Tampa, FL 33620, USA
Venue:
Engineering Applications of Artificial Intelligence
Year:
2006

Citing 3
Cited 3

Fuzzy self-tuning of PID controllers

Fuzzy Sets and Systems
Fuzzy controllers as gain scheduling approximators

Fuzzy Sets and Systems - Special issue on methods for data analysis in classificatin and control
Process Control Systems: Application, Design and Tuning

Process Control Systems: Application, Design and Tuning

Iterative performance improvement of fuzzy control systems for three tank systems

Expert Systems with Applications: An International Journal
On accurate localization and uncertain sensors

International Journal of Intelligent Systems
Stable and convergent iterative feedback tuning of fuzzy controllers for discrete-time SISO systems

Expert Systems with Applications: An International Journal

Quantified Score

Hi-index	0.00

Visualization

Abstract

Unstable behavior of a control loop is the most feared situation in control engineering. Since PID controllers are linear in nature, but still used in 90% of control loops in the process industry, their use in nonlinear processes increases the risk of closed-loop unstable response. This paper presents the design of an instability avoidance module for SISO PID control loops to automatically adjust the controller's tuning parameters when undesired oscillatory behavior is monitored. This module is a Fuzzy-based Supervisor composed of a dynamic parameter identification module that operates on-line, and a Fuzzy Inference System (FIS) with fuzzy rules to modify PID tuning. Tests in nonlinear process models are performed, demonstrating that this Fuzzy Supervisor-Type 1 (non-intrusive) provides the PID controller the ability to adapt its tuning to eliminate observed oscillatory behavior.