Identification of Fuzzy Systems Using a Continuous Ant Colony Algorithm
I. A. Khodashinskii, P. A. Dudin
Tomsk State University of Control Systems and Radio Electronics
Keywords: identification, fuzzy systems, continuous ant colony algorithm
Pages: 63-71
Abstract
A continuous ant colony algorithm is first proposed to identify fuzzy systems. It is used to construct five fuzzy systems to approximate nonlinear functions of one, two, and three variables. The effect of algorithm parameters on the approximation error is studied. A comparative analysis with other identification algorithms showed an advantage of the proposed algorithm.
|
