加强局部搜索能力的人工蜂群算法

针对人工蜂群算法在求解过程中存在收敛速度慢、易陷入局部最优解等缺点,提出了基于加强局部搜索策略的人工蜂群算法(ABC Based On Enhancing Local Search Ability, LSABC).一方面,在雇佣蜂搜索阶段,利用两种不同的搜索公式得到两组解,并将适应度最佳者作为候选解,增加解的多样性;同时,在搜索公式中加入个体的双重认知能力平衡算法的勘探和开发能力.另一方面,在侦察蜂搜索阶段,采用禁忌搜索策略,将局部极值存入禁忌表中,帮助算法跳脱局部最优解,达到避免算法早熟的同时加快算法收敛速度的目的.由于LSABC算法的改进与粒子群算法相似,为验证LSABC算法的寻优性能,针对8个经典基准函数,选取标准ABC算法、PSO算法、EABC算法、RLPSO算法及LSABC算法分别进行对比测试.计算实验结果表明,LSABC算法在求解精度和收敛速度方面明显提高,易于跳脱局部最优解.

An Improved artificial bee colony algorithm for enhancing local search ability

Aiming at the shortcomings of artificial bee colony algorithm in solving process, such as slow convergence rate and easy to fall into local optimal solution, this paper proposes an artificial bee colony algorithm based on enhanced local search ability(LSABC). On the one hand, in the employed bees search stage, two different search formulas are used to obtain two sets of solutions, and the best fitness value is used as a candidate solution to increase the diversity of the solutions. At the same time, the search formula is added with individual self-cognition ability(exploration ability)and social cognition ability(exploitation ability)to balance the exploration and exploitation ability of the algorithm, so as to speed up the convergence of the artificial bee colony algorithm. On the other hand, in the scout bees search stage, the Tabu search strategy is adopted and the local optimal solution is stored in the taboo table to help the algorithm to jump off the local optimal solution, so as to avoid the premature algorithm and accelerate the convergence speed of the algorithm. Because the improvement of LSABC algorithm is similar to particle swarm optimization algorithm, in order to verify the optimization performance of LSABC algorithm, the standard ABC algorithm, PSO algorithm, EABC algorithm, RLPSO algorithm and LSABC algorithm are selected for comparative test for eight classic benchmark functions. The experimental results show that LSABC algorithm can improve the accuracy and convergence speed, and is easy to jump out of the local optimal solution.