求解函数优化问题的改进的人工蜂群算法

摘要/Abstract

摘要： 为提高人工蜂群算法求解复杂函数优化问题的性能,分析了算法中侦察蜂逃逸行为的不足,并对其进行改进:定义了逃逸指标,使其能准确地反映个体状态对算法早熟的影响；重新设计选择机制,让侦察蜂不需要参数控制,能自适应地选择可能导致算法早熟收敛的个体执行逃逸操作；改进了逃逸算子,降低了逃逸操作的盲目性。通过9个典型测试问题的实验结果表明:在指定误差精度下,本改进算法均能有效收敛；同时与基本人工蜂群算法和已有的典型改进相比,本改进算法在收敛精度和速度上均有明显提高。说明提出的改进策略能有效提高算法求解复杂函数优化问题的能力。

关键词: 人工蜂群算法,早熟收敛,逃逸指标,选择机制,逃逸算子

Abstract: In order to enhance the performance of artificial bee colony algorithm in solving complex function optimization problems,this paper analysed the shortcoming of escape behavior of scout bees,and improved it．The improved algorithm defines escape index,making it precisely reflecting the effect of individual status on the premature convergence of algorithm,redesigns the selection scheme,making scout bees choosing individual escape operation that might result in algorithm premature convergence adaptively,improves the escape operator,reducing the blindness of escape operation．Nine typical experiments prove that the improved algorithm could converge efficiently under assignment convergence accuracy,and the improved algorithm could converge with more convergence accuracy and speed compared with basic artificial colony algorithm and existing typical improved versions,thus proves the improved strategy proposed in this paper could boost capability of solving complex function optimization problems．

Key words: Artificial bee colony algorithm,Premature convergence,Escape index,Selection scheme,Escape operator

葛宇,梁静,王学平,谢小川. 求解函数优化问题的改进的人工蜂群算法[J]. 计算机科学, 2013, 40(8): 252-257. https://doi.org/

GE Yu,LIANG Jing,WANG Xue-ping and XIE Xiao-chuan. Improved Artificial Bee Colony Algorithms for Function Optimization[J]. Computer Science, 2013, 40(8): 252-257. https://doi.org/

参考文献

[2] Karaboga D,Basturk B．On the performance of artificial bee co-lony(ABC)algorithm[J].Applied Soft Computing,2008,8(1):687-697
[3] Karaboga D,Akay B．A comparative study of artificial bee colony algorithm[J]．Applied Mathematics and Computation,2009,214(1):108-132
[4] ,仅有Sphere和Schwefel’s 2.22函数的优化结果优于SABC,但SABC的求解复杂度明显大于文献[4](SABC为100维,文献[4]为50维),并且SABC对以上两函数的求解结果也达到了10－100以上精度,说明SABC与文献[4]的改进策略相比同样占有一定优势。表4 SABC与参考文献的对比结果函数算法 SABC文献[4]文献[8]文献[9]文献[10] Sphere7.7E－24201.3E－333.0E－16－ Quartic4.1E－46.5E－22.5E－1－－ Step01.2E－15－－－ Schwefel’s 2.213.8E－9619.67－－－ Schwefel’s 2.222.5E－1050－－－ SumSquares5.7E－247－－－3.3E－21 Griewank007.2E－172.7E－160 Rastrigin001.4E－1606.8E－14 Ackley8.8E－168.8E－163.4E－132.9E－145.3E－13结束语为提高人工蜂群算法求解复杂函数优化问题的性能,本文对其进行改进,分析了原算法侦察蜂逃逸行为的不足,并提出了改进方案。具体地,设计了逃逸指标、自适应选择机制,改进了逃逸算子。数值实验结果表明:在函数优化问题求解中,本文改进方案能有效帮助算法逃离早熟收敛,使算法的收敛速度和求解精度得到提高,达到了预期效果。 Karaboga D．An idea based on honey bee swarm for numerical optimization[R]．Kayseri:Erciyes University,2005[2]Karaboga D,Basturk B．On the performance of artificial bee co-lony(ABC)algorithm[J].Applied Soft Computing,2008,8(1):687-697[3]Karaboga D,Akay B．A comparative study of artificial bee colony algorithm[J]．Applied Mathematics and Computation,2009,214(1):108-132[4]Li C Q,Niu P F,Xiao X J．Development and investigation of efcient articial bee colony algorithm for function optimization numerical function optimization[J]．Applied Soft Computing,2012,12:320-332
[5] 罗钧,王强,付丽.改进蜂群算法在平面误差评定中的应用[J].光学精密工程,2012,0(2):422-430
[6] Horng M H.Multilevel thresholding selection based on the artificial bee colony algorithm for image segmentation[J]．Expert Syst Appl,2011,38(11):13785-13791
[7] ztürk C,Karaboˇga D,Grkemli B．Artificial bee colony algorithm for dynamic deployment of wireless sensor networks[J]．Turk J Electr Eng Comput Sci,2012,0(2):1-8
[8] 文献
[9] 文献
[10] Sphere7.7E－24201.3E－333.0E－16－ Quartic4.1E－46.5E－22.5E－1－－ Step01.2E－15－－－ Schwefel’s 2.213.8E－9619.67－－－ Schwefel’s 2.222.5E－1050－－－ SumSquares5.7E－247－－－3.3E－21 Griewank007.2E－172.7E－160 Rastrigin001.4E－1606.8E－14 Ackley8.8E－168.8E－163.4E－132.9E－145.3E－13结束语为提高人工蜂群算法求解复杂函数优化问题的性能,本文对其进行改进,分析了原算法侦察蜂逃逸行为的不足,并提出了改进方案。具体地,设计了逃逸指标、自适应选择机制,改进了逃逸算子。数值实验结果表明:在函数优化问题求解中,本文改进方案能有效帮助算法逃离早熟收敛,使算法的收敛速度和求解精度得到提高,达到了预期效果。 Karaboga D．An idea based on honey bee swarm for numerical optimization[R]．Kayseri:Erciyes University,2005[2]Karaboga D,Basturk B．On the performance of artificial bee co-lony(ABC)algorithm[J].Applied Soft Computing,2008,8(1):687-697[3]Karaboga D,Akay B．A comparative study of artificial bee colony algorithm[J]．Applied Mathematics and Computation,2009,214(1):108-132[4]Li C Q,Niu P F,Xiao X J．Development and investigation of efcient articial bee colony algorithm for function optimization numerical function optimization[J]．Applied Soft Computing,2012,12:320-332[5]罗钧,王强,付丽.改进蜂群算法在平面误差评定中的应用[J].光学精密工程,2012,0(2):422-430[6]Horng M H.Multilevel thresholding selection based on the artificial bee colony algorithm for image segmentation[J]．Expert Syst Appl,2011,38(11):13785-13791[7]ztürk C,Karaboˇga D,Grkemli B．Artificial bee colony algorithm for dynamic deployment of wireless sensor networks[J]．Turk J Electr Eng Comput Sci,2012,0(2):1-8[8]暴励,曾建潮.一种双种群差分蜂群算法[J].控制理论与应用,2011,8(2):266-272[9]Wu B,Fan S H．Improved artificial bee colony algorithm with chaos[M]．Computer science for environmental engineering and ecoinformatics,Berlin:Springer,2011:51-56[10]罗钧,肖向海,付丽,等.基于分段搜索策略的改进蜂群算法[J].控制与决策,2012,7(9):1402-1405
[11] 张超群,郑建国,王翔.蜂群算法研究综述[J]．计算机应用研究,2011,28(9):3201-3205
[12] Chen S H,Chen M C,Chang P C,et al．Guidelines for developing effective estimation of distribution algorithms in solving single machine scheduling problems[J]．Expert Systems with Applications,2010,37(9):6441-6451

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed