Dissimilarity Clustering Algorithm for Designing the PID-like Fuzzy Controllers

  • Essam Natsheh Dep. of Informatics Engineering, College of Engineering, AMA International University, Bahrain
Keywords: Three-term controller, PID controller, fuzzy systems, fuzzy logic controller, clustering algorithms

Abstract

Fuzzy logic controller is one of the most prominent research fields to improve efficiency for process industries, which usually stick to the conventional proportional-integral-derivative (PID) control. The paper proposes an improved version of the three-term PID-like fuzzy logic controller by removing the necessity of having user-defined parameters in place for the algorithm to work. The resulting non-parametric three-term dissimilarity-based clustering fuzzy logic controller algorithm was shown to be very efficient and fast. The performance study was conducted by simulation on armature-controlled and field-controller DC motors, for linguistic type and Takagi-Sugeno-Kang (TSK) models. Comparison of the created algorithm with fuzzy c-means algorithm resulted in improved accuracy, increased speed and enhanced robustness, with an especially high increase for the TSK type model.

Published
2021-06-15
How to Cite
[1]
E. Natsheh, ‚ÄúDissimilarity Clustering Algorithm for Designing the PID-like Fuzzy Controllers‚ÄĚ, JIOS, vol. 45, no. 1, Jun. 2021.
Section
Articles