基于精英候选池的蝙蝠算法

Bat algorithm based on elite candidate pool

针对蝙蝠算法搜索后期容易陷入局部最优,从而导致算法精度不高、停滞等不足,提出了精英候选池策略为每只蝙蝠选取待追随的最优蝙蝠,从而增强了种群的多样性,避免算法过早成熟。在10个Benchmark测试函数中进行仿真实验,实验结果表明,基于候选池的蝙蝠算在在收敛精度和速度上都有较大的提高。

Bat algorithm( BA) has such limitations as low accuracy and stagnation,because it is prone to fall into the local optimum during the late searching period. In order to improve the diversity of population,bat algorithm with elite candidate pool( EBA) is proposed to make each bat select the pursed bat from the elite bats pool. The elite pool method enables the algorithm to avoid successfully premature. Ten Benchmark testing functions have been conducted. The experiment results show that EBA outperforms BA on convergence accuracy and convergence speed on most test functions.