基于动态整体更新和试探机制的蜂群算法

Artificial bee colony algorithm based on dynamic wholly updating and tentative mechanism

蜂群算法已被证明其效率高于多数传统优化算法,但是对于不可分离变量的函数则优势不明显。为平衡单维更新与整体更新,避免算法在某一方面开采过深陷入局部最优,通过计算单维开采成功率动态地控制参数limit,提出了一种单维更新和整体更新交替进行的混合算法。该算法在整体更新阶段采用基于试探机制的粒子群算法,避免种群飞向错误的方向。采用多种不同类型的基准函数对改进算法进行测试,数值实验结果验证了该算法的有效性。

Artificial bee colony（ABC）algorithm has been proven to be a better heuristic algorithm compared with other evolution algorithms.however,ABC has little advantage when used to optimize nonseparable functions.In order to balancing single dimesnsion updating（SDU）stage and wholly updating（WU） stage for avoiding local optimization,proposed a hybrid algorithm named artificial bee colony algorithm particle swarm optimization（ABCPSO） to implement the two stages by turns and balance the exploitation depths of the two stages by controlling the parameter ＇limit＇ dynamically in line with the success rate of SDU stage.In the WU stage,adopted a tentative PSO for avoiding colony flying toward wrong direction.The results on the benchmark functions show the effectiveness of the proposed algorithm.