Elman动态递归神经网络在陀螺仪系统建模中的应用

本文针对Ｅｌｍａｎ 动态递归神经网络的特点,提出了一种基于Ｅｌｍａｎ 动态递归神经网络建立陀螺仪系统模型的方法。文中给出了Ｅｌｍａｎ 网络的网络结构和学习方法,并对建立起的网络模型进行了仿真,仿真结果表明,该方法是可行的。     

随机激励下柔性智能梁结构振动的模糊自适应控制

对柔性智能梁结构提出了一种具有结构和参数学习能力的模糊神经网络控制方法,该法摒弃了常规的以BP算法来优化整个网络参数的作法,利用遗传算法对网络全局性参数进行离线优化,利用BP学习算法对网络局部性参数进行在线调节。以柔性智能悬臂梁为例,实现了对其在随机激励下的振动控制。仿真结果表明,模糊神经网络控制算法对智能结构的振动控制具有一定的鲁棒性和自适应性。     

基于并行卷积神经网络的地磁方向适配性分析

针对地磁方向适配性分析时人工特征提取主观性较强、所取特征难以表达深层的结构性特征的问题,并为了进一步提高方向适配性分析的准确率,提出了一种基于并行卷积神经网络的地磁方向适配性分析方法。首先,从不同角度建立了地磁场在6个代表方向上的适配性分析图;然后,从同一磁场的不同角度出发,利用卷积神经网络自动完成了特征学习,得到了更为全面的方向适配性特征描述;最后,在并行卷积神经网络所得特征的基础上,利用BP网络建立了地磁方向适配性的分析模型。仿真结果证明,该方法可以有效避免人工特征提取和计算等复杂步骤,实现了地磁方向适配性分析的自动化,而且可以获得优于传统网络和单路卷积神经网络的准确率。     

工程结构优化的神经网络方法

本文阐述了神经网络优化计算的基本原理，构造了工程结构优化的神经网络模型。采用模拟退火技术进行模型求解，且巧妙地将退火温度Ｔ的倒数作为Ｌａｇｒａｎｇｅ乘子，以改善增广目标函数的收敛性。实例计算表明，由非线性模拟神经元组成的大规模并行、互连的网络在工程结构的优化设计中是可行且有效的。     

遗传优化小波神经网络在组合导航系统中的应用

实时性是组合导航系统的一个重要指标，而神经网络的优化学习问题是决定网络效率的关键技术。遗传优化小波神经网络不仅继承了小波分析良好的局部性及其神经网络的学习和推广能力，而且具有遗传算法全局寻优的特点，是多层前向神经网络学习的一种理想算法。将它应用于组合导航系统中并进行了仿真，结果表明，该算法能够根据实际情况自适应确定网络结构，实时性好，精度与常规方法相当。     

金属栅格加筋结构的并行协同优化方法

栅格加筋板的主要失效形式为整体屈曲和局部屈曲,据此提出了一种栅格加筋板布局的并行协同优化方法,将设计变量分为整体设计变量与局部设计变量,在两个子空间中并行优化,然后进行系统级协调,反复迭代至收敛。两个典型算例结果表明,本文方法优化流程清晰、算法稳定且结果可靠,可以获得最优解。     

人工神经网络在GPS自主式完整性检测中的应用

基于ＢＰ网络，结合ＧＰＳ接收机自主式完整性检测问题，给出一种用于ＧＰＳ自主式完整性检测的ＢＰ神经网络，通过实验优化了网络参数，经仿真实验表明这种方法是有效可行的。     

基于遗传算法优化神经网络的多源信息融合室内定位方法

为了解决复杂室内环境中单一定位技术误差较大的问题,提出了一种基于遗传算法优化BP神经网络的多源信息融合室内定位方法。首先利用Wi Fi定位结果约束地磁匹配范围进行组合定位,降低误匹配率;再采用遗传算法寻找网络全局最优解对BP神经网络的初始权值和阈值进行优化,提升网络精度并加快收敛;使用优化后的网络对组合定位结果和推算定位结果向真实位置坐标方向训练融合,得到最优定位结果。数据显示,经遗传算法优化后BP神经网络预测均方误差降低了约75%,融合定位精度较单一定位方式定位精度平均提升约47%。结果表明,所提的方法可有效提升定位精度,具有更优的定位性能。     

基于TH神经网络的自适应滤波在船舶姿态测量信号处理中的应用研究

自适应滤波是提高滤波性能的主要方法之一。自适应滤波要求实时跟踪输入信号的变化,实时地计算滤波器的权系数。这就大大增加了运算量,因此很难用基于数字计算的方法给出所希望的实时解。神经网络所具有的高度运算能力为解决这一问题展示了光明前景。中利用根据线性规划神经网络导出的TH网络,就船舶姿态测量信号进行滤波计算,仿真结果表明基于TH神经网络的自适应滤波是有效的。     

基于RBF网络的激光陀螺标度因数数据建模研究

前向神经网络中的径向基函数(RBF)网络是一种局部逼近网络,它用局部逼近的总和达到对训练数据的全局逼近,在理论上可以实现全局最优.该文利用径向基函数神经网络对某一温度段的陀螺标度因数的温度数据进行建模处理,并利用各组数据建立一种两因素RBF网络,这两个输入因素选择为温度以及各个温度值对于所属组初始温度的增量.仿真结果表明,所建立的两因素RBF网络可以精确地拟合各温度下的标度因数温度数据,仿真数据的误差与均方差比用BP网络训练的数据效果要好,在数值上提高了近一个数量级.     

ASYMPTOTIC STABILITIES OF STOCHASTIC FUNCTIONAL DIFFERENTIAL EQUATIONS

Asymptotic characteristic of solution of the stochastic functional differential equation was discussed and sufficient condition was established by multiple Lyapunov functions for locating the limit set of the solution. Moreover, from them many effective criteria on stochastic asymptotic stability, which enable us to construct the Lyapunov functions much more easily in application, were obtained. The results show that the well-known classical theorem on stochastic asymptotic stability is a special case of our more general results. In the end, application in stochastic Hopfield neural networks is given to verify our results.     

Transient multiboiling in a pin fin with temperature dependent thermal conductivity

Transient heat conduction in fins undergoing different kinds of convective processes (film, transition, nucleated boiling and natural convection) as in multiboiling processes take place, is a strongly non-linear problem because of the abrupt changes in the heat transfer coefficient that occur at certain temperatures. Transient equations for the thermal fields and fluxes are solved simultaneously, giving the time constant of the process; the stationary solution is compared with the numerical or experimental values of other authors. Temperature dependencies of the heat transfer coefficient and the thermal conductivity is assumed due to the large interval of temperatures occurring. Network Simulation Method is used for the numerical solution, which gives simultaneously thermal field of temperatures and heat fluxes.     

复模态理论在近海平台中的应用

运用复模态理论对考虑海洋平台结构与流体相互作用而引起的非经典阻尼问题进行了理论分析。水中耦合阻尼矩阵运用复模态解法，可以不必近似对角化。推导了随机波浪荷载作用下结构各种反应的能量谱密度函数及结构反应的统计特性，得到了结构反应的均方差和峰值的表达式。针对某典型平台在随机波浪荷载作用下动力反应用复模态法进行了计算并与运用实模态方法多重叠代方法的结果进行了比较。     

Low-rank solution of an optimal control problem constrained by random Navier-Stokes equations

We develop a low-rank tensor decomposition algorithm for the numerical solution of a distributed optimal control problem constrained by two-dimensional time-dependent Navier-Stokes equations with a stochastic inflow. The goal of optimization is to minimize the flow vorticity. The inflow boundary condition is assumed to be an infinite-dimensional random field, which is parametrized using a finite- (but high-) dimensional Fourier expansion and discretized using the stochastic Galerkin finite element method. This leads to a prohibitively large number of degrees of freedom in the discrete solution. Moreover, the optimality conditions in a time-dependent problem require solving a coupled saddle-point system of nonlinear equations on all time steps at once. For the resulting discrete problem, we approximate the solution by the tensor-train (TT) decomposition and propose a numerically efficient algorithm to solve the optimality equations directly in the TT representation. This algorithm is based on the alternating linear scheme (ALS), but in contrast to the basic ALS method, the new algorithm exploits and preserves the block structure of the optimality equations. We prove that this structure preservation renders the proposed block ALS method well posed, in the sense that each step requires the solution of a nonsingular reduced linear system, which might not be the case for the basic ALS. Finally, we present numerical experiments based on two benchmark problems of simulation of a flow around a von Kármán vortex and a backward step, each of which has uncertain inflow. The experiments demonstrate a significant complexity reduction achieved using the TT representation and the block ALS algorithm. Specifically, we observe that the high-dimensional stochastic time-dependent problem can be solved with the asymptotic complexity of the corresponding deterministic problem.     

基于摄动法的不确定性有限元模型修正方法研究

开展了考虑不确定性的有限元模型修正方法的研究。基于摄动法推导了待修正参数均值和协方差矩阵的迭代格式,其中协方差的迭代格式包括是否考虑试验数据与修正参数之间相关性的两种形式。在理论研究基础上开展数值仿真研究,实现了不确定性有限元模型修正的摄动法,并研究了试验数据样本数量对修正误差的影响。仿真结果表明,该方法适用于解决系统参数与试验数据存在不确定性的模型修正问题,试验样本数量对待修正参数标准差的修正精度影响较大;忽略试验模态参数与待修正参数不确定性之间的相关性,能够避免计算二阶灵敏度矩阵,在保证修正结果准确性的前提下减少计算量。     

Stochastic phase decoupling in dynamical networks

Network models and their theories play a central role in the understanding of complex systems, in particular complex social systems such as societies and organizations. An important problem is to understand how agent attributes become organized within the connectivity structure of a network. The effective matching of agent attributes is important for the expression of functionality by a network. The creation of static networks relative to some control parameter has been extensively studied and gives rise to order-disorder phase transitions. This paper extends this work to dynamic networks. Several models of dynamic networks are created relative to two control parameters and their associated stochastic phase transitions are examined. Under conditions of weak coupling between the control parameters, it is shown that the relevant stochastic phase transitions become decoupled from one another, each qualitatively distinct and dependent on a single (distinct) control parameters.     

Bifurcation and Chaos Analysis of Stochastic Duffing System Under Harmonic Excitations

The Chebyshev polynomial approximation is applied to the dynamic response problem of a stochastic Duffing system with bounded random parameters, subject to harmonic excitations. The stochastic Duffing system is first reduced into an equivalent deterministic non-linear one for substitution. Then basic non-linear phenomena, such as stochastic saddle-node bifurcation, stochastic symmetry-breaking bifurcation, stochastic period-doubling bifurcation, coexistence of different kinds of steady-state stochastic responses, and stochastic chaos, are studied by numerical simulations. The main feature of stochastic chaos is explored. The suggested method provides a new approach to stochastic dynamic response problems of some dissipative stochastic systems with polynomial non-linearity.     

Neural networks for computing in the elastoplastic analysis of structures

A neural network model is proposed and studied for the treatment of elastoplastic analysis problems. These problems are formulated as Q.P.P.s with inequality subsidiary conditions. In order to treat these conditions the Hopfield model is appropriately generalized and a neural model is proposed covering the case of inequalities. Finally, the parameter identification problem is formulated as a supervised learning problem. Numerical applications close the presentation of the theory and the advantages of the neural network approach are illustrated.

---

Sommario Si propone un modello di rete neurale con l'obiettivo di usarlo per la trattazione di problemi di analisi elastoplastica, formulati come problemi di programmazione quadratica con disequazioni sussidiarie. Allo scopo di trattare queste condizioni si generalizza il modello di Hopfield e si propone un modello neurale che copre il caso di disequazioni. Inoltre il problema di identificazione parametrica viene formulato come un problema di apprendimento guidato. La presentazione della teoria è seguita da esempi di applicazioni numeriche e dalla illustrazione dei vantaggi dell'uso delle reti neurali.

     

Stochastic modeling and identification of an uncertain computational dynamical model with random fields properties and model uncertainties

This paper is devoted to the construction and to the identification of a probabilistic model of random fields in the presence of modeling errors, in high stochastic dimension and presented in the context of computational structural dynamics. Due to the high stochastic dimension of the random quantities which have to be identified using statistical inverse methods (challenging problem), a complete methodology is proposed and validated. The parametric–nonparametric (generalized) probabilistic approach of uncertainties is used to perform the prior stochastic models: (1) system-parameters uncertainties induced by the variabilities of the material properties are described by random fields for which their statistical reductions are still in high stochastic dimension and (2) model uncertainties induced by the modeling errors are taken into account with the nonparametric probabilistic approach in high stochastic dimension. For these two sources of uncertainties, the methodology consists in introducing prior stochastic models described with a small number of parameters which are simultaneously identified using the maximum likelihood method and experimental responses. The steps of the methodology are explained and illustrated through an application.     

基于单源模糊数的模糊随机动态有限元方程的解法

本文提出一种模糊随机动态有限元方程的解法,指出利用单源模糊数和它的运算法则,可以把一个不含阻尼项的模糊随机动态有限元平衡方程转化为两类不同集合下的方程组,一种是模糊数方程,另一种是普通的动态有限元平衡方程。前者可用模糊数运算法则求解。通常这类方程的表达式非常简单,故很容易求解,后者可利用现有的求解随机动态随机有限元平衡方程的方法计算,这时求解该方程的计算量几乎等同于求解相应的普通随机动态有限元平衡方程的计算量。最后的算例表明,本文提出的方法与通常所用的γ截集法计算结果基本相同,而且所用的计算量远远小于用γ截集法所用的计算量。