J. Martikainen and S.J. Ovaska (Finland)
Evolution strategy, multipopulation, adaptive evolution ary algorithm, dynamic fuzzy system
This paper introduces an evolutionary optimization algo rithm taking advantage of multiple populations and an adaptive aging parameter to achieve faster and more ro bust convergence. As challenging test cases, the evolu tionary algorithm is used to optimize parameters for dy namical fuzzy systems. Our results show that the pro posed algorithm is capable of outperforming the tradi tional reference algorithm. The effect of sampling the membership functions of the dynamical fuzzy system in feed-forward and feedback configurations is also studied.
Important Links:
Go Back
