遗传算法的收敛性统一判据

本文针对遗传算法的过早收敛或者收敛缓慢甚至不收敛,进行了详细的分析;理论分析了算法发生过早收敛的原因,收敛速度与各个控制参数和遗传操作的关系;提出并严格证明了与编码方式和选择策略无关的判断遗传算法收敛性判据。     

Local control strategies for groups of mobile autonomous agents

The problem of achieving a specified formation among a group of mobile autonomous agents by distributed control is studied. If convergence to a point is feasible, then more general formations are achievable too, so the focus is on convergence to a point (the agreement problem). Three formation strategies are studied and convergence is proved under certain conditions. Also, motivated by the question of whether collisions occur, formation evolution is studied.     

Estimate of exponential convergence rate and exponential stabilityfor neural networks

Estimates of exponential convergence rate and exponential stability are studied for a class of neural networks which includes Hopfield neural networks and cellular neural networks. Both local and global exponential convergence are discussed. Theorems for estimation of the exponential convergence rate are established and the bounds on the rate of convergence are given. The domains of attraction in the case of local exponential convergence are obtained. Simple conditions are presented for checking exponential stability of the neural networks.     

A note on causal and CITE iterative learning control algorithms

The convergence properties of causal and current iteration tracking error (CITE) discrete time iterative learning control (ILC) algorithms are studied using time and frequency domain convergence criteria. Of particular interest are conditions for monotone convergence, and these are evaluated using a discrete-time version of Bode's integral theorem.     

Logistic阻滞增长模型的计算机模拟

将Logistic阻滞增长模型的差分形式进行简化,对Logistic模型的简化差分形式进行迭代求解,做出了随固有增长率r的变化,序列狖yk狚收敛、2倍周期、4倍周期……直至一片混乱的图形。以参数b为横坐标、序列狖yk狚的收敛点为纵坐标,用MATLAB数学软件模拟展示了这一简单差分方程从收敛、分叉、2n倍周期收敛进入混沌现象的过程。     

基于改进遗传算法的PID参数寻优与控制器设计

相比传统的调节方法,遗传算法具有更好的鲁棒性、最优性,能较好的实现参数的自动化调节。对标准遗传算法（SGA）进行了分析、研究,并在SGA的基础上进行了改进。改进的遗传算法从提高全局搜索性能和加快收敛速度出发,提出了改进的选择算子、交叉算子和变异算子,仿真结果表明,改进的遗传算法的全局搜索性能和收敛速度远远优于标准遗传算法。     

三网融合安全问题及对策研究

文章全面分析三网融合中的安全隐患,对三网融合中的安全问题进行了全面深入地分析,并提出相应的对策和建议,。在三网融合潜在隐患的分析过程中,重点探讨了EPON技术、IPTV业务和VoIP业务中存在的安全隐患和风险,以及融合网络之间安全威胁相互扩散的问题,挖掘出了其发展过程中存在的一些安全隐患和风险问题,并分别提出了相应的对策和建议。     

On exponential convergence and robustness of least-squares identification

Robustness and convergence properties of exponentially weighted least-squares identification are studied. It is shown that exponential convergence in the noiseless case can be obtained for a class of increasing or decreasing regression vectors. The rate of change of the limits in the regressors affect the convergence rates, which are explicitly given. It is demonstrated that for a sub-class of regressors decreasing-in-the-norm exponential convergence without the noise does not guarantee robustness subject to a bounded noise. Instead, exponential divergence of the estimate is shown in a specific case.     

Convergence of sequences of fuzzy perceptive elements defined on different possibility spaces

Properties of the convergence of sequences of fuzzy perceptive elements defined on possibility spaces are studied, and results are obtained that can be applied to the investigation of a possibilistic analog of the law of large numbers for the convergence in possibility and convergence with necessity 1.     

最优保留遗传算法的研究与分析

最优保留GA（EGA）是目前GA收敛性研究中比较典型的一类。文章在已有的研究成果基础上,给出了EGA更一般的规范化定义,指明了EGA全局收敛的本质及其两种实现方式,并分别对它们进行了收敛性分析。     

广义信息熵测度在医学图像配准中的应用

针对互信息测度在配准医学图像时易陷入局部极值的缺点,将Shannon熵扩展到广义熵,提出了三种基于广义熵的信息测度。对于收敛性能的评价,提出收敛宽度和收敛半径的概念。通过人体脑部CT/MR和MR-T1/T2图像的刚体配准实验,从计算时间、收敛性能和配准精度方面,对归一化互信息、广义熵信息测度进行了比较与分析。实验结果表明,在不损失计算时间和配准精度的前提下,广义信息熵测度SRI_0.9和GMI_0.9的收敛性能优于归一化互信息测度,对噪声有很强的鲁棒性。     

元胞遗传算法的收敛性分析和收敛速度估计

元胞遗传算法是空间结构化种群的遗传算法,将遗传操作限制在相邻个体之间进行,限制优势基因的扩散速度,保持种群的多样性,改善遗传算法的性能。但是,目前有关元胞遗传算法收敛性的分析还较缺乏。文中根据元胞遗传算法的特性,建立元胞遗传算法的吸收态 Markov链模型,证明元胞遗传算法的收敛性。提出元胞遗传算法的首达最优解期望时间的估算方法,并估计标准同步元胞遗传算法首达最优解期望时间的上下界。     

Identification of discrete Hammerstein systems by the Fourier series regression estimate

The identification of a single-input, single-output (SISO) discrete Hammerstein system is studied. Such a system consists of a non-linear memoryless subsystem followed by a dynamic, linear subsystem. The parameters of the dynamic, linear subsystem are identified by a correlation method and the Newton-Gauss method. The main results concern the identification of the non-linear, memoryless subsystem. No conditions are imposed on the functional form of the non-linear subsystem, recovering the non-linear using the Fourier series regression estimate. The density-free pointwise convergence Of the estimate is proved, that is.algorithm converges for all input densities The rate of pointwise convergence is obtained for smooth input densities and for non-linearities of Lipschitz type.Globle convergence and its rate are also studied for a large class of non-linearities and input densities     

On convergence properties of piecewise affine systems

In this paper convergence properties of piecewise affine (PWA) systems are studied. In general, a system is called convergent if all its solutions converge to some bounded globally asymptotically stable steady-state solution. The notions of exponential, uniform and quadratic convergence are introduced and studied. It is shown that for non-linear systems with discontinuous right-hand sides, quadratic convergence, i.e., convergence with a quadratic Lyapunov function, implies exponential convergence. For PWA systems with continuous right-hand sides it is shown that quadratic convergence is equivalent to the existence of a common quadratic Lyapunov function for the linear parts of the system dynamics in every mode. For discontinuous bimodal PWA systems it is proved that quadratic convergence is equivalent to the requirements that the system has some special structure and that certain passivity-like condition is satisfied. For a general multimodal PWA system these conditions become sufficient for quadratic convergence. An example illustrating the application of the obtained results to a mechanical system with a one-sided restoring characteristic, which is equivalent to an electric circuit with a switching capacitor, is provided. The obtained results facilitate bifurcation analysis of PWA systems excited by periodic inputs, substantiate numerical methods for computing the corresponding periodic responses and help in controller design for PWA systems.     

Characterization of the convergence of stationary Fokker–Planck learning

The convergence properties of the stationary Fokker–Planck algorithm for the estimation of the asymptotic density of stochastic search processes is studied. Theoretical and empirical arguments for the characterization of convergence of the estimation in the case of separable and nonseparable nonlinear optimization problems are given. Some implications of the convergence of stationary Fokker–Planck learning for the inference of parameters in artificial neural network models are outlined.     

Generalization of Linear Cyclic Pursuit With Application to Rendezvous of Multiple Autonomous Agents

Cyclic pursuit is a simple distributed control law in which agent i pursues agent i+1 modulo n. We generalize existing results and show that by selecting the gains of the agents, the point of convergence of these agents can be controlled. The condition for convergence, the range of controller gains and the reachable set where convergence can occur are studied. It is also shown that the sequence in which an agent pursues another does not affect the point of convergence     

蚁群优化算法的收敛性分析

有关蚁群优化算法收敛性分析的研究还很少．不利于进一步改进其算法．为此,较详细地分析了用蚁群优化算法求解TSP问题的收敛性．证明了当0〈q0〈1时．算法能够收敛到最优解．分析了封闭路径性质、启发函数、信息素和q0对收敛性的影响．据此给出了提高算法收敛速度的几点结论．     

The convergence of Hamming memory networks

The convergence properties of Hamming memory networks are studied. It is shown how to construct the network so that it probably converges to an appropriate result, and a tight bound is given on the convergence time. The bound on the convergence time is largest when several stored vectors are at the minimum distance from the input vector. For random binary vectors, the probability for such situations to occur is not small. With a specific choice of parameter values, the worst-case convergence time is on the order of p ln (pn), where p is the memory capacity and n is the vector length. By allowing the connection weights to change during the computation, the convergence time can be decreased considerably.     

On convergence of double splitting methods for non-Hermitian positive semidefinite linear systems

Some convergence results for double splitting iterations for (possibly non-Hermitian) positive semidefinite linear systems are established. Furthermore, the convergence of double splitting methods for generalized saddle point systems is studied, and a convergence condition for double splitting methods applied to this type of system is given.     

Order barriers for the B-convergence of ROW methods

The convergence of ROW methods is studied when these methods are applied to the stiff model equation of Prothero and Robinson. It turns out that there are barriers of the attainable order of convergence. Furthermore, the existence of ROW methods is shown the accuracy of which increases asymptotically with the stiffness of the model. The theoretical results are demonstrated by numerical examples.