The particle swarm optimization via cultural algorithm with fuzzy knowledge evolution

The Particle Swarm Optimization (PSO) is a population-based evolution, algorithm and the inertia weight plays a key role in PSO. To, this paper. a novel class of adaptive PSOs is proposed by using the Cultural Algorithm. (CA) with Jazzy knowledge evolution. The fuzzy rules of the fuzzy knowledge represent the experiences of the particles, and these rules are shared in the population to form the culture. While the popalation is evolving, the culture which is encoded as cultural gene is evolved by the Genetic Algorithm (GA). From the culture, the fuzzy systems are constructed by the fuzzy rules in, the belief space of the CA and used to approximate the fittest controllers of the inertia weights of the particles in the PSO far a given optimization problem. The simulation. results illustrate that the FCAPSO is More accuracy than both the LPSO and the FPSO, when the dimension of the optimization problems are low, or when the evolution time is long enough for the optimization problems with high dimensions.