Identification of fuzzy systems by means of space search evolutionary algorithm (SSEA) and Information granulation
Wei Huang, 오성권 · 정보 및 제어 심포지엄 논문집 · 2009
In this paper, we introduce a hybrid optimization of fuzzy inference systems based on space search evolutionary algorithm (SSEA) and information granulation (IG). SSEA is exploited here to carry out the parameter estimation of the fuzzy models as well as to realize structure optimization which is described as a function optimization problem with inequality constraints by using a new chromosome for structure identification. Compared with the chromosome commonly used in fuzzy modeling, the new chromosome records the same information with a simpler structure. The whole hybrid optimization mechanisms Structural optimization and parametric optimization. The structural optimization is developed by SSEA and HCM while the parametric optimization is realized via SSEA and a standard least square method. As two representative numerical examples, gas furnace and Mackey-Glass time series are considered to evaluate the performance of the proposed model. Experimental results show that the proposed model leads to superior performance in comparison with some other fuzzy models reported in the literature.