Abstract: Artificial bee colony algorithm is easy to fall into local optimum and to premature convergence. In order to solve these problems, an Artificial Bee Colony algorithm based on the Law of the jungle and Feedback（LFABC） is proposed in this paper. The algorithm introduces feedback mechanism, so the algorithm can directly search area where there is possibly an optimal solution, improve the development ability and convergence speed. In order to balance the development capacity and the exploration capacity in the different stages, LFABC introduces linear increasing differential strategy into the global search equation. In the nature, law of the jungle is the only criterion for survival. It simulates this natural phenomena, LFABC algorithm randomly selects an individual to initialize, effectively prevent the algorithm into local optimum. Experimental results show that LFABC effectively improve the convergence precision and its convergence speed is outstanding.

Key words: feedback mechanism, law of jungle, self-adaptive, linear increasing differential strategy

摘要： 为了解决人工蜂群算法（ABC）容易陷入局部最优、易早熟收敛等问题，提出一种基于反馈机制和丛林法则的人工蜂群算法（Artificial Bee Colony algorithm based on Feedback and the Law of the jungle，LFABC）。该算法在全局搜索公式中引入反馈机制，直接搜索最优解可能存在的区域，提高了算法的开发能力和收敛速度。同时加入线性微分递增策略，平衡算法各个阶段的开发能力和探索能力。根据丛林法则，该算法随机选择较差个体进行初始化，有效防止算法陷入局部最优。实验结果证明，LFABC算法有效提高了算法的收敛精度，且其收敛速度非常突出。

关键词: 反馈机制, 丛林法则, 自适应, 线性微分递增策略