Simple Modification of Takagi-Sugeno Model
The Takagi-Sugeno model is very popular in fuzzy logic area. It can approximate different functions and physical processes. However, it is known that the model has one inconvenience. If simple triangular or trapezoidal membership functions are used for fuzzy sets then in intermediate area, described example by two fuzzy sets (conclusions of two rules) and two approximate functions used in these rules, the model can not approximate the process well. In the paper simple modification of the model is presented as one giving very good result.
KeywordsMembership Function Fuzzy System Approximate Function Intermediate Area Universal Approximators
Unable to display preview. Download preview PDF.
- 1.Babuska R., (2001) Construction of fuzzy systems - interplay between precision and transparency, European Symposium on Intelligent Technics (ESIT), Aachen, Germany, 445 - 452Google Scholar
- 2.Butkiewicz B. S., (2001) Inference in Fuzzy Models of Physical Processes, in Computational Intelligence Theory and Applications, Lecture Notes in Computer Science 2206, B. Reusch (ed.), Springer, Berlin, New York, 782 - 790Google Scholar
- 6.Takagi T., Sugeno M., (1983) Derivation of fuzzy control rules form human operator’s control system, Proc. of IFAC Symp. on Fuzzy Information, Knowledge Representation and Decision Analysis, Marseilles, France, 55 - 60Google Scholar
- 8.Wang X. L., (1992) Fuzzy systems are universal approximators, Proc IEEE Int. Conf. on Fuzzy systems, San Diego (CA), 1163 - 1169Google Scholar
- 9.Wang X. L., Mendel J. M., (1992) Fuzzy basis functions, universal approximation, and orthogonal least-square learning, IEEE Trans. on Neural Network, 3, No. 5, 807 - 814Google Scholar