摘要
灰狼优化算法是一种模拟灰狼捕食行为的群智能优化算法。基于灰狼捕食行为的包围、追捕、攻击三个阶段提出了一种小生境灰狼优化算法(Niche Grey Wolf Optimization, GWO)。该算法利用基本GWO计算各灰狼的适应度值,以小生境半径作为限制,比较灰狼个体的适应度值,通过对适应度值较差的灰狼个体施以罚函数,来提高全局搜索能力。分析了NGWO算法的时间复杂度,利用NGWO对5个基准函数进行了测试,并与基本灰狼算法和粒子群算法的结果进行了比较,表明NGWO算法无论是在收敛速度还是求解精度上均有明显改善。
Grey wolf optimization algorithm is a new swarm intelligence algorithm with simulation of grey wolf predation behavior. Based on the encircling, hunting and attacking of grey wolves prey be-havior in nature, a niche grey wolf optimization (NGWO) algorithm is proposed. In this algorithm, calculating the fitness of individual using basic GWO, to poor fitness with penalty function through compares fitness of individual with niche radius as limit, thus improve the global search ability. From the above, according to algorithmic flows analyze the time complexity of NGWO. Moreover, tested the NGWO by 5 benchmark functions and compared with two intelligent algorithms, grey wolf optimization and particle swarm optimization. Simulation results indicate that the algorithm has significant improvement in aspects of convergence speed and accuracy.
出处
《计算机科学与应用》
2017年第6期562-571,共10页
Computer Science and Application