A novel Fuzzy Particle Swarm Optimization
Ehsan Aminian, Mohammad Teshnehlab · 2013
In this paper, we introduce a novel Fuzzy particle Swarm Optimization method in which the inertia weight as well as the cognitive and social coefficients are adjusted for each particle separately according to the information coming from a Fuzzy Logic Controller. We illustrate the efficiency of our method in comparison with the origin version of inertia Weight Particle Swarm Optimization. Although the curse of dimensionality has always been one of significant weaknesses in Evolutionary Algorithms, we show it has the least influence on the proposed method compared to PSO and original WPSO. We also prove that our method outperforms other Fuzzy-PSO versions by testing two benchmark functions.