The Particle Swarm Algorithm Based On Group Decision-making

Particle swarm optimization (PSO) is a population-based swarm intelligent algorithm by simulating bird flocking and fish schooling. Therefore in decision view, the equation of Particle swarm optimization could be considered as the process of individual decision, so in this paper, we use the idea of group decision making to improve their performance through several waysIn this algorithm, the particle uses the best individual historical position and the swarm historical position to make decision for next expectation position, but population diversity drop quickly, the algorithm is easy to fall into premature convergence. So we improve algorithm through the idea of group decision making, it use the best individual historical position for making decision to receive a new position and replace the best swarm historical position. So in early stage, the algorithm has large population diversity, which cause them avoid premature convergence.The improvement in last chapter merely use the best individual historical position, but it rarely uses individual current position, so in this paper we further discuss the way of a new group decision-making based on current position, which take the individual position liner weighted, then a new deciding position takes as a disturbance position for influencing the way of individual direction. Particle swarm optimization (PSO) is a stochastic optimization algorithm imitating animal behavior which was proposed by them, a large number of academics quoted them, because of its popular characters, such as the high speed of searching and the simple structure, which is applied to every aspect of society, because the algorithm fall into premature convergence, so we use any way to improve the performance from any aspect. This paper we use the idea of group decision making to improve their performance.In the end, this paper explains the algorithm from the growth view of plant, every particle takes as a plant, In order to grow better, they will move to the best individual to get more light, and other resources. This paper will introduce a single monocular competition index to design the weight of individual, ipropose the group decision-making algorithm based on single monoculacompetition index model, Simulation results show that it is effective.