摘要
针对传统人工蜂群算法存在收敛速度慢和易陷入局部最优的问题,提出一种基于云模型的改进人工蜂群算法。通过正态云算子计算候选位置,自适应调整算法的局部搜索范围,以提高算法的收敛速度和勘探能力。为保持种群多样性,引入一个新的概率选择策略,使较差的个体具有较大的选择概率,并且利用历史最优解探索新的位置。标准复合函数测试表明,改进算法的收敛速度和求解精度得到提升,优于一些新近提出的改进人工蜂群算法。
Traditional Artificial Bee Colony (ABC) algorithms suffer from the problem of slow convergence and easy stagnation in local optima. An improved ABC algorithm based on cloud model, was proposed to solve the problem. By calculating a candidate food source through the normal cloud particle operator and by reducing the radius of the local search space, the proposed algorithm can enhance the convergence speed and exploitation capability. In order to maintain diversity, a new selection strategy that makes the inferior individual have more chances to be selected was introduced. In addition, the best solution found over time was used to explore a new position in the algorithm. A number of experiments on composition functions show that the proposed algorithm has been improved in terms of convergence speed and solution quality, and is better than some recently proposed improved ABC algorithms.
出处
《计算机应用》
CSCD
北大核心
2012年第9期2538-2541,共4页
journal of Computer Applications
基金
福建省自然科学基金资助项目(2010J01329)
福建省高校产学合作科技重大项目(2010H6012)
关键词
云模型
人工蜂群算法
全局优化
群体智能
早熟收敛
cloud model
Artificial Bee Colony (ABC) algorithm
global optimization
swarm intelligence
premature convergence