Abstract: Moth-Flame Optimization（MFO） algorithm, which is inspired by social behaviors of individuals in moth flame, is a nature-inspired and easy to implement global optimization algorithm. The MFO has shown good performance for many real-world optimization tasks. However, MFO has problems with premature convergence and easy trapping into local optimum solutions. In order to overcome these deficiencies, an Ameliorative MFO（AMFO） algorithm based on Kent chaotic dynamic inertia weight is proposed. In the AMFO algorithm, a Kent chaotic map search strategy is introduced to help the current optimal solution jump out of the local optimal solution. In addition, the dynamic inertia weight strategy based on fitness value and iteration number is used to balance the development and exploration ability and further improve the performance of MFO algorithm. The search accuracy and performance of AMFO algorithm are verified on 8 classical benchmark functions. The experimental results show that AMFO technique has superior search performance compared with standard MFO algorithm, particle swarm optimization algorithm and differential evolution algorithm.

Key words: swarm intelligence, moth-flame optimization algorithm, Kent chaotic, dynamic inertia weight, numerical function

摘要： 飞蛾火焰优化算法（Moth-Flame Optimization，MFO）是一种自然激励且易于实现的全局优化算法，在许多实际优化任务中表现出良好的性能。然而，MFO算法存在早熟收敛和容易陷入局部最优解的问题，针对这些不足，提出了一种Kent混沌动态惯性权值的改善飞蛾火焰优化算法（Ameliorative MFO，AMFO）。在AMFO算法中，引入Kent混沌映射搜索策略帮助当前最优解跳出局部最优；采用基于适应度值和迭代次数的动态惯性权值策略来平衡算法的开发和探索能力，以进一步提升MFO算法性能。在8个经典benchmark函数上验证AMFO算法的搜索精度和性能，并将其结果与标准飞蛾火焰优化算法、粒子群算法和差分进化算法进行比较，仿真结果表明AMFO算法具有较好的搜索性能。

关键词: 群智能, 飞蛾火焰优化算法, Kent混沌, 动态惯性权值, 数值函数