大规模光电混合神经网络中的可编程拓扑重构

提出了一种在大规模光电混合神经网络系统实现可编程拓扑重构的有效方法，文中介绍了这种方法的原理，运用技术以及采用这种方法在已有的ＮＰ１０２４光学／数字神经网络处理器中运行单层反馈型和多层前馈型等不同网络拓扑结构的实验结果。     

基于加强型BP算法的神经网络故障诊断研究

文章给出了利用神经网络进行故障诊断的可行性、基本原理、诊断过程分析、和参数优化过程 ,并得出较为理想的诊断结果 .     

Permeability of Clustered Fiber Networks: Modeling of the Unit Cell

The paper is focused on estimating the permeability of a clustered fiber network by variational methods. First, a laminar flow in ducts is considered by using polynomial trial functions. Then, a longitudinal flow through a square array is described by expanding the flow-rate field in trigonometric and Laurent series. Finally, a formal scheme for estimating the longitudinal permeability in a cluster with an irregular distribution of fibers is given. The irregular distribution is modeled by setting an individual effective radius for each fiber and then letting this fiber reach its minimum gravitational energy. The results obtained here form a basis for future predictions of the permeability of fibrous reinforcements.     

EXISTENCE AND ATTRACTIVITY OF ALMOST PERIODIC SOLUTION FOR SHUNTING INHIBITORY CELLULAR NEURAL NETWORKS WITH CONTINUOUSLY DISTRIBUTED DELAYS AND VARIABLE COEFFICIENTS

In this paper, a class of two-dimensional shunting inhibitory cellular neural networks with distributed delays and variable coefficients system is studied. By using the Schauder's fixed point theorem and Lyapunov function, we obtain some sufficient conditions about the existence and attractivity of almost periodic solutions to the above system, and all its solutions converge to such almost periodic solution. An example is given to illustrate that the conditions of our results are feasible.     

基于模拟退火的文件优化分配

在局域网环境下，或是在分布布系统环境下，文件优化分配是一个非常重要的问题，本文提出一种基于模拟退火的文件优化分配方法，利用该方法能较好地解决工程中的文件优化分配，该方法也可以推广应用到其它类型的网络系统中。     

基于径向基函数网络的虚拟物流企业伙伴选择方法研究

在对虚拟物流企业系统研究的基础上 ,针对虚拟物流企业伙伴选择问题的特点 ,给出了一个面对虚拟物流企业伙伴选择问题的较为全面的选择过程框架 .并提出了一个基于径向基函数网络算法的虚拟物流企业伙伴选择模型 ,实例仿真说明了该算法和模型的有效性 .     

A Subtree-Partitioning Algorithm for Inducing Parallelism in Network Simplex Dual Updates

This paper reports on thedevelopment of a very efficient method for partitioning the networksimplex basis subtree in which dual values must be updated duringa pivot. The partitioning procedure may be concurrently executedby multiple processes. The resulting rapid decomposition of thesubtree allows an arbitrary number of processes to be utilized inthe actual dual update. This approach alleviates a primary limitationof the most efficient parallel network simplex implementationpublished to date. The new code performs at least as well as theprevious implementation on medium-scale problems and reduces averagesolution time by over 34% on large-scale problems.     

Finite attraction for nonlinear uncertain mechanical systems

This paper deals with the robust control problem for uncertainnonlinear mechanical systems with ideal and holonomic constraints.A robust control law having the form of state feedback is obtained.The control law consists of two parts: the first one is a nominalcontrol law which affects the behaviour of the nominal dynamicsfor the considered mechanical system; the second one is a learningcontrol law which reduces the influence of uncertainties onthe stability of the mechanical system. Under the action ofthe robust control law, the closed-loop system for the uncertainmechanical system reaches a region specified a short time inadvance.     

Robust stability of a class of neural networks with time delays

A stability analysis of the equilibrium position for a given class of Hopfield neural networks with time delays is presented. The robustness of the equilibrium stability with respect to variations in the time delays, system parameters, and interconnection matrix is analyzed. Three approaches are presented which account in various ways for stability of the equilibrium with respect to these perturbations.Work conducted while visiting the University of Bremen, supported by the Deutscheforschungsgemeinschaft.     

Structure-reversibility and departure functions of queueing networks with batch movements and state dependent routing

We consider characterizations of departure functions in Markovian queueing networks with batch movements and state-dependent routing in discrete-time and in continuous-time. For this purpose, the notion of structure-reversibility is introduced, which means that the time-reversed dynamics of a queueing network corresponds with the same type of queueing network. The notion is useful to derive a traffic equation. We also introduce a multi-source model, which means that there are different types of outside sources, to capture a wider range of applications. Characterizations of the departure functions are obtained for any routing mechanism of customers satisfying a recurrent condition. These results give a unified view to queueing network models with linear traffic equations. Furthermore, they enable us to consider new examples as well as show limited usages of this kind of queueing networks. This revised version was published online in June 2006 with corrections to the Cover Date.