A Study on Turbo—rotor Multi—fault Diagnosis Based on a Neural Network

The multi-fault phenomena are common in the turbo-rotor system of a liquid rocket engine.As it has many excellent qualities,the neural network might be used to solve the problems of multi-fault diagnasis of a turbo-rotor system.First,the feature expression of a common turbo-rotor fault was studied in order to build up the standard fault pattern and satisfy the need of neural network studying and diagnosing.Then.the turbo-rotor fault identification and diagnosis problems were investigated by using a BP(back-propaga-tion)neural network.According to the BP neural network problems,the parallel BP neural network method of multi-fault diagnosis and classification was presented and investigated.The results indicated that the parallel Bp neural network method could solve the turbo-rotor multi-fault diagnosis problems.     

斜拉索-磁流变阻尼器系统半主动控制的神经网络法

磁流变阻尼器具有良好的可变阻尼特性，能用于结构的半主动控制。针对斜拉桥拉索-磁流变阻尼器系统，本文提出了利用神经控制器进行半主动控制的神经网络法，该神经控制器根据过去时刻的加速度响应和控制信号预测下一时间步的控制信号，并给出了训练方法及开展了数值仿真研究。仿真结果显示，采用神经网络半主动控制能取得很好的减振效果，证明该方法是可行的。     

黑龙江省电子政务建设及思考

黑龙江省电子政务"王"字型广域网络结构采用MPLSVPN技术,解决了网络互联互通、允许内部IP地址重复问题,为推广网上应用系统实现信息共享奠定了良好的基础;反思了黑龙江省电子政务建设的一些做法,为兄弟省电子政务建设提供一些可借鉴的模式。     

有线电视宽带网边界路由系统安全技术研究

提出在当今有线电视宽带网络环境下,从构建安全的边界路由系统出发,研究在边界路由器上采取针对有线电视宽带网内部网络的报文过滤和针对路由系统本身的安全策略,并在某有线电视宽带网络中进行了典型应用,实践证明可以达到事半功倍的网络安全目的.     

Study of the Improvement on Network Management Performance with Mobile Agent

1　IntroductionWiththerapiddevelopmentofInternet,thenetworkisbecominglargerandmorecomplex ,whichcallsformoresophisticatednetworkmanage mentstrategiestobeadoptedinordertoguaranteeadequatelevelsofperformance .Confrontedwiththepresentstatusofnetworkmanagement,manyschol arsmakegreateffortstoinventnewapproachesforthepurposeofimprovingtheperformanceofnet workmanagement.MobileAgenttechnologiesandtheirapplicationtonetworkmanagementaread vancedunderthisbackground .Althoughalotofworkhasstilltobedone ,t…     

Integrated Survivability Strategies of IP／GMPLS／Opticai Multi——layer Network

In last decade,due to that the popularity of the internet, data--central traffic kept growing, some emerging networking requirements have been posed on the today‘s telecommunication networks,especially in the area of network survivability. Obviously, as a key networking problem, network reliability will be more and more important. The integration of different technologies such as ATM, SDH, and WDM in multilayer transport networks raises many questions regarding the coordination of the individual network layers.This problem is referred as multilayer network survivability. The integrated multilayer network survivability is investingated as well as the representation of an interworking strategy between different single layer survivability schemes in IP via generalized multiprotocol label switching over ootical netwnrk     

高炉开炉装料计算软件的设计

高炉开炉装料计算工作繁琐、复杂,手工计算耗费时间长、容易出错,难以得到最佳方案,影响高炉顺利开炉。通过高炉开炉装料计算软件的开发实践,介绍了装料计算模型的建模步骤;讨论了软件的功能需求;阐述了选用三层网络结构的原因,从而描述了装料计算软件的设计方法。软件投入运行后,能够快速、准确地计算、分析开炉数据,取得了满意的效果。     

人工神经网络理论及其应用

简要介绍了人工神经网络的发展过程和基本理论，从神经网络具有自学习功能、联想存储功能和高速寻找优化解的能力三个方面论述了其特点和优越性，详细阐述了在模式识别、信号处理、自动控制、人工智能、自适应的人机接口、优化计算、通信以及其它方面的应用，探讨和分析了其发展前景。     

半合成发动机油基础油调配的仿真研究

采用矿物油、聚α-烯烃、多元醇酯和高压加氢BS光亮油调配半合成发动机油基础油,并测定不同配比下基础油的油膜强度、摩擦系数和粘度指数,通过双隐层BP神经网络对基础油调配进行仿真研究,建立了半合成发动机油基础油调配配方优化模型,经过优化训练后,最终得出综合性能优良的配方组合。按49.9957∶4.3693∶42.4578∶3.1772的配比调配基础油,可以使基础油的油膜强度、摩擦系数和粘度指数分别达到56 kg,0.065和112,经实验后仿真结果与实验结果(56 kg;0.065598;112)极其吻合。     

The effect of annealing on the elastoplastic and viscoelastic responses of isotactic polypropylene

Observations are reported on isotactic polypropylene (i) in a series of tensile tests with a constant strain rate on specimens annealed for 24 h at various temperatures in the range from 110 to 150 °C, (ii) in two series of creep tests in the subyield region of deformations on samples not subjected to thermal treatment and on specimens annealed at 140 °C, and (iii) in a series of tensile relaxation tests on non-annealed specimens. Constitutive equations are derived for the elastoplastic and non-linear viscoelastic responses of semicrystalline polymers. A polymer is treated as an equivalent transient network of macro-molecules bridged by junctions (physical cross-links, entanglements and lamellar blocks). The network is assumed to be highly heterogeneous, and it is thought of as an ensemble of meso-regions with different activation energies for separation of strands from temporary nodes. The elastoplastic behavior is modelled as sliding of junctions in meso-domains with respect to their reference positions driven by macro-deformation. The viscoelastic response is attributed to detachment of active strands from temporary junctions and attachment of dangling chains to the network. Constitutive equations for isothermal deformations with small strains are derived by using the laws of thermodynamics. Adjustable parameters in the stress–strain relations are found by fitting the experimental data.