量子谐振子优化算法  

Quantum harmonic oscillator optimization algorithm

作　　者：肖黎彬[1] 王鹏[1,2] 陈磊[1] 郭又铭[2,3] 

机构地区：[1]成都信息工程学院并行计算实验室,成都610225 [2]中国科学院成都计算机应用研究所,成都610041 [3]中国科学院研究生院,北京100049

出　　处：《计算机应用》2012年第A02期1-4,44,共5页journal of Computer Applications

基　　金：国家自然科学基金资助项目(60702075);广东省科技厅科技计划高新技术产业化科技攻关项目(2011B010200007);成都市科技局创新发展战略研究项目(11RKYB016ZF)

摘　　要：量子谐振子的振动物理过程与智能算法的工作机制有内在的相似性,结合量子谐振子振动空间稳定的收敛性和基态高斯曲线分布的特性,提出了基于量子谐振子基态最优性的优化算法模型。从理论上分析了量子谐振子基态的最优特性以及它和智能算法的对应关系,将这种关系对应到算法模型的构建,理论上证明了由量子谐振子模型构建的算法能够在解空间形成高斯曲线的分布形式,并能够在势阱的约束下快速收敛到最优解。最后将该算法应用于求解旅行商问题(TSP),通过选取三组实验数据,将该算法与同等规模下的模拟退火算法进行比较,实验结果表明量子谐振子算法具备更好的收敛性和寻优能力。An optimization algorithm named quantum harmonic oscillator optimization algorithm based on quantum harmonic oscillator ground state of optimality was proposed, combined with the stable convergence of the vibration space and the ground state Gaussian curve distribution characteristics of quantum harmonic oscillator. The vibration of the physical processes of the quantum harmonic oscillator has an inherent similarity with the working mechanism of the intelligent algorithm. This paper mainly analyzed the characteristics of quantum harmonic oscillator and its relation to optimization algorithm, and the reiationship between them is one of the reasons to construct the algorithm model. It is theoretically proved that the algorithm based on quantum harmonic oscillator will make the solution space form the Gaussian distribution, and will provide a better searching capability. In order to test the performance, the algorithm was used to solve the corresponding scale of Traveling Saleman Problem （TSP）, and compared with the simulated anneal algorithm under equal scale, the results of experiment show that the algorithm has better convergence and enhanced optimization capability.

关 键 词：智能优化算法 量子谐振子 基态 解空间 高斯分布 旅行商问题 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]