论文部分内容阅读
针对果蝇优化算法(FOA)寻优精度不高、容易陷入局部极值的缺陷,提出了一种具有莱维飞行搜索策略和精英反向学习的果蝇优化算法(LOBL-FOA)。首先,采用莱维飞行搜索模式对果蝇寻优过程中位置更新方式进行改进,使得算法具有较强的全局寻优能力,并在一定程度上避免了算法的过早收敛;其次,对精英果蝇个体进行反向学习生成反向解,保留具有较优味道浓度的果蝇个体,从而提高了算法搜索精度;最后,对5个经典测试函数在固定迭代次数和固定寻优精度条件下进行仿真测试,并同参考文献算法进行对比,结果表明本文提出的改进果蝇优化算法相较于传统果蝇优化算法具有较强的寻优精度和收敛效率。
In view of the defect that the FOA optimization accuracy is not high and is easy to fall into the local extreme, a fruit fly optimization algorithm (LOBL-FOA) with Levi flight search strategy and elite reverse learning is proposed. Firstly, the Levi flight search mode is used to improve the location updating method of Drosophila melanogaster, which makes the algorithm have strong global optimization ability and avoids premature convergence to some extent. Secondly, Inverted learning of fly individuals leads to the inverse solution and preserves Drosophila individuals with better taste concentration, which improves the searching accuracy of the algorithm. Finally, the simulation results of five classical test functions under the conditions of fixed iteration times and fixed optimization precision The results show that the improved Drosophila optimization algorithm proposed in this paper has better precision and convergence efficiency than the traditional Drosophila optimization algorithm.