【摘要】 提出了一种基于邻域极值数的协同粒子群优化算法。该算法将种群分为若干个独立进化的子种群。根据邻域极值数确定各子种群的生存状态。根据子种群的生存状态对子种群实施相应的控制操作,提高子种群的搜索能力,实现子种群之间的信息共享,共同进化。测试结果表明基于邻域极值数的协同粒子群优化算法是一种高效稳健的全局优化算法。更多还原

【Abstract】 A cooperative particle swarm optimization based on the neighborhood extremum number is proposed. In this algorithm, the whole population is divided into several sub-populations evolving independently. The survival state of each sub-population is determined in terms of the neighborhood extremum number. Based on the survival state of each sub-population, corresponding control operation is implemented so as to improve the search ability of each sub-population and realize information sharing so that the sub-populations coevolve. The experimental results show that the cooperative particle swarm optimization based on the neighborhood extremum number is an effective and steady global optimization algorithm.更多还原