一种量子模拟退火算法

为扩展量子智能算法的研究领域,根据模拟退火算法的思想,提出量子模拟退火算法(QSA).定义了量子染色体相位邻域空间,缩小了算法搜索范围;引入信息熵的概念,避免了搜索的盲目性;给出一个量子的旋转角增量的表达式,简化了计算过程;采用Boltzmann概率分布原则接受新解,提高了算法的搜索性能;同时增加了量子变异操作和量子随机行为,可以防止算法早熟现象.研究结果表明:该算法具有较强的全局收敛性和搜索能力.     

量子进化算法和免疫算法的比较研究

量子进化算法和免疫算法都是解决优化问题的强有力算法,.在分析了量子进化算法搜索的特点和免疫算法的机理基础上,对它们进行了比较,阐明了了二者的不同特点,并通过仿真实例总结出它们在求解多峰值函数优化问题上各自的优缺点.     

一种新的量子蚁群优化算法

 针对蚁群算法在求解连续空间优化问题时易于陷入局部最优和收敛速度慢的问题,提出了一种新的基于量子进化的蚁群优化算法。 该算法采用量子比特的概率幅表示蚂蚁当前位置信息;设计了一种新的量子旋转门更新蚂蚁位置, 完成蚂蚁的移动;最后采用量子 非门实现蚂蚁所在位置的变异, 增加位置的多样性。不仅从理论上证明了所提出算法的收敛性,而且通过仿真实验表明该算法可使 搜索空间加倍,比传统的蚁群算法具有更好的种群多样性,更快的收敛速度和全局寻优能力。     

概率门量子进化算法

量子进化算法(QEA)比传统进化算法(EA)有更好的种群多样性和全局寻优能力，但它采用概率操作过程，具有随机性和盲目性．将量子进化算法中的旋转门以概率门代替，在概率分析及实例验证的基础上，说明概率门量子进化算法(PGQEA)能使得对种群选取过程控制在全局优化的方向下，并且能更快地收敛于最优解。     

混合量子进化算法及其在多用户检测中的应用

将量子进化算法(QEA)和粒子群算法(PSO)互相结合,提出了两种混合量子进化算法.通过对多用户检测问题的求解表明,新的算法不仅操作更简单,而且全局搜索能力有了显著的提高.     

基于学习的并行免疫量子进化算法及收敛性

提出了基于学习的多宇宙并行免疫量子进化算法,算法中将种群分成若干个独立的子群体,称为宇宙。宇宙内采用免疫量子进化算法,宇宙间采用基于学习机制的移民、模拟量子纠缠的种群交叉等信息交互方式,使得进化算法具有更好的种群多样性,更快的收敛速度和全局寻优能力。不仅从理论上证明了该算法的收敛,而且通过仿真实验表明了该算法的优越性。     

量子进化算法在BP网络模型优化中的应用

BP网络的研究多年来主要集中于网络的结构与参数优化上,却忽略了对训练过的BP网络模型本身的优化.针对上述问题,提出了一种解决BP网络模型优化的量子进化算法.通过裙座锻造结构参数优化设计实例,表明量子进化算法较好地实现了BP网络模型的优化设计,可有效解决实际工程的优化问题.     

一种新型群体智能优化算法——微进化算法

提出了一种新型群体智能优化算法——微进化算法.该算法采用实数编码,基于个体自身历史最优位置,以群体中最优个体与当前个体的矢量差异信息作为指导,进行启发式搜索.数值实验结果表明:微进化算法简单有效、计算精度高、收敛速度快、鲁棒性强;此外,还具有参数设置简便、计算简单等特点.     

采用群体统计学习的量子进化算法

针对传统量子进化算法采用精英个体作为吸引子,存在种群学习范围窄、优秀基因易丢失的缺陷,提出了一种采用群体统计学习的量子进化算法.该算法抛弃了传统量子进化算法中的精英保留策略,通过截断、比例、竞赛选择等方式对进化过程中优秀群体统计分析后构建整个种群的吸引子,避免了以单一个体为单位的学习方式,能较为全面地从整个优秀种群学习知识,并保留群体的优秀基因信息.同时,吸引子每代更新,避免了采用精英保留策略易陷入局部极值的问题.通过测试实验表明,提出的算法搜索精度和效率提高,收敛速度更快,算法综合性能提高.     

基于改进量子进化算法的末端配送任务动态分配模型

大多数物流快递企业的配送业务末端会按照固定的配送服务区进行配送任务分配,无法针对变化频繁、分布不均的动态配送需求进行合理的配送资源设置,造成了各个末端配送节点工作负荷不均衡的现象,并进一步导致了配送调度管理混乱等问题。针对末端配送任务分配问题建立了一种考虑配送成本,资源利用率以及工作量配比差异的配送任务分配模型,对量子进化算法进行改进。对此问题求解,提出采用量子群稳定度作为算法退出判定条件,来避免算法的早退与无效迭代问题,并引入量子变异与淘汰机制,加强了算法对可行解的搜索能力。实验结果表明,与按配送区进行分配的方案相比,算法给出的方案有效缓解了配送任务分配不均的现象,同时也有效降低了总体配送成本。相关模型和算法可以根据动态的配送需求合理地分配各个末端网点的配送任务,有助于配送业务的下一步配送路径优化和科学调度。     

NIQGA改进及基于可变角距离的新型量子进化算法

为了提高量子进化算法的执行效率,在NIQGA算法基础上,通过改进△θi和S(αi,βi)参数表提出了一种改进算法INIQGA.又通过引入量子比特间角距离定义,提出了一种基于可变角距离旋转的量子进化算法QEA-VAR,该算法采用旋转门操作进行种群进化时,依据当前染色体中量子比特|φ〉i与最优解对应基态| 0〉或| 1〉的...     

A new evolutionary algorithm for function optimization

A new algorithm based on genetic algorithm(GA) is developed for solving function optimization problems with inequality constraints. This algorithm has been used to a series of standard test problems and exhibited good performance. The computation results show that its generality, precision, robustness, simplicity and performance are all satisfactory. Foundation item: Supported by the National Natural Science Foundation of China (No. 69635030), National 863 High Technology Project of China, the Key Scientific Technology Development Project of Hubei Province. Biography: GUO Tao(1971-), male, Ph D, research interests are in evolutionary computation and network computing.     

一种小生境正交遗传算法研究

针对标准遗传算法的不足,借助正交试验法的全局均衡设计思想和二元变异操作对初始种群产生方式、交叉算子和变异算子进行了改进,提高了种群的多样性;借助最优保留策略和自然界的小生境思想,对选择算子进行了改进,提高了算法的全局收敛性能;另外还通过引入加速正交搜索操作,提高了算法的收敛速度.在此基础上,提出了一种小生境正交遗传算法,并进行了实例研究.研究结果表明,该算法不但可以有效地克服标准遗传算法的缺陷,而且计算速度、计算精度和算法稳定性也得到了显著提高.     

解决图着色问题的膜进化算法

图着色问题是图论中比较热门的NP难问题之一。针对该问题,有许多启发式求解算法,但都存在求解的质量不高,计算时间较长等问题。近些年提出的膜进化算法,在处理NP难问题中展现出了独特的优势。基于膜进化算法框架,提出了解决图着色问题的膜进化算法,把图着色问题和膜结合,设计了复制、融合、分裂、溶解、融合分裂、禁忌搜索6种膜进化算子。这些算子在演变的过程中使膜和膜结构发生进化,从而找到更优解,最后求得解决方案。在DIMACS的40个挑战数据集上面进行了实验,与3个最新的图着色算法比较的结果表明：在保证解的质量的情况下,文中提出的膜进化算法能有效降低求解的时间,其中有58%的实例占优。     

进化算法在船体结构优化设计中的应用

结构优化问题在计算上的难点是计算复杂、存贮量大、计算时间长,解决问题的关键是如何提高处理整数与离散型变量的有效性.本文针对船舶结构优化设计问题的特点与计算上的难点,采用可以求解多峰性连续函数全局最优解的分配区间型进化算法进行结构优化.相对简单遗传算法(SGA)在解     

A parallel global-local mixed evolutionary algorithm for multimodal function optimization based on domain decomposition

This paper presents a parallel two-level evolutionary algorithm based on domain decomposition for solving function optimization problem containing multiple solutions. By combining the characteristics of the global search and local search in each sub-domain, the former enables individual to draw closer to each optima and keeps the diversity of individuals, while the latter selects local optimal solutions known as latent solutions in sub-domain. In the end, by selecting the global optimal solutions from latent solutions in each sub-domain, we can discover all the optimal solutions easily and quickly. Foundation item: Supported by the National Natural Science Foundation of China (60133010,60073043,70071042) Biography: Wu Zhi-jian(1963-), male, Associate professor, research direction: parallel computing, evolutionary computation.     

Asynchronous parallel evolutionary algorithms for constrained optimizations

Recently Guo Tao proposed a stochastic search algorithm in his PhD thesis for solving function optimization problems. He combined the subspace search method (a general multi-parent recombination strategy) with the population hill-climbing method. The former keeps a global search for overall situation, and the latter keeps the convergence of the algorithm. Guo's algorithm has many advantages, such as the simplicity of its structure, the higher accuracy of its results, the wide range of its applications, and the robustness of its use. In this paper a preliminary theoretical analysis of the algorithm is given and some numerical experiments has been done by using Guo's algorithm for demonstrating the theoretical results. Three asynchronous parallel evolutionary algorithms with different granularities for MIMD machines are designed by parallelizing Guo's Algorithm. National Laboratory for Parallel and Distributed Processing Foundation item: Supported by the Natonal Natural Science Foundation of China (No. 70071042, 50073043), the National 863 Hi-Tech Project of China (No. 863-306-ZT06-06-3) and the National Laboratory for Parallel and Distributed Processing. Biography: Kang Li-shan (1934-), male, Professor, research interests: parallel computing and evolutionary computation.     

Novel evolutionary algorithm for global numerical optimization with continuous variables

Evolutionary algorithms (EA) are a class of general optimization algorithms which are applicable to functions that are multimodal, non-differentiable, or even discontinuous. In this paper, a novel evolutionary algorithm is proposed to solve global numerical optimization with continuous variables. In order to make the algorithm more robust, the initial population is generated by combining determinate factors with random ones. And a decent scale function is designed to tailor the crossover operator so that it can not only find the decent direction quickly but also keep scanning evenly in the whole feasible space. In addition, to improve the performance of the algorithm, a mutation operator which increases the convergence-rate and ensures the convergence of the proposed algorithm is designed. Then, the global convergence of the presented algorithm is proved at length. Finally, the presented algorithm is executed to solve 24 benchmark problems. And the results show that the convergence-rate is noticeably increased by our algorithm.     

Novel evolutionary algorithm for global numerical optimization with continuous variables

Evolutionary algorithms (EA) are a class of general optimization algorithms which are applicable to functions that are multimodal, non-differentiable, or even discontinuous. In this paper, a novel evolutionary algorithm is proposed to solve global numerical optimization with continuous variables. In order to make the algorithm more robust, the initial population is generated by combining determinate factors with random ones. And a decent scale function is designed to tailor the crossover operator so that it can not only find the decent direction quickly but also keep scanning evenly in the whole feasible space. In addition, to improve the performance of the algorithm, a mutation operator which increases the convergence-rate and ensures the convergence of the proposed algorithm is designed. Then, the global convergence of the presented algorithm is proved at length. Finally, the presented algorithm is executed to solve 24 benchmark problems. And the results show that the convergence-rate is noticeably increased by our algorithm.     

遗传算法参数的正交设计

遗传算法参数的设置问题一直是遗传算法中的重要研究课题之一。本文探索用正文法对遗传算法中的参数进行设置。