基于随机扰动的自适应布谷鸟算法

An Adaptive Cuckoo Algorithm Based on Random Disturbance

布谷鸟搜索算法(CS)是模仿布谷鸟的繁殖行为所建的一种元启发式算法。这是一种新兴启发算法,通过模拟某些种属布谷鸟的寄生育雏崽来有效地求解最优化问题。针对该算法计算精度不高,收敛速度慢,容易陷入局部最优等缺陷,提出了一种基于自适应步长随机扰动的布谷鸟搜索算法(ASCS)。在增加鸟窝位置变化活力的基础上,对鸟窝位置之间的距离引入自适应的调整步长因子,可以防止算法在运行过程中陷入局部最优。同时为了更大程度地提高鸟窝的计算精度与搜索速度,在寻找最优鸟窝的时候增加一个扰动因子,提高了算法的收敛速度。通过7个测试函数进行仿真实验,结果证明了该算法的可行性,其性能显著优于原始的布谷鸟算法。

The cuckoo search (CS) is a kind of meta-heuristic algorithm constructed by imitating the breeding behavior of cuckoos.This is an emerging heuristic algorithm that effectively solves optimization problems by simulating the breeding of some species of cuckoos.Aiming at the shortcomings of this algorithm,such as low computational accuracy,slow convergence speed and easy to be trapped in local optimum,we propose a cuckoo search based on adaptive step size random perturbation (ASCS).On the basis of increasing the vitality of the bird nest position,an adaptive adjustment step factor is introduced to the distance between the bird nest positions to prevent the algorithm from falling into a local optimum during the operation.At the same time,in order to improve the calculation accuracy and search speed of the bird nest to a greater extent,an interference factor is added to find the optimal bird nest,which improves the convergence speed of the algorithm.The simulations are performed by seven test functions.The experiment shows that the ASCS is feasible and its performance is significantly better than that of CS.