Research on updating algorithms in particle swarm optimization
Zhengqiang Li · Journal of Jiangsu University of Science and Technology · 2008
PSO(Particle Swarm Optimization) is a stochastic optimization algorithm inspired by social behavior of bird flocking or fish schooling.However,the standard PSO has some shortcomings,such as premature convergence,searching precision lowness and so forth.Based on the simulation of natural death process of 5% B-cell in biology clone selection,this paper proposes 8 kinds of updating algorithms according to intergeneration differential,theory of chaos,principle of mutation respectively,and selects the updated particles in terms of simulated annealing method.Numerical experiments show that the updating algorithm by using intergeneration differential and chaotic mutation(algorithm 8) is a good selection.Simultaneously,the updating effect is perfect when the updated particles are about 20%.The algorithm may overcome effectively the premature problem and speed up the convergence.