Parameter identification using memetic algorithms for fuzzy systems

Zhu, N., Ong, Y.S., Wong, K.W. and Seow, K.T. (2003) Parameter identification using memetic algorithms for fuzzy systems. In: Proceedings of the Fourth International Conference on Intelligent Technologies (InTech' 03), 17 - 19 December, Chang Mai, Thailand pp. 833-839.

Abstract

In recent years, fuzzy modelling has become very popular because of its ability to assign meaningful linguistic labels to fuzzy sets in the rule base. However, in order to achieve better performance in fuzzy modelling, parameter identification often needs to be performed. In this paper, we address this optimization problem using memetic algorithms (MAs) for Sugeno and Yasukawa's (SY) qualitative (fuzzy) model. MAs are essentially variants of Genetic Algorithms incorporated with local search methods (or memes) that could better improve the search control accuracy. In addressing the parameter identification problem, MAs are utilized to perform search exploitation within the neighbourhood of the prior knowledge extracted via the Improved SY fuzzy modelling approach. The use of MAs in performing parameter identification is examined empirically, and found to produce better solutions attributable to the extraction and proper use of prior knowledge.

Item Type:	Conference Paper
Publisher:	Chiang Mai University. Institute for Science and Technology Research and Development
URI:	http://researchrepository.murdoch.edu.au/id/eprint/22803

Item Control Page

Downloads