摘要
分析了人工蜂群算法及部分国内外学者提出的改进算法,针对局部搜索能力差和容易陷入局部最优解的缺点,根据马尔可夫链预测已知解空间的发展趋势,提出了一种基于马尔可夫链的改进人工蜂群算法(MABC),通过伪代码给出了算法的运行过程,从收敛性能和算法复杂度2个方面分析了人工蜂群算法、一种典型的改进算法和MABC算法的性能.最后以10个典型函数为测试用例,从结果精度、收敛速度、分割参数和运行时间4个方面进行验证,实验结果表明,MABC算法在求解精度和收敛速度上高于ABC算法,但运行时间略长,验证了理论分析的结果.
To overcome the shortcomings of existing local search ability and to easily obtain the local optimal solution of artificial bee colony algorithm(ABC),a new modified artificial bee colony algorithm(MABC)is proposed using the development trend of known solution space predicted by Markov Chain.The running process of the algorithm is provided through a pseudo code.The performances of the ABC and MABC are analyzed from two aspects:convergence performance and algorithm complexity.Using 10 typical functions as test cases,Experiments are carried out in four aspects:result precision,convergence speed,segmentation parameters and running time.It is shown that the MABC algorithm is superior to the ABC algorithm in terms of accuracy and convergence speed.
作者
郭佳
马朝斌
苗萌萌
张绍博
GUO Jia;MA Chao-bin;MIAO Meng-meng;ZHANG Shao-bo(School of Computer and Information Technology,Beijing Jiaotong University,Beijing 100044,China;National Secrecy Science and Technology Evaluation Center,Beijing 100044,China)
出处
《北京邮电大学学报》
EI
CAS
CSCD
北大核心
2020年第1期54-60,共7页
Journal of Beijing University of Posts and Telecommunications
关键词
马尔可夫链
人工蜂群算法
函数优化
Markov chain
artificial bee colony algorithm
function optimization
作者简介
郭佳(1983—),女,工程师,E-mail:m13581902161@163.com.
