一种新的量子行为衍生布谷鸟搜索算法  

Novel Quantum-behaved Inspired Cuckoo Search Algorithm

作　　者：曲宝[1] 李荟[1] Qu Bao;Li Hui(Northeast Petroleum University,Daqing 163318)

机构地区：[1]东北石油大学,大庆163318

出　　处：《黑龙江八一农垦大学学报》2021年第4期86-92,共7页journal of heilongjiang bayi agricultural university

基　　金：黑龙江省自然科学基金(LH2020F003)。

摘　　要：为提高布谷鸟搜索算法的优化能力,从研究布谷鸟算法的实现机制入手,提出一种量子行为衍生布谷鸟算法。算法中的鸟窝位置采用Bloch球面描述的量子比特编码,个体更新采用Bloch球面上的绕轴旋转方式,引入莱维飞行随机走动控制绕轴旋转角度大小,搜索范围限定到Bloch上半球,最后根据种群退化程度建立自适应发现概率。函数极值优化的仿真结果表明,该算法的单步迭代时间为普通布谷鸟算法的9倍左右,但寻优能力有明显提高,相对于其他量子进化算法也有一定程度的提升。To enhance the performance of cuckoo search algorithm,by studying the implementation mechanism of the cuckoo search algorithm,a new quantum-behaved inspired cuckoo algorithmwas proposed.The bird's nest location was encoded by the qubits described on the Bloch sphere.The individual evolution was achieved by the rotating aroud the axis.It used the Lévy fight random walk to control the size about rotating angle.The search scope was limited to the Bloch upper hemisphere.According the degenerative degree of populization,it constructed the adaptive discovery probability.Finally,the simulation results of optimizing the extreme value of functions indicates that,for one interative step,the average time of the proposed approachwas 9 times as long as that of the common cuckoo search algorithm.However,it was obviously efficient in optimization ability than cuckoo algorithm and also got a certain degree of improvement than other quantum evolutionary algorithms.

关 键 词：布谷鸟搜索 莱维飞行 量子计算 量子比特 Bloch球面 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程]