Aiming at the shortcoming of artificial bee colony algorithms, such as the low convergence rate and easy to be trapped into the local optimums, an improved algorithm is proposed. First, a new crossover strategy is designed to make the group close to the optimal solution as soon as possible. Then, considering that the parameter of controlling the behavior of the scouts to avoid falling into local optimal setting is difficult and of a greater impact on the performance of the algorithm, a mutation strategy based on opposition-based learning is proposed to replace the scouts' behavior. The simulation results on 10 standard test functions show that this new improved algorithm can obtain the global optimal solutions for almost all the functions, with fast convergence and good robustness. The performance of this algorithm is significantly better than the existing artificial bee colony algorithms.%针对人工蜂群算法(artificial bee colony algorithm,ABC)存在的收敛速度慢、易陷入局部最优的缺点,提出了一种改进算法.首先,设计新的选择策略和交叉策略,使群体快速向最优解靠近；然后,鉴于控制侦查蜂行为的参数难于确定,且对算法性能影响较大,提出了基于反向学习的变异策略代替侦查蜂行为,同样达到避免陷入局部最优的效果.通过对10个标准测试函数的仿真表明,改进算法几乎都可以得到各测试函数的全局最优解,而且收敛速度快、鲁棒性好.改进性能明显优于现有人工蜂群算法.
展开▼
