一种协调勘探和开采能力的粒子群算法

提出一种新的协调勘探和开采能力的粒子群优化算法. 该算法将种群分为随机子群和进化子群, 随机子群增加了算法全局解空间的勘探能力, 在运行过程中通过随机子群进化信息生成解优胜区域指导进化粒子向着最优解子空间逼近. 为了提高算法收敛速度, 算法只在进化子群进入收敛阶段时才对其进行指导, 以防止增加种群多样性导致算法开采能力下降的问题. 将此算法与其他改进粒子群算法进行比较, 实验结果表明, 该算法有较好的全局收敛性, 不仅能有效地克服其他算法易陷入局部极小值的缺点, 而且算法收敛速度和稳定性都有显著提高.

Particle-swarm algorithm coordinating the exploration and the exploitation

A novel particle-swarm optimization(PSO) algorithm which coordinates the exploration ability and the exploitation ability(EEPSO) is presented. This algorithm divides the population of the swarm into the evolutionary sub-swarm and the randomized sub-swarm. During the evolution, the randomized sub-swarm reinforces the global space-exploration ability of the PSO algorithm, and uses the multi-species evolution information to generate the best-result-value space, guiding the particles of the evolutionary sub-swarm to approach this space. In order to improve the convergence rate, the guidance will be effective only when the evolutionary sub-swarm particles are in the convergence status. This limits the diversity of the population swarm, preventing the reduction in exploitation ability. The comparison experiments have been made between the proposed approach with the dissipative PSO and other cooperative particle swarm algorithm. The experimental results show that the proposed method not only effectively solves the premature convergence problem, but also significantly speeds up the convergence and improves the stability.