The Simulation and Improvement of Particle Swarm Optimization Based on Cellular Automata
Yu Fengxia, Li Gang · Procedia Engineering · 2012
We observed that PSO algorithm was easily falling into local extreme point by using cellular automata simulation, and developed several improvement strategies which drew inspiration from simulation, such as adding additional particle into updated formula, introducing crossover operator and mutation disturbance when the premature convergence occurs. The experiments validated new algorithm on 10 benchmark functions, and results showed that the improved PSO algorithm has better stability and higher accuracy.