Error and attack tolerance of evolving networks with local preferential attachment

Networks generated by local-world evolving network model display a transition from exponential network to power-law network with respect to connectivity distribution. We investigate statistical properties of the evolving networks and the responses of these networks under random errors and intentional attacks. It has been found that local world size M has great effect on the network's heterogeneity, thus leading to transitional behaviors in network's robustness against errors and attacks. Numerical results show that networks constructed with local preferential attachment mechanism can maintain the robustness of scale-free networks under random errors and concurrently improve reliance against targeted attacks on highly connected nodes.     

A novel hard decision decoding scheme based on genetic algorithm and neural network

A novel hard decision decoding scheme based on a hybrid intelligent algorithm combining genetic algorithm and neural network, named as genetic neural-network decoding (GND), is proposed. GND offsets the reliability loss caused by channel transmission error and hard decision quantization by making full use of the genetic algorithm's optimization capacity and neural network's pattern classification function to optimize the hard decision outputs of received matched filter and restore a more likelihood codeword as the input of hard decision decoder. As can be seen from the theoretical analysis and computer simulation, GND scheme is close to the traditional soft decision decoding in error correction performance, while its complexity, compared with the traditional soft decision decoding, is greatly reduced because its decoding process does not need to use the channel statistical information.     

A Single-Star Multi-Ring Structure of Self-Healing Wavelength Division Multiplexing Optical Access Network

Abstract

A simple single-star multi-ring structure of multi-wavelength optical access network for enlarging network scale and protecting distribution fiber is presented. In this scheme, the ring structure enables optical network units to switch automatically to a protection link when the distribution fiber fails, and a star structure is adopted to weaken the sphere of the distribution fiber failure's influence. The simple combination of these two structures greatly increases the number of subscribers and minimizes costs of network. The network's survivability can also be better than that of a star network or ring network. In addition, it can easily be upgraded from an existing star network, so this article also proposes a method to update the existing star network. The simulation result shows the structure is of good performance.     

基于遗传神经网络的耙吸挖泥船泥泵转速预测

耙吸挖泥船泥泵管线模型是一个复杂的、非线性的动态模型,影响模型准确性的参数较多。为了根据当前施工条件和流量的优化值准确地预测转速,为施工人员提供参考,提高疏浚效率,采用了遗传算法改进的BP神经网络对泥泵转速进行预测。首先,遗传算法对BP神经网络的初始权值和阈值进行优化。然后,BP神经网络根据优化值对网络进行训练并对转速进行预测。为了验证该方法的有效性,将遗传BP神经网络的预测输出和实测泥泵转速进行对比。仿真结果表明：遗传BP神经网络具有很强的非线性拟合能力和全局搜索能力,能够准确地预测泥泵转速。该预测输出可为施工人员提供参考,以便改变泥泵转速,提高疏浚效率。     

基于热图像的种蛋气室变化监测算法

种蛋气室的大小是监测种蛋孵化过程的重要指标之一。根据种蛋的热力学结构,种蛋在孵化过程中,包裹气室部分蛋壳会与其他部分蛋壳产生温差,从而可通过热红外图像进行观察。针对在种蛋孵化过程中,人工照蛋检测气室效率低的问题,探索设计了一种基于热图像的种蛋气室变化俯视监测算法。监测种蛋气室热图像的算法主要包括种蛋目标检测,种蛋图像分割和种蛋气室面积计算3个部分,其中种蛋的目标检测采用Faster-RCNN算法实现;种蛋图像分割采用BP神经网络算法实现;种蛋气室面积是在种蛋图像分割的基础上进行计算。使用孵化5天及以上的种蛋作为研究对象,并拍取种蛋的热图像进行试验。试验结果表明：种蛋热图像的目标检测的平均精度（mAP）为99.85%,拥有较好的检测效果。使用BP网络对种蛋进行图像分割。BP神经网络经过调参后,其网络最佳的结构为三层隐藏层,每个隐藏层拥有1 000个神经元,最优初始学习率为0.000 1,最优最大迭代次数为500。以F1-measure作为分割效果的评价指标,BP神经网络的图像分割总体结果为87.02%,Otsu算法的总体结果为65.25%。其中只有一个蛋的情况下,BP神经网络的分割结果为87.17%,Otsu算法的结果为68.86%。存在其他种蛋的干扰条件下,BP神经网络的分割结果为86.94%,Otsu算法的结果为61.64%,BP神经网络的分割效果优于Otsu分割算法,BP神经网络拥有更强的抗干扰能力。最后提取了孵化5～19 d种蛋的气室变化,通过观察种蛋气室大小曲线来监测种蛋的孵化情况,可看出随着天数的增加,气室有着明显变大的趋势。人工测量法与热红外测量法比较结果说明两者相关性为0.934 3,拥有较好的相关性。基于热图像的种蛋气室变化监测算法可在实际生产中实现种蛋的识别与气室大小的快速监测,为实现监测种蛋孵化的自动化提供了技术参考。     

Sediment transport mechanics upstream of an orifice

Abstract  

Mechanics of sediment transport, maximum scour depth, and scoured volume upstream of a circular orifice was explored under constant head condition. A high-speed camera was used to investigate the sediment transport mechanism. It was found that sediment transport had two distinct phases. In the first phase, the sediment transport was due to excess bed shear stress and occurred within an area that was elliptical in shape, and the mobilized sediment was transported out of the orifice. In the second stage, vortices appear in the area that was scoured during the first phase. The sediment particles were lifted up into the main flow by these vortices, which were then transported out of the orifice by the main flow. The scour depth and scoured volume were found to be dependent on sediment size and head over the orifice.     

硅基支撑系统的光刻工艺

为提高硅基支持系统制备中光刻过程的线宽精度,用正交试验分析了前烘、曝光、显影主要步骤中一些主要参数对制备结果的影响,得到了它们的影响程度的规律以及较优的参数组合。在此基础上设计和训练了合适的前向误差反向传播神经网络,对主要工艺参数进行了进一步的分析、预测和优选,并用实验加以验证。最终得到在胶厚约1.55 m时,前烘温度100 ℃,时间90 s; 曝光时间5 s; 显影温度15 ℃,时间90 s时,光刻后图形的线宽偏差最小,达到了0.3 m以下。     

Sensitivity of the resilience of congested random networks to rolloff and offset in truncated power-law degree distributions

Random networks were generated with the random configuration model with prescribed truncated power-law degree distributions, parameterized by an exponent, an offset, and an exponential rolloff. As a model of an attack, each network had exactly one of its highest degree nodes removed, with the result that in some cases, one or more remaining nodes became congested with the reassignment of the load. The congested nodes were then removed, and the “cascade failure” process continued until all nodes were uncongested. The ratio of the number of nodes of the largest remaining cluster to the number of nodes in the original network was taken to be a measure of the network's resiliency to highest-degree node removal. We found that the resiliency is sensitive to both rolloff and offset (but not to cutoff) in the degree distribution, and that rolloff tends to decrease resiliency while offset tends to increase it.     

基于图像处理和模式识别的网络图像监控系统

随着科学技术和生活水平不断的提高,人们的财产和人身安全问题已经不仅仅来自于传统的经验,而是逐渐的转变到虚拟的网络当中;为此,提高人们的隐私安全和财产安全需要加强网络监控,其中图像监控是较为主要的方面;加强图像监控的可靠性和实时性,是实现社会稳步发展和人民网络隐私安全的首要目标;应用计算机网络技术和现代化通讯技术相结合,综合图像处理技术和模式识别技术的应用,设计应用于网络图像监控系统是当前社会发展的安全技术保障之一;文章通过对图像监控进行简析,阐述图像处理技术和模式识别技术的相关方面,探讨研究网络图像监控中图像识别和处理方面的技术;通过对图像处理中的色彩均化和模糊识别实验,得出其对于网络中传播的图片辨识度和提取信息能力具有很高的发展。     

Scour monitoring system of subsea pipeline using distributed Brillouin optical sensors based on active thermometry

A scour monitoring system of subsea pipeline is proposed using distributed Brillouin optical sensors based on active thermometry. The system consists in a thermal cable running parallel to the pipeline, which acquires frequency shift of optical sensors during heating and cooling, directly indicating temperature change. The free spans can be detected through the different behaviors of heat transfer between in-water and in-sediment scenarios. Three features were extracted from temperature time histories including magnitude, spatial continuity and temporal stability. Several experimental tests were conducted using the proposed system. The results substantiate the monitoring technique.     

复杂网络系统拓扑连接优化控制方法

为了增加实际网络系统连接增益、减少网络连接成本, 提出了一种基于网络效率和平均连接度的网络拓扑连接优化控制方法, 该方法利用网络效率来表征网络连接收益、用网络平均连接度来表征网络连接成本, 并提出了其计算优化算法, 该算法的时间复杂性为O(Mpn²). 实验分析表明, 可以采取一定的方式对实际复杂网络拓扑连接进行优化控制, 小世界和无标度网络均存在一个最佳的网络平均度值能够使网络连接增益达到最大. 关键词： 复杂网络 拓扑连接 优化控制 连接增益     

基于LM-BP神经网络的气阀故障诊断方法

提出一种基于LM（Levenberg-Marquardt）算法优化的 BP (Back Propagation)神经网络的多级往复式压缩机压缩机气阀故障诊断方法。以6M25-185/314氢氮气压缩机的 6级压差和6级温差作为网络的输入向量,建立可对往复式压缩机一至六级气阀故障进行在线监测及故障诊断的LM-BP神经网络模型。以100组故障数据作为网络训练样本,30组数据作为网络检测样本进行故障诊断,结果表明,LM-BP神经网络相比于变梯度BP神经网络和RBF神经网络诊断更快速稳定且准确率达到96%以上。利用Matlab软件平台建立的LM-BP 神经网络故障诊断模型,模型简单便于在工程实际中应用。     

基于异步延迟采样的光通信性能监测方法

基于异步延迟采样和人工神经网络统计学习提出了一种光通信性能监测方法。通过对高速光信号进行异步延迟采样,获得信号二维幅度直方图,然后提取其中特征参数并对人工神经网络进行训练,最后以人工神经网络的预测输出实现对光信号损伤的监测。构建10 Gb/s非归零码开关键控,40 Gb/s光学双二进制码和归零码差分移相键控光通信仿真系统,并对光信噪比、色散和偏振模色散损伤进行监测。仿真结果表明,所提方法对被监测光信号的速率、码型调制格式透明,可同时准确监测多种并存的传输损伤,损伤参数监测误差小于5%。该方法具有电域处理带宽要求低、采样机制简单的特点,适用于分布式在线光性能监测。     

基于三维荧光谱特征分析的油种鉴别技术的研究

介绍了运用神经网络进行模式识别的基本原理,将主成分分析法和BP神经网络相结合,提出矿物油三维荧光谱鉴别方案,并进行了系统设计,建立了基本的模型框架.选取矿物油三维荧光谱的特征参量,组成原始特征向量,采用主成分分析法进行预处理,而后选取主成分运用BP神经网络实现油种鉴别.该方法减少了输入变量的维数,消除了各输入变量的相关性,同时简化了网络结构,提高了程序运行的速度.通过实例进行了分析,结果证明该方法有效地实现了矿物油三维荧光谱的油种鉴别,同时该系统也可用于其它物质的光谱识别技术领域.     

基于径向基神经网络的航空发动机气路故障诊断

基于径向基神经网络对民用高涵道比航空发动机风扇、增压级、高压压气机、高压涡轮、低压涡轮5大气路部件的效率降低故障进行诊断。采用Gasturb进行故障训练样本和测试样本库的生成,诊断结果显示,采用径向基神经网络进行航空发动机气路故障诊断的计算时间短、精度较高,不仅能定性的定位故障部位,而且在大多数情况下可以定量的给出该部件的性能衰退程度。某些情况下诊断结果与测试样本不尽一致,但都是方程的合理解,这是因为航空发动机的数学模型是一个多解的复杂方程,一个总性能的衰减对应着多组部件性能衰退的组合。随噪声幅值加大,诊断精度变差,同时研究发现诊断精度受噪声影响的敏感系数在不同的噪声幅值水平下是不同的。     

基于灰色关联分析的BP神经网络对混流闭式冷却塔出水温度的预测

通过控制变量法对混流闭式冷却塔进行测试, 采用灰色关联分析法对影响出水温度的因素进行筛选, 将关联度较大的5个因子作为输入参数, 进而建立灰色_BP神经网络预测模型, 对混流闭式冷却塔的出水温度进行预测。操作参数包括进水温度、湿球温度、补水温度、循环水流量和风量, 输出值为出水温度。网络采用三层结构, 隐含层神经元数为4个, 迭代次数为30 000次, 使用不涉及训练阶段的实验数据来验证所建立的模型。结果表明, 灰色_BP神经网络模型比传统BP神经网络模型的预测结果更加准确, 其预测值与实际值的相关系数、平均相对误差、均方根误差, 分别为0.998 9、0.293 4%和0.152 9, 因而可认为灰色_BP神经网络是预测混流闭式冷却塔出水温度的有效工具。     

基于人工神经网络的制冷系统工况模拟

针对人工神经网络技术在制冷空调系统中的仿真应用,本文建立了单回路制冷系统的性能仿真系统。通过实验模拟制冷系统在夏季的负荷变动情况,得到了用于神经网络模型训练的样本数据。对制冷系统进行多种神经网络结构的建模,并进行了神经网络中各种结构参数对模型精度影响的分析。利用训练好的双隐层神经网络模型,研究了空调机组性能的影响因素,包括压缩机频率、室内外温度等。模拟结果表明,机组EER随着压缩机频率增加先增加后减少,随着室内温度升高而增加,随着室外温度升高而减少。结果表明,人工神经网络方法是分析制冷机组性能的一种有效途径。     

大型病患智能远程电子监控系统的设计与实现

现在医疗事业当中,远程医疗作为先进科学技术应用;病患科学治疗监控理念;快速信息化时代体现,正在成为医疗事业当中热门的研究方向。其依靠计算机技术、通讯技术以及信息处理技术为基础,同医疗活动相结合,在当今的医疗事业当中有着广阔的发展前景。本文设计出用于大型病患的智能远程监控系统,系统从生理数据采集、智能终端、IOCP网络通讯器以及数据库和监控软件为设计出发点,利用无线通信网络实现医生或是家属通过监控软件对病患进行实时远程监控和监护。提出系统的设计需求和设计结构,并通过实际检测观察此系统在实际应用中的可靠性以及对各项需求的完成度。结果显示,此系统可以很高程度的完成对与大型病患的远程实时监控和智能化需区。     

An investigation of the changes in the natural frequency of a pile affected by scour

Scour around bridge foundations is one of the leading causes of bridge failure. Up until recently, the monitoring of this phenomenon was primarily based around using underwater instrumentation to monitor the progression of scour holes as they develop around foundation systems. Vibration-based damage detection techniques have been used to detect damage in bridge beams. The application of these vibration based methods to the detection of scour has come to the fore in research in recent years. This paper examines the effect that scour has on the frequency response of a driven pile foundation system, similar to those used to support road and rail bridges. The effect of scour on the vibration characteristics of the pile is examined using laboratory and field testing. It is clear that there is a very clear reduction in the natural frequency of the pile as the severity of scour increases. It is shown that by combining state-of-the-art geotechnical techniques with relatively simple finite element modelling approaches, it is possible to accurately predict the natural frequency of the pile for a given scour depth. Therefore, the paper proposes a method that would allow the estimation of scour depth for a given observed pile frequency.     

The criteria for selecting a method for unfolding neutron spectra based on the information entropy theory

To further expand the application of an artificial neural network in the field of neutron spectrometry, the criteria for choosing between an artificial neural network and the maximum entropy method for the purpose of unfolding neutron spectra was presented. The counts of the Bonner spheres for IAEA neutron spectra were used as a database, and the artificial neural network and the maximum entropy method were used to unfold neutron spectra; the mean squares of the spectra were defined as the differences between the desired and unfolded spectra. After the information entropy of each spectrum was calculated using information entropy theory, the relationship between the mean squares of the spectra and the information entropy was acquired. Useful information from the information entropy guided the selection of unfolding methods. Due to the importance of the information entropy, the method for predicting the information entropy using the Bonner spheres' counts was established. The criteria based on the information entropy theory can be used to choose between the artificial neural network and the maximum entropy method unfolding methods. The application of an artificial neural network to unfold neutron spectra was expanded.