融合莱维飞行和小波变异的蜣螂优化算法  

Dung Beetle Optimization Algorithm Integrating Levy Flight and Wavelet Variation

作　　者：郭琴 郑巧仙 GUO Qin;ZHENG Qiaoxian(School of Computer and Information Engineering,Hubei University,Wuhan 430062,China)

机构地区：[1]湖北大学计算机与信息工程学院,武汉430062

出　　处：《小型微型计算机系统》2024年第12期2882-2890,共9页Journal of Chinese Computer Systems

基　　金：国家自然科学基金项目(61803149)资助。

摘　　要：针对蜣螂优化算法(Dung Beetle Optimizer, DBO)搜索能力不足,易陷入局部最优,收敛速度慢等问题,提出了一种融合莱维飞行和小波变异的蜣螂优化算法(Dung beetle optimization algorithm integrating Levy flight and wavelet variation, LWDBO).在偷窃蜣螂中引入莱维飞行策略,莱维飞行的随机步长特点可以扩大搜索空间,加快算法的收敛速度;采用融入小波理论的变异策略,根据算法的迭代深度动态调整变异概率和变异程度,增大了算法跳出局部最优解的概率.对12个基准函数进行仿真实验,对2种改进策略分别进行有效性分析,结果表明2种改进策略均提高了算法的收敛精度和稳定性,同时融合2种策略的LWDBO则进一步提升了算法的寻优性能;将LWDBO与其他新型智能优化算法进行比较,进一步证明了该算法的可靠性与高效性.将该算法应用于2种工程约束优化问题中,验证了该算法的有效性及工程实用性.Aiming at the problems of insufficient search ability of Dung Beetle Optimizer(DBO),easy to fall into local optimization and slow convergence speed,a Dung beetle optimization algorithm integrating Levy flight and wavelet variation(LWDBO)is proposed.The Levy flight strategy is introduced in the stealing dung beetle,and the random step size feature of Levy's flight can expand the search space and accelerate the convergence speed of the algorithm;The mutation strategy integrated into wavelet theory is adopted,and the variation probability and degree are dynamically adjusted according to the iteration depth of the algorithm,which increases the probability of the algorithm jumping out of the local optimal solution.Simulation experiments are carried out on 12 benchmark functions,and the effectiveness analysis of the two improvement strategies shows that the convergence accuracy and stability of the two improved strategies are improved,and LWDBO integrating the two strategies further improves the optimization performance of the algorithm.Comparing LWDBO with other new intelligent optimization algorithms further proves the reliability and efficiency of the algorithm.The algorithm is applied to two engineering constraint optimization problems,and the effectiveness and engineering practicability of the algorithm are verified.

关 键 词：蜣螂优化算法 莱维飞行 随机步长 小波理论 局部最优解 

分 类 号：TP301[自动化与计算机技术—计算机系统结构]