基于模拟退火的量子进化算法

针对量子进化算法全局搜索能力强而局部寻优能力弱的特点,提出一种基于模拟退火的量子进化算法。该方法将模拟退火算法引入到量子进化算法中,在采用量子进化算法进行解空间全局搜索的同时,用模拟退火算法加强局部寻优能力,以有效平衡算法的开采与勘探能力。采用著名的NP难组合优化问题———背包问题为例进行实验,结果表明:本文方法获得了比量子进化算法更好的解,证实了其有效性。

Quantum-inspired Evolutionary Algorithms Based on Simulated Annealing

Aiming at the features that quantum-inspired evolutionary algorithms(QEA) have good global search capability and bad local search performance,this paper proposes a simulated annealing based QEA(SA-QEA).The SA-QEA combines simulated annealing with QEA,and applies QEA to explore the whole solution space and employs simulated annealing to exploit the neighbor domains.Thus,SA-QEA is able to balance the exploration and exploitation.Knapsack problems,well-known NP-hard combinatorial optimization problems,are taken for examples to test SA-QEA performances.Experimental results show that SA-QEA obtains better solutions than QEA,which proves the effectiveness of SA-QEA.