The artificial bee colony （ABC） algorithm is a com- petitive stochastic population-based optimization algorithm. How- ever, the ABC algorithm does not use the social information and lacks the knowledge of the proble...The artificial bee colony （ABC） algorithm is a com- petitive stochastic population-based optimization algorithm. How- ever, the ABC algorithm does not use the social information and lacks the knowledge of the problem structure, which leads to in- sufficiency in both convergent speed and searching precision. Archimedean copula estimation of distribution algorithm （ACEDA） is a relatively simple, time-economic and multivariate correlated EDA. This paper proposes a novel hybrid algorithm based on the ABC algorithm and ACEDA called Archimedean copula estima- tion of distribution based on the artificial bee colony （ACABC） algorithm. The hybrid algorithm utilizes ACEDA to estimate the distribution model and then uses the information to help artificial bees to search more efficiently in the search space. Six bench- mark functions are introduced to assess the performance of the ACABC algorithm on numerical function optimization. Experimen- tal results show that the ACABC algorithm converges much faster with greater precision compared with the ABC algorithm, ACEDA and the global best （gbest）-guided ABC （GABC） algorithm in most of the experiments.展开更多
针对人工蜂群(ABC)算法开发能力弱的缺点,提出一种基于适应度分割机制和自适应搜索策略的ABC算法(FSABC)。首先,在雇佣蜂和跟随蜂阶段开始前,根据适应度值将种群划分为高适应度子种群和低适应度子种群,并通过动态调整子种群大小,更好地...针对人工蜂群(ABC)算法开发能力弱的缺点,提出一种基于适应度分割机制和自适应搜索策略的ABC算法(FSABC)。首先,在雇佣蜂和跟随蜂阶段开始前,根据适应度值将种群划分为高适应度子种群和低适应度子种群,并通过动态调整子种群大小,更好地平衡算法的开发性和探索性,并更合理地分配搜索资源;其次,对跟随蜂中的高适应度子种群提出一个策略池和一种新的自适应搜索方式,以避免算法陷入局部最优解;再次,为了加强算法的开发能力,根据高适应度子种群的特点,设计一个新的搜索策略和一个策略池,以发挥该子种群的优势,从而提高算法的性能;最后,对于复杂的多峰问题,在适应度景观中存在许多局部最优解,其中一些可能接近全局最优解,因此,搜索一个好的解的邻域将有助于找到更好的解,甚至可能找到全局最优解,鉴于此,使用一个邻域搜索算子加强算法的开发能力。基于22个经典测试函数进行比较实验的结果表明,在30维和50维问题上,与ABCLGII(ABC algorithm with Local and Global Information Interaction)相比,所提算法的Friedman检验的秩次等级分别提高了30.8%和11.7%,可见,所提算法的性能求解精度更优,并能有效处理全局数值优化问题。展开更多
基金supported by the National Natural Science Foundation of China(61201370)the Special Funding Project for Independent Innovation Achievement Transform of Shandong Province(2012CX30202)the Natural Science Foundation of Shandong Province(ZR2014FM039)
文摘The artificial bee colony （ABC） algorithm is a com- petitive stochastic population-based optimization algorithm. How- ever, the ABC algorithm does not use the social information and lacks the knowledge of the problem structure, which leads to in- sufficiency in both convergent speed and searching precision. Archimedean copula estimation of distribution algorithm （ACEDA） is a relatively simple, time-economic and multivariate correlated EDA. This paper proposes a novel hybrid algorithm based on the ABC algorithm and ACEDA called Archimedean copula estima- tion of distribution based on the artificial bee colony （ACABC） algorithm. The hybrid algorithm utilizes ACEDA to estimate the distribution model and then uses the information to help artificial bees to search more efficiently in the search space. Six bench- mark functions are introduced to assess the performance of the ACABC algorithm on numerical function optimization. Experimen- tal results show that the ACABC algorithm converges much faster with greater precision compared with the ABC algorithm, ACEDA and the global best （gbest）-guided ABC （GABC） algorithm in most of the experiments.
文摘针对人工蜂群(ABC)算法开发能力弱的缺点,提出一种基于适应度分割机制和自适应搜索策略的ABC算法(FSABC)。首先,在雇佣蜂和跟随蜂阶段开始前,根据适应度值将种群划分为高适应度子种群和低适应度子种群,并通过动态调整子种群大小,更好地平衡算法的开发性和探索性,并更合理地分配搜索资源;其次,对跟随蜂中的高适应度子种群提出一个策略池和一种新的自适应搜索方式,以避免算法陷入局部最优解;再次,为了加强算法的开发能力,根据高适应度子种群的特点,设计一个新的搜索策略和一个策略池,以发挥该子种群的优势,从而提高算法的性能;最后,对于复杂的多峰问题,在适应度景观中存在许多局部最优解,其中一些可能接近全局最优解,因此,搜索一个好的解的邻域将有助于找到更好的解,甚至可能找到全局最优解,鉴于此,使用一个邻域搜索算子加强算法的开发能力。基于22个经典测试函数进行比较实验的结果表明,在30维和50维问题上,与ABCLGII(ABC algorithm with Local and Global Information Interaction)相比,所提算法的Friedman检验的秩次等级分别提高了30.8%和11.7%,可见,所提算法的性能求解精度更优,并能有效处理全局数值优化问题。
