一种高效的混合蝙蝠算法

Efficient hybrid bat algorithm

针对基本蝙蝠算法存在收敛速度慢,易陷入局部最优,求解精度低等缺陷,提出一种融合局部搜索的混合蝙蝠算法用于求解无约束优化问题。该算法利用混沌序列对蝙蝠的位置和速度进行初始化,为全局搜索的多样性奠定基础;融合Powell搜索以增强算法的局部搜索能力,加快收敛速度;使用变异策略在一定程度上避免算法陷入局部最优。选取几个标准测试函数进行仿真实验,结果表明:与基本蝙蝠算法和粒子群优化算法相比,混合蝙蝠算法具有更好的寻优性能。

The Bat Algorithm（BA）has a few disadvantages in the global searching, including slow convergence speed, high possibility of being trapped in local optimum and low solving precision. A hybrid bat algorithm based on Powell search method is proposed to solve unconstrained optimization problems. Firstly, chaotic sequence is used to initiate indi-vidual position and velocity, which strengthens the diversity of global searching. It combines the bat algorithm and Powell search to enhance the ability of local search and improve the convergence speed of algorithm. Mutation strategy is used to prevent the algorithm into a local optimum in a certain extent. The experimental results show that the proposed algorithm is more effective and feasible than the standard BA and Particle Swarm Optimization（PSO）algorithm.