Reinforcement Swarm Intelligence in the Global Optimization Method via Neuro-Fuzzy Control of the Search Process

摘要

The new modification method of the particle swarm optimization (PSO) is presented. Intensified adaptation properties of this stochastic computer method are based on the hybridization it with the weighted average coordinates method and reinforcement swarm intelligence via the neuro-fuzzy control of the agent particles behavior. The results of computer experiments of the global optimization on the test functions of 2, 50, 100 variables with multiple extremes are presented.