具有Lévy飞行特征的蝙蝠算法  

Bat algorithm with the characteristics of Lévy flights

作　　者：刘长平[1,2] 叶春明[1] 

机构地区：[1]上海理工大学管理学院,上海200093 [2]淮阴工学院经济管理学院,江苏淮安223001

出　　处：《智能系统学报》2013年第3期240-246,共7页CAAI Transactions on Intelligent Systems

基　　金：国家自然科学基金资助项目(71271138);教育部人文社会科学规划基金资助项目(10YJA630187);上海市教委科研创新重点资助项目(12ZS133);教育部高校博士学科点专项科研基金资助项目(20093120110008)

摘　　要：针对基本蝙蝠算法易早熟、收敛精度低等不足,在分析蝙蝠算法优化机理和局限性的基础上,从算法仿生原理入手,采用Lévy飞行搜索策略更为真实地模拟蝙蝠的捕食行为,取代原有算法的速度和位置更新方式,充分利用Lévy飞行会产生较大跳跃这种不均匀随机游走的特性,有效避免局部极值的吸引.通过标准测试函数对所提算法进行仿真测试,结果表明所提算法有效克服了原算法易早熟、收敛精度低等缺陷,在寻优精度和全局收敛性能方面明显优于基本蝙蝠算法和粒子群优化算法,是解决复杂函数优化问题的一种有效工具.The basic bat algorithm (BA) in the past research studies reveal deficiencies as apt to be premature and low precision of convergence .This paper first analyzed the optimization mechanism and deficiency of bat algorithm (BA), and then considering the Lévy flight behaviors of bats can simulate predatory more realistically , the study proposed substituting for the speed and location updating pattern of former algorithm .The proposed algorithm fully explored the trait of uneven random walks , so that clusters of short steps were connected by rare long steps , to a-void being trapped in local optimal solution .Simulation results for benchmark functions show that the proposed algo-rithm improved the global optimization ability remarkably and outperformed the basic BA and particle swarm optimi -zation (PSO) in accuracy and convergence property .Therefore, the proposed algorithm is an effective tool for sol-ving the optimization of complex functions.

关 键 词：蝙蝠算法 Lévy飞行 函数优化 粒子群优化算法 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构] N945[自动化与计算机技术—计算机科学与技术]