A simple method for identification of singleton fuzzy models

Authors:
C.-L. Chen;S.-H. Hsu;C.-T. Hsieh;T.-C. Wang
Affiliations:
Department of Chemical Engineering, National Taiwan University, Taipei, Taiwan, Republic of China;Department of Chemical Engineering, National Taiwan University, Taipei, Taiwan, Republic of China;Department of Chemical Engineering, National Taiwan University, Taipei, Taiwan, Republic of China;Center for Environmental, Safety and Health, Industrial Technology Research Institute, Chutung, Hsinchu, Taiwan, Republic of China
Venue:
International Journal of Systems Science
Year:
2005

Citing 4
Cited 1

Fuzzy modeling and control of multilayer incinerator

Fuzzy Sets and Systems - Special issue: Dedicated to the memory of Richard E. Bellman
A nonlinear fuzzy controller with linear control rules is the sum of a global two-dimensional multilevel relay and a local nonlinear proportional-integral controller

Automatica (Journal of IFAC)
Essentials of fuzzy modeling and control

Essentials of fuzzy modeling and control
Theoretical analysis of crisp-type fuzzy logic controllers using various t-norm sum-gravity inference methods

IEEE Transactions on Fuzzy Systems

Identification of transparent, compact, accurate and reliable linguistic fuzzy models

Information Sciences: an International Journal

Quantified Score

Hi-index	0.00

Visualization

Abstract

This article presents a simple method for constructing a singleton fuzzy model from a given set of input/output data. The method consists of three computational steps: the initial phase, the growth phase, and the optional refining phase. The universe of discourse and two linguistic terms for each input variable and a rule base are established during the initial phase. Additional linguistic terms and rules are then appended sequentially during the growth phase to modify the model structure and to elevate the performance. During the optional refining phase the overall modelling performance can be further improved by adjusting the singleton outputs of the rule set in the sense of least squares. The proposed identification method can simultaneously provide an appropriate model structure and parameters without any time-consuming optimisation. Several numerical examples demonstrate the effectiveness of the proposed identification method.