一种改进双链量子遗传算法及其应用.docVIP

一种改进的双链量子遗传算法及其应用 　　摘 要:针对目前双链量子遗传算法中保持种群多样性和改善优化效率问题提出了三种改进方法。通过在量子比特概率幅三角函数表达式中引入常数因子,使搜索过程在多个周期上同时进行,以改善算法的优化效率;提出了一种基于单比特量子Hadamard的变异策略,可提高保持种群多样性的概率;改进了量子旋转门转角步长函数,能够有效避免算法震荡,增强算法的适应性。以多变量函数极值优化问题为例,仿真实验结果表明上述三种改进措施是有效的。 　　关键词:量子计算; 量子遗传算法; 优化算法 　　中图分类号:TP183文献标志码:A 　　文章编号:1001-3695(2010)06-2090-03 　　doi:10.3969/j.issn.1001-3695.2010.06.027 　　 　　Improved quantum genetic algorithm with double chains and its application 　　 　　XU Shao-hua, XU Chen, HAO Xing, WANG Ying, LI Pan-chi 　　 　　(School of Computer Information Technology, Daqing Petroleum Institute, Daqing Heilongjiang 163318, China) 　　 　　Abstract:Aiming at the problems that how to keep population diversity and improve optimization efficiency in double chains quantum genetic algorithm, this paper proposed three improvements. Firstly, by adding the constant factor to the trigonometric expressions of quantum bit probability amplitudes, performed the search in a number of trigonometric functions cycle at the same time, which enhanced the optimization efficiency of the proposed algorithm. Secondly, the mutation strategy applying the single bit quantum Hadamard gates enhanced the diversity of population. Thirdly, enhanced the adaptability of the proposed algorithm by redesigning the step function of rotation angle of quantum rotation gates, and this also avoided the oscillation effectively. Finally, with application of function extremum optimization with multi-variables, the simulation results show that the three improvements are efficient. 　　Key words:quantum computing; quantum genetic algorithm(QGA); optimization algorithm 　　 　　量子计算是信息科学和量子力学相结合的新兴交叉学科,自1994年Shor提出第一个求解大数质因子分解的量子算法[1]和1996年Grover提出随机数据库搜索的量子算法[2]之后,量子计算以其独特的计算性能引起了广泛关注,迅速成为国际上研究的热点[3～5]。量子遗传算法(QGA)是一种基于量子计算原理的概率优化算法,它以量子计算的一些概念和理论为基础,用量子位编码来表示染色体,用量子门更新染色体来完成进化搜索,具有种群规模小、收敛速度快、全局搜索能力强的优点。目前,量子计算与遗传算法的融合已呈现出众多模式。Narayanan等人[6]首先提出了量子衍生遗传算法并成功求解了TSP, 虽然该方法仍属传统遗传算法, 但激发了量子计算原理与遗传算法相结合的研究;Han等人[7]采用量子位编码和量子门更新染色体,提出了遗传量子算法和并行量子遗传算