摘要
针对标准布谷鸟算法的相关问题,提出了一种基于logistic模型的动态步长控制因子和动态发现概率的改进布谷鸟算法。改进的算法在运行时可以自动调节步长控制因子和发现概率的大小,且在算法初期可以使种群保持多样性,提高了全局最优值的搜索能力;随着局部最优值搜索能力的增强,算法在后期逐渐趋于稳定。通过用几种典型的Benchmarks函数进行模拟试验,试验结果证实了所提出的算法计算精度高、收敛速度快。
On the basis of related presentations of the standard cuckoo algorithms, an improved cuckoo algorithm of dynamic step-size control factor and dynamic find-statistic based on logistic model is presented. The improved algorithms can auto-matically adjust step-size control factor and find-statistic during the running time, so it can keep the individuals diversity and improve searching ability of global optimum in the population at the initial generations. However, the algorithm is gradually stabilized with searching ability of local optimum improved at a later time. Several classic Benchmarks func-tions are tested and the results show that the proposed algorithms have fast convergence and higher calculation accuracy.
出处
《计算机工程与应用》
CSCD
北大核心
2015年第20期31-35,共5页
Computer Engineering and Applications
基金
国家自然科学基金(No.41474100)
山东省自然科学基金(No.ZR2013DM015)
