净油储罐运行状态识别与故障诊断技术研究

为了实时、准确地判断净油储罐运行状态，本文提出了一种基于聚类分析和专家系统相结合的状态模式识别方法，有效地解决了传统判别方法存在的误判问题，并且利用离散Hopfield网络对联合站所有净油储罐的运行过程故障状态进行了正确识别，确保了原油集输系统的稳定、安全、高效运行，提高了整个原油集输系统的工作效率。实践表明，本文提出的状态识别与故障诊断方法具有很好实用价值。     

基于Hopfield网络的渐进型漂移自相关过程质量控制方法

为提高自相关过程的统计过程控制方法的灵敏度与可靠性,提出利用Hopfield网络来检测自相关过程的均值渐进型漂移.首先将质量特性观测值分解为原形与背景噪声,通过动态编码将原形存储于网络;而后采用"相对增加"和"大于均值"原则对观测值编码,再利用网络的联想学习功能滤去背景噪声,提取原形,并判断均值是否发生渐进型漂移.研究表明,所提方法适用于具有不同参数的自相关过程,既无需过程统计模型,也无需大量的历史样本进行权值训练,具有较高的灵敏度与可靠性.     

Convergence of higher-order two-state neural networks with modified updating

The Hopfield network is a standard tool for maximizing aquadratic objective function over the discrete set {− 1,1} ⁿ . It is well-known that if a Hopfield network is operated in anasynchronous mode, then the state vector of the network converges to a local maximum of the objective function; if the network is operated in asynchronous mode, then the state vector either converges to a local maximum, or else goes into a limit cycle of length two. In this paper, we examine the behaviour ofhigher-order neural networks, that is, networks used for maximizing objective functions that are not necessarily quadratic. It is shown that one can assume, without loss of generality, that the objective function to be maximized ismultilinear. Three methods are given for updating the state vector of the neural network, called the asynchronous, the best neighbour and the gradient rules, respectively. For Hopfield networks with a quadratic objective function, the asynchronous rule proposed here reduces to the standard asynchronous updating, while the gradient rule reduces to synchronous updating; the best neighbour rule does not appear to have been considered previously. It is shown that both the asynchronous updating rule and the best neighbour rule converge to a local maximum of the objective function within a finite number of time steps. Moreover, under certain conditions, under the best neighbour rule, each global maximum has a nonzero radius of direct attraction; in general, this may not be true of the asynchronous rule. However, the behaviour of the gradient updating rule is not well-understood. For this purpose, amodified gradient updating rule is presented, that incorporates bothtemporal as well as spatial correlations among the neurons. For the modified updating rule, it is shown that, after a finite number of time steps, the network state vector goes into a limit cycle of lengthm, wherem is the degree of the objective function. Ifm = 2, i.e., for quadratic objective functions, the modified updating rule reduces to the synchronous updating rule for Hopfield networks. Hence the results presented here are “true” generalizations of previously known results.     

NAROAS: a neural network-based advanced operator support system for the assessment of systems reliability

We have developed and implemented a computerized reliability monitoring system for nuclear power plant applications, based on a neural network. The developed computer program is a new tool related to operator decision support systems, in case of component failures, for the determination of test and maintenance policies during normal operation or to follow an incident sequence in a nuclear power plant. The NAROAS (Neural Network Advanced Reliability Advisory System) computer system has been developed as a modularized integrated system in a C++ Builder environment, using a Hopfield neural network instead of fault trees, to follow and control the different system configurations, for interventions as quickly as possible at the plant. The observed results are comparable and similar to those of other computer system results. As shown, the application of this neural network contributes to the state of the art of risk monitoring systems by turning it easier to perform online reliability calculations in the context of probabilistic safety assessments of nuclear power plants.     

基于神经网络的图象序列特征点匹配

利用神经网络优化技术解决图象序列的特征点匹配问题，将特征点匹配归结为一个带约束的优化问题，并用2D Hopfield网络实现，在Hopfield网络的能量函数的设计中，综合考虑了特征点的预测结果、特征点的遮挡等情况，从而克服了现有的多数方法所存在的误匹配现象，对于特征点的跟踪，头3帧图象的正确匹配是十分关键的。本文提出了一种3D Hopfield网络用以解决头3帧图象的特征点匹配，并提出了一个运动平滑性的代价函数用以构造3D Hopfield网络的能量函数，实际图象序列的实验结果证明了本方法的有效性。     

神经网络A／D转换器电路的两种改进方法

本文对Ｈｏｐｆｉｅｌｄ神经网络Ａ／Ｄ转换器电路进行了分析。对照转换条件及电路的稳态方程，讨论了输出状态与参考输入必须满足的条件。在此基础上提出了两种改进Ａ／Ｄ转换电路的途径：（１）对原电路选择适宜的参考输入；（２）根据Ｈｏｐｆｉｅｌｄ线性规划网络提出的一种新电路。文中还给出了ＰＳＰＩＣＥ模拟计算，所得结果与理论分析吻合。     

IPA模型对Hopfield模型神经网络的改进

考虑Ｈｏｐｆｉｅｌｄ模型中假态之间的布尔逻辑关系，提出利用ＩＰＡ模型对神经网络中的假态进行抑制，并和原网络的性能进行了比较，结果表明：改进后的网络性能比Ｈｏｐｆｉｅｌｄ模型和ＩＰＡ模型均有较大提高。     

近线性神经和元非线性神经元

人工神经网络(ANN)中普遍采用的神经元模型具有单调有界的I／O特性,该文称其为近线性神经元(ALN,approximate－linear neuron),ALN既简化了网络的复杂性,也是限制ANN能力的一个瓶径.通过比较Hopfield神经网络的稳定性和Conway生命游戏的复杂行为,该文提出非线性神经元(NLN,non－linear neuron)的概念,其I／O特性是完全非线性的.ALN可以进一步提高ANN的非线性,从而增强它的功能,余数制神经网络是一个例子.     

基于一类T-模的模糊Hopfield网络的稳定性分析

本文建立了基于一类T-模的一种动态模糊神经网络-(∨,T)Hopfield网.文章首先证明了系统本身的稳定性及系统平衡态的Lyapunov稳定性,然后建立了吸引子的一个非平凡吸引域,使系统具有良好的容错性.最后,通过例子验证了得到的结论.     

用多层Hopfield神经网络构造分组密码

本文引入了多层Hopfield神经网络,用它的输入、输出关系构造分组加密变换,并分析了所得到的分组密码的安全性。