查询词典 neural network

For further study to improve performance, an adaptive control algorithm based on Neural Network is tried in stabilization control. A new method of selecting the initial weights of Neural Network is put forward, which is that the coefficients of discrete equivalent of continuous transfer function in traditional control are used as the initial weights of Neural Network. Adaptive Neural Network control is combined with traditional control by using different method in different segment.

作为进一步提高稳定系统性能的探讨，对基于神经网络的参数自适应调整控制方法在稳定控制中的应用进行了研究，提出了将传统校正方法经离散化后所得到的数字控制算法的系数作为神经网络权值初值的新方法，并采用分段控制，将传统校正方法和神经网络控制方法相结合，取得了较好的控制效果。

In chapter 4, basing theories and methods of scientific visualization, and artificial neural network BP algorithm, we integrate the Visual C++, OpenGL graphics library and Excel VBA technique to develop the program of artificial neural network and to make the BP algorithm visually, this program works can be divided into four parts: Using C language to develop program about BP algorithm; Using Visual C++, develop the GUI Interface, make input parameter visually; Using OpenGL graphic technique to display the training sample point in three dimension; at last using Excel DDE technique display the error graphic tables in Excel system In chapter 5, on the view of engineering application, we establish new method of surface reconstruction basing artificial neural network, develop interface program between module and commercial CAD/CAM system, meantime deeply discuss some key problems, for example, setting up the base plane, using the API technique, cutting and editing surface boundary, and also discuss the more compliant problem: how to intersect surface, at end we finish the work of translation from our surface reconstruction module to commercial CAD/CAM system, then make reverse engineering system basing artificial neural network more useful.

第四章基于科学计算可视化理论，依据人工神经网络BP算法理论模型，综合Visual C++，OpenGL图形库以及Excel VBA等多项软件开发技术，编制了人工神经网络程序，实现了BP算法的可视化映射。具体工作分为四部分：利用C语言实现人工神经网络BP算法；利用VisualC++的GUI技术开发图形用户界面，实现参数设置可视化；利用OpenGL图形技术进行三维映射，显示学习样本及训练样本点；利用微软电子表格DDE动态数据交换技术，在Excel上动态显示学习误差曲线图。第五章从工程应用的角度出发，提出了一种新的基于人工神经网络算法的曲面裁剪重构方法，完成了曲面重建模块与通用CAD/CAM系统的接口设计工作，对其中的若干关键问题进行了深入讨论，例如基平面设定、API技术的应用、边界裁剪等问题，同时，对曲面计算中较为困难的曲面相交问题也进行的专门探讨，最终完成了曲面重建模块向CAD/CAM系统的数据传输工作，使人工神经网络逆向工程系统趋向实用。

It points out the current situation and existing problems of this method. This paper actualizes the integration of Expert System and Neural Network. It excellently utilizes the advantages of Expert System and BP Neural Network, and overcomes some of their disadvantages.Neural network expert system is designed based on neural network high identification ability and theory research.

本文介绍了专家系统应用于故障诊断的原理，详细阐述了专家系统的结构、各部分功能、工作流程等，并指出了该方法在故障诊断中的应用现状和存在的问题；实现了神经网络和专家系统的结合，很好地利用专家系统和BP神经网络的优点并且克服了它们的一些缺点。

The main contents are as follows:Firstly, starting with the general situation of soil erosion and the harms caused by it, the causes leading to the local soil erosion problem are analyzed comprehensively in the paper. And connecting with the measures taking place, sticking points towards the career of soil and water conservation are expatiated upon.Secondly, Back-Propagation Neural Network, One of Artificial Neural Network is used to set up a modal about the connection of the soil erosion modulus and seven factors impacts on it, such as, rainfall, rainfall largest intensity in 30 minutes, runoff coefficient, vegetation cover percent, rate of granule, rate of physical viscidity-clay, the rate of organic matter. Through the comparison with linear regression model, the second regression model, the Chinese Soil Loss Equation, it illustrated that BP Neural Network modal is more accurate than the other three modals in forecasting the mount of soil erosion, and the BP Neural Network will have some applicability in forecasting in soil erosion.

本文以霍山县作为皖西大别山区的典型区域，主要研究了以下内容：(1)从介绍霍山县土壤侵蚀状况以及所造成的危害入手，全面分析了导致当地土壤侵蚀发生的原因，并结合当地采取的水土保持相关措施，阐述了当地水土保持工作的症结所在；(2)结合上土市水土保持试验站多年实测资料和2005年实验资料，应用BP神经网络理论，建立了次降雨土壤侵蚀量与次降雨量、最大30min雨强、径流系数、植被覆盖度等因子之间关系的模型，并通过BP神经网络的预测模型与一次回归模型、二次回归模型、CSLE模型之间的对比分析，说明了建立的BP神经网络模型在土壤侵蚀预测可以取得较回归模型和CSLE模型更高的精度，也说明了BP神经网络理论在土壤侵蚀预报中具有一定的适用性。

The whole PPT-reports of the Forum have put into "Complex Network Science Forum Circle" on Science Network in China (Please see: hettp//www.sciencene.cn/g/ComplexNets ) and also appear on the CCAST-WL Webside (Please see: www.ccast.ac.cn/workshop ). Main issues of the Forum are associated main subject of Network Science with application, including: review and outlook of network science ,spreading model of the H1N1 influenza virus , dynamics of neural network , brain functional networks,dynamics of topological structure, military network science with application, space structure of complex networks, community structure,,stationary e fficiency of coevolutionary networks , significance of community structure,,relationshipe between structure and dynamics of complex network, generalized synchronization and control, consensus of multi-agent systems, evolutionary game and friendship network, s preading of epidemic based on human and animal mobility pattern, complex network engineerings, and so on.

本届论坛文集共收编了32篇 ppt 报告，已经公布在中国科学网"复杂网络论坛圈"上（请看： hettp//www.sciencene.cn/g/ComplexNets 并且也发表在中国高等科学技术中心网站上（ the CCAST-WL Webside ）(请看： www.ccast.ac.cn/workshop )，这些报告内容丰富，各具特色，颇有创新，基本涵盖了近年来我国网络科学领域若干重要课题的研究进展和丰硕成果，课题研究颇有深度，不仅具有现实意义的重要课题（如甲型流感全球大爆发之经验统计及其传播模型），而且有前沿课题：生物神经网络系统的动力学与大脑网络，还有一系列重要课题：网络拓扑结构的动力学识别与估计、军事网络科学、网络科学在工程应用、多个体一致性、网络结构与动力学之间关系、同步-协同-中性稳定性和流型的稳定性、网络系统的协调动力学与控制、复杂网络的空间结构、连续系统和离散系统中的广义同步、疾病免疫算法、网络节点匹配、电力网、供应链网、化工网、通讯网与保密通信，等等。

In this thesis, we firstly introduce several class of discrete-time neural network models and the research progress of the neural networks. By Schauder fixed-point principle we prove the existence of an equilibrium (i.e. a fixed point) of a discrete-time neural network with generalized input-output function and by using the converse theorem of Lyapunov function we study the uniformly asymptotical stability of equilibrium in this discrete-time neural network with variable weight and give some sufficient conditions that guarantee the stability of it.

本文首先介绍了几类离散神经网络模型的由来及其研究概况，利用Schauder不动点原理证明了一类具有广义输入输出函数的离散神经网络模型平衡点的存在性，利用Lyapunov函数逆定理给出了这类离散神经网络模型在时变权值下的一致渐近稳定性的充分条件。

Through above study, the theoretical analysis of the inherent fault tolerance of feed-forward neural networks has been extended from hard-limit activation functions to differentiable activation functions; 2. By using above method, fault tolerance of discrete Hopfield feedback neural networks have been also analyzed. It means that the above method can be applied not only for feed-forward neural networks but also for feedback neural networks; 3. For the feed-forward neural network with the Sigmoid activation function, Chebyshev Inequality method was presented.

这项研究将以前仅限于对硬限幅作用函数前向神经网络固有容错性能的理论分析推广到具有可微作用函数的前向神经网络； 2 针对离散型Hopfield反馈神经网络，利用上面提出的方法，对其同样进行了容错性分析，得出许多有用的结论和计算公式，说明上面提出的方法不仅适用于前向神经网络，同时也适用于反馈神经网络； 3 针对具有Sigmoid作用函数的前向神经网络提出了一种切比雪夫不等式法。

For the data set with noises, a regularization intropolation method is proposed according to regularization theory. The relation between the regularization intropolation method and radial basis function method is analysed and structure of regularization neural networks is proposed. RBF neural network is introduced by mortifying the regularization neural networks. Finally the approximation capacity of RBF neural networks is analysed. 4. A method of selecting the centers of hidden layer neurons of RBF neural networks is proposed.

首先从精确内插问题开始对RBF神经网络进行讨论，然后根据正则化理论提出了在数据集带有噪声的情况下的内插方法，并分析了这种内插方法和径向基函数方法之间的密切联系以及其对应的正则化神经网络结构，其次对正则化神经网络进行了修改，得到正则化神经网络的简化形式—RBF神经网络，最后分析了RBF神经网络的逼近性能。

In this paper chip shapes were recognized by using of radbas neural network. We put forward area ratio feature, Euler number feature, etc. geometry feature of chip shape image and thing of those features as inputting vector of neural network. Adopt radbas neural network and training the network using RLS.

研究了径向基函数神经网络在硬质合金刀具切屑形态图像识别中的应用，提出了面积比、欧拉数、分散度等硬质合金刀具切屑形态图像的几何特征，以上述特征作为神经网络的输入矢量，利用径向基函数网络，采用了递推最小二乘法训练该网络。

The engine monitor requires accurate predictions of rational velocities. A prediction model is constructed on the basis of flat neural network A fast recursive algorithm is used to improve the learning speed by forgetting factor. Accordingly, the present neural network algorithm leads to considerable reduction of effort in network weighting matrix as well as increase of self-learning and adaptability of neural network algorithms to nonlinear and nonstationary systems.

发动机监测器以精确的转速预测为前提，通过构建扁平化神经网络预测模型，并根据遗忘因子思想改进快速递推算法，既减少网络权矩阵的计算量，又充分发挥神经网络算法对非线性和非平稳系统的自学习能力。

Now she was hungry and angry.She began to smoulder.

现在她又饿又气，她开始流露难以抑制的怒火。

You have placed our iniquities before You, Our secret sins in the light of Your presence.

诗90：8 你将我们的罪孽摆在你面前、将我们的隐恶摆在你面光之中。