摘要
标准的粒子群优化算法作为一种随机全局搜索算法,因其在种群中传播速度过快,易陷入局部最优解。基于KRTG的动态拓扑结构的粒子群算法(KRTG-PSO),从粒子间的拓扑结构出发,动态地调整种群的拓扑结构,增加种群的多样性,使算法收敛于全局最优解。通过测试函数以及与其他算法的比较,并通过实验表明,该算法在收敛速度与数据精度上收到了满意的效果。
The standard particle swarm optimization algorithm as a random global search algorithm, because of its rapid propagation in populations, easily into the local optimal solution. Based on the dynamic topology structure of KRTG particle swarm optimization （PSO）, KRTG- between particles from the topological structures, dynamically adjust the topological structure of population, can increase the diversity of population, the method converge to the global optimal solution. Through the test function and the comparison with other algorithm, experimental results show that the algorithm convergence picked up and the effect is satisfied.
出处
《计算机与数字工程》
2010年第2期25-27,81,共4页
Computer & Digital Engineering
关键词
动态
拓扑结构
粒子群
KTPG
适应度
dynamic, topology structure, particle swarm, KRTG, fitness
