精英策略个体优势遗传算法研究

针对遗传算法存在的局部搜索能力差、早熟收敛和进化后期收敛速度慢的问题,提出了一种改进精英策略的个体优势遗传算法（Individual Advantages Genetic Algorithm,IAGA）。IAGA通过在精英子种群更新中不断增加精英个体数量和多样性,在保持算法全局收敛性的同时,增强算法在最优解区域的局部搜索能力。引入半粒子群变异算子,提高了算法前期向全局最优解靠拢的速度;引入个体优势算子,提高种群优势个体的多样性,有效改善了进化后期收敛速度慢的问题;与已有同类算法相比,平衡了收敛速度和全局收敛性之间矛盾的同时,进一步提高了收敛速度和精度。     

基于遗传算法求解应急决策系统中的最优路径

提出了一种将模拟退火算法和遗传算法相结合的进化算法GASA,利用Boltzmann机制 接收交叉和变异后的个体,避免遗传算法中存在的早熟收敛问题,增强了算法的全局收敛性,并对遗 传算子(选择、交叉、变异算子)进行重构,引入新的交叉算子和变异算子能根据种群的进化情况动态 调整遗传算子,加速进化后期搜索效率。实验表明,将此算法用于应急决策系统的最优路径的求解中 与传统算法相比,能加速进化速度和全局寻优能力,提高应急决策效率。     

一种改进的基于NSGA-II和DE的多目标混合进化算法

为提高非支配排序遗传算法（NSGA-II）的搜索精度和多样性,本文借鉴差分进化中加强局部搜索的策略,提出了一种改进的NSGA-II算法（LDMNSGA-II）。该算法利用拉丁超立方体抽样技术对解种群进行初始化,保证种群的初始分布能够均匀,采用差分进化中的变异引导算子和交叉算子替换NSGA-II的交叉算子,加强局部搜索能力和提高搜索精度,同时保留NSGA-II中的变异算子,保留算法多样性。四个经典测试函数的仿真结果表明,文中算法LDMNSGA-II在解决多目标优化问题中表现出良好的综合性能。     

可用于演化硬件的改进自适应遗传算法研究

为提高演化硬件在演化过程中的收敛速度,以解决其可扩展性问题,研究了标准遗传算法的3个遗传算子,分析了进化不同阶段对遗传算子的不同要求及其对收敛速度的影响.在Srinivas的自适应策略和基于阶段进化的自适应策略的基础上,提出一种新的针对变异算子的自适应策略,并在轮盘赌选择方式中加入适应值标度变换.结合实例,对改进后的算法进行了仿真,结果表明了加入适应值尺度变换和新的自适应策略后,算法的收敛性有所提高.     

混沌蜜蜂双种群进化遗传算法

利用混沌运动的遍历性、随机性和规律性等特点,提出混沌蜜蜂双种群进化遗传算法。该算法在基于蜜蜂双种群进化遗传算法的基础上,利用混沌优化进行改善初始种群质量和利用混沌退化变异算子代替常规算法中的变异算子,避免搜索过程陷入局部极值。实验结果表明,该算法计算速度快、收敛性好,提高了常规遗传算法的收敛速度和优化效果。     

多目标进化算法中变异算子的比较与研究

提出了一种适应于多目标进化算法的变异越界处理策略,成功地将这些变异算子应用到多目标进化优化问题中,从多目标优化收敛性的角度比较了这些变异算子的性能。通过一组实验表明这种越界处理方法是非常有效的,单目标优化中的这些变异算子具有与多项式变异算子相当的分布性,同时取得了更好的收敛性能。     

一种改进的生物地理学优化算法

生物地理学优化算法（BBO）作为一种新型的智能算法,在其提出不到十年的时间内受到学界的广泛关注和研究,并显示出了广阔的应用前景。为了提高算法的优化性能,对BBO算法提出一种改进,该算法在将差分优化算法（DE）中的局部搜索策略同BBO算法中的迁移策略相结合的基础上,针对迁移算子和变异算子分别进行改进,提出了二重迁移算子和二重变异算子,使得栖息地个体在进化过程中得到更高的进化概率,从而使得算法的寻优能力得到进一步提升。通过6个高维函数的测试,结果表明该算法在优化高维优化问题时,较其他几种生物地理学优化算法具有更好的收敛性和稳定性。     

进化神经网络中的变异算子研究

针对进化神经网络中遗传算法收敛速度慢和容易早熟这两个难题,提出了一个启发性的变异算子.该算子采用了自适应的变异率和启发式的变异位的选择策略.在多代无进化时,通过提高变异率扩大搜索范围,同时减小变异量进行更细致的搜索.求解XOR问题的实验表明,该算法既具有很快的收敛速度又能自动维持群体的多样性.     

基于改进遗传算法的可重构计算任务划分

为实现可重构计算中的软硬件任务自动划分,引入了遗传算法来搜寻最优解。为解决标准遗传算法可能出现种群早熟和种群进化后期收敛速度慢的问题,使用了小生境技术来保护种群中基因的多样性。设计了能够随适应度自动改变的自适应遗传算子（杂交算子和变异算子）。对算法进行了50次随机实验,并对结果进行分析。实验表明,改进后的遗传算法搜寻到全局最优任务划分的概率和搜寻到最优任务划分时的进化代数都要优于标准遗传算法。     

混沌变异算子的改进遗传算法及其应用

为解决标准遗传算法（SGA）收敛缓慢等缺点，提出一种混沌变异算子的改进遗传算法，进化过程中，为防止局部早熟收敛，对较优个体的变异操作中引入一个混沌变异算子，并把混沌运动的遍历范围“放大”到优化变量的取值范围，通过一代代地不断进化，收敛到一个最适合环境的个体上，求得问题的最优解；建立精英个体序列库，防止最优解的丢失。采用实际算例进行仿真试验，仿真结果证明了该算法的有效性。     

Refinable interpolatory and quasi-interpolatory operators

In this paper, we present a new class of operators, which are refinable, quasi-interpolatory and satisfy some interpolation conditions. The refinability is achieved by using as functional bases the B-bases corresponding to totally positive refinable functions. We analyze the main properties of the constructed refinable operators and give some convergence results. Some examples are also displayed.     

一种防止浮点遗传算法早熟收敛的父代选择策略

通过对浮点遗传算法早熟收敛现象的分析，提出了一种新的父代选择策略，既使用当前代的子代个体作为下代的父代个体，可使交叉算子持续地探索和开发新空间。引入对个体的代数保护策略，即在它发生变异前保证有足够的演化，可以避免对新空间不成熟的开发。通过与其它父代选择策略的对比，并通过实验和GENOCOP系统比较，表明本方法能得到较好的结果。     

A novel hybrid algorithm of gravitational search algorithm with genetic algorithm for multi-level thresholding

The multi-level thresholding is a popular method for image segmentation. However, the method is computationally expensive and suffers from premature convergence when level increases. To solve the two problems, this paper presents an advanced version of gravitational search algorithm (GSA), namely hybrid algorithm of GSA with genetic algorithm (GA) (GSA-GA) for multi-level thresholding. In GSA-GA, when premature convergence occurred, the roulette selection and discrete mutation operators of GA are introduced to diversify the population and escape from premature convergence. The introduction of these operators therefore promotes GSA-GA to perform faster and more accurate multi-level image thresholding. In this paper, two common criteria (1) entropy and (2) between-class variance were utilized as fitness functions. Experiments have been performed on six test images using various numbers of thresholds. The experimental results were compared with standard GSA and three state-of-art GSA variants. Comparison results showed that the GSA-GA produced superior or comparative segmentation accuracy in both entropy and between-class variance criteria. Moreover, the statistical significance test demonstrated that GSA-GA significantly reduce the computational complexity for all of the tested images.     

Convergence theorems for stochastic automata and learning systems

It is shown that every stochastic automaton generates a semigroup of continuous linear operators in order to give conditions for the convergence of certain operators connected with infinite input sequences. The main results of this note are conditions for the infinitesimal stability of stochastic automata. Since under suitable conditions the behaviour of a learning system can be represented by a stochastic automaton, these results apply to the asymptotic stability of learning.     

Morphological representation of nonlinear filters

Morphological operators provide very efficient algorithms for signal (image) processing. The efficiency of morphological operators has been captured by using them as approximations of nonlinear operators in numerous applications (e.g., image restoration). Our approach to the approximation of nonlinear operators is the construction of morphological bounds on them. We present a general theory on the morphological bounds on nonlinear operators, propose conditions for the existence of these bounds, and derive several fundamental morphological bounds.We also derive morphological bounds on iterations of nonlinear operators, which are superior to the original nonlinear operator in some applications. Because obtaining the results of the convergence of iterations of a nonlinear operator is often particularly desirable, we provide morphological bounds on the convergence of such iterations, and propose conditions for their convergence based on morphological properties. Finally, we propose several criteria for the morphological characterization of roots of nonlinear operators.This work was supported by U.S. Office of Naval Research award N00014-91-J-1725.     

辅助变量最小二乘辨识的均方收敛性

利用随机过程理论,首次证明了递推辅助变量最小二乘(RIVLS)的收敛性,研究了RIVLS算法的收敛速率,给出估算RIVLS算法均方参数估计误差上界的计算公式.分析表明,当辅助矩阵与信息矩阵的乘积是非奇异阵,且关于辅助向量的弱持续激励条件成立时,均方参数估计误差以(1/t)的速率收敛于零.这一研究结果对于提高RIVLS算法的实际应用效果具有重要意义.数字仿真例子表明了该结论的正确性.     

用于间歇化工过程最优设计的遗传算法

间歇化工过程的最优设计问题是一类复杂且难以求解的组合优化问题。通过把这类问题分解为只包含离散变量的主导问题和只含连续变量的子问题,把遗传算法和线性规划法结合起来对其进行求解。并在算法中引入了一类新的算子,显著地提高了收敛概率、算例表明,该方法可以避免直接求解过程的复杂性和困难,并且具有很好的全局收敛性。     

Optimisation With Real-Coded Genetic Algorithms Based On Mathematical Morphology

The goal of this work is to propose a novel approach to function optimisation by evolutionary techniques, in particular, real-coded genetic algorithms. A new genetic crossover operator, suitable for real codification, has been designed. This operator is called morphological crossover as it is based on mathematical morphology theory. The morphological crossover includes a new genetic diversity measure that has low computational cost. This operator is presented along with the resolution of a set of optimisation problems, including neural network training. The results are compared to other optimisation approaches as gradient descent methods or binary and real-coded genetic algorithms using different crossover operators. These tests show that the properties exhibited by the proposed operator when using real-coded genetic algorithms give higher convergence speed and less probability of being trapped in a local optimum.     

Comparison of discrete Hodge star operators for surfaces

We investigate the performance of various discrete Hodge star operators for discrete exterior calculus (DEC) using circumcentric and barycentric dual meshes. The performance is evaluated through the DEC solution of Darcy and incompressible Navier–Stokes flows over surfaces. While the circumcentric Hodge operators may be favorable due to their diagonal structure, the barycentric (geometric) and the Galerkin Hodge operators have the advantage of admitting arbitrary simplicial meshes. Numerical experiments reveal that the barycentric and the Galerkin Hodge operators retain the numerical convergence order attained through the circumcentric (diagonal) Hodge operators. Furthermore, when the barycentric or the Galerkin Hodge operators are employed, a super-convergence behavior is observed for the incompressible flow solution over unstructured simplicial surface meshes generated by successive subdivision of coarser meshes. Insofar as the computational cost is concerned, the Darcy flow solutions exhibit a moderate increase in the solution time when using the barycentric or the Galerkin Hodge operators due to a modest decrease in the linear system sparsity. On the other hand, for the incompressible flow simulations, both the solution time and the linear system sparsity do not change for either the circumcentric or the barycentric and the Galerkin Hodge operators.     

Frequency domain synthesis of trajectory learning controllers for robot manipulators

Trajectory learning control is a method for generating near to optimal feedforward control for systems that are controlled along a reference trajectory in repeated cycles. Iterative refinements of a stored feedforward control sequence corresponding to one cycle of the control trajectory is computed based upon the recorded trajectory error from the previous cycle. Several learning operators have been proposed in earlier work, and convergence proofs are developed for certain classes of systems, but no satisfactory method for design and analysis of learning operators under the presence of uncertainties in the system model have been presented. This article presents frequency domain methods for analyzing the convergence properties and performance of the learning controller when the amplitude and phase of the system transfer function is assumed to be within specified windows. Experimental results with an industrial robot manipulator confirm the theoretical results.