具有自适应步长与协同寻优的蝙蝠烟花混合算法

摘要

论文提出了一种新的群体智能优化算法——蝙蝠烟花混合算法.该算法采用蝙蝠算法在全局最优附近的位置信息、蝙蝠发出的频率、全局最优以及烟花的位置信息构造了新的爆炸半径,使烟花算法在寻优的过程中能够自动地调整步长;并且使蝙蝠个体与烟花个体实现协同寻优;最后,采用"精英-随机"策略选择下一代烟花,增加了烟花种群的多样性.与其他算法(如蝙蝠算法、标准粒子群算法、烟花算法等)相比,函数优化问题以及0-1背包问题的对比实验结果表明:论文提出的混合算法的整体性能优于其他五种算法.%This paper proposes a new swarm intelligent optimization algorithm called hybrid Bat and Fireworks algorithm. This hybrid algorithm adopts the position information around the global optimum,the frequency of the bat particle,the global best particle and the fireworks particle′s location information to construct a new explosion radius,so that the fireworks algorithm can adjust the step size au-tomatically during the optimization process,and make the individual bat and the individual fireworks achieve cooperatrive optimiza-tion. Finally,adopting the Elite-Random strategy to select the next generation of fireworks,which can increase diversity of fireworks populations. Comparing with other algorithms,such as bat algorithm,standard particle swarm algorithm,fireworks algorithm and so on,the experimental results of function optimization problems and 0-1 knapsack problems show that the hybrid algorithm proposed in this paper outperforms the other five algorithms.