高压杆塔受迫振动数据采集系统设计

输电线路实时在线监测对于保证电网的安全稳定运行具有重大的意义,针对高压输电杆塔可能受到的外力撞击振动的实际情况,设计了以C8051F060为核心主控单元的高速数据采集系统。系统主要由信号调理、采集与分析三个部分构成。实验测试表明,系统运行稳定、可靠,采集高压输电杆塔振动数据送Matlab工具显示分析,能区别输电杆塔可能受到的外力破坏类型及撞击轻重,实时监测其振动状况,从而实现对杆塔运行状况的预警和评估,保障输电设备安全运行。     

一种面数据挖掘的雷达辐射源参数分布规律分析方法

通过分析电子侦察中数据挖掘的特点,对雷达辐射源的参数测量误差分布特点进行分析.在传统直方图统计算法的基础上,提出一种面向雷达辐射源特征分布的优化校正算法,通过理论分析和仿真手段验证了校正算法对提升雷达特征参数分布规律分析性能的改善.     

空心微珠基红外低发射率材料研究

利用化学镀镍工艺制备空心微珠基红外低发射率材料,对其表面形貌,并重点对其吸收率、发射率和热辐射能等红外性能进行表征.SEM表征表明:材料表面包覆完整、毛刺较少.研究表明:材料在8～14 μm波段平均红外吸收率从34.2%降至13.9%;红外发射率从0.85降至0.65;红外热辐射能较包覆前明显降低,且低于常用金属填料,在低发射率红外隐身涂料研制领域具有广泛应用前景.     

微波反射光电导衰减法测量InGaAs 吸收层的均匀性

外延材料的均匀性对制备大型线列和二维面阵焦平面阵列(FPA)探测器非常重要.用微波反射光电导衰减法(μ-PCD)分别在300 K和85 K温度下测量了用分子束外延技术生长的p-InP/n-InGaAs/n-InP双异质结中掺杂InGaAs吸收层的载流子寿命分布图,并详细论述了这种测试技术的理论基础.在300 K和85 K时,平均寿命分别为168.2 ns和149.4 ns,结果与ZnS/n-InGaAs/n-InP的基本一致.寿命变温曲线表明,中等掺杂InGaAs载流子寿命在低温下变化较小.μ-PCD法可以非接触无损伤快速测量p-InP/n-InGaAs/n-InP双异质结中InGaAs的寿命图,对于表征InGaAs吸收层的均匀性有潜在的应用,这对研制均匀性良好的InGaAs焦平面探测器非常重要.     

基于多信号流图的飞控系统实时故障诊断

针对飞控系统构造复杂、解析建模困难的特点,给出一种基于多信号流图模型进行实时故障诊断的方法。该方法首先建立系统的多信号流图模型,然后根据模型得出相关矩阵,在相关矩阵和各实时测试结果基础上按照实时故障诊断基本算法进行诊断推理,判定系统各模块所处的状态。对某型飞机飞控系统的建模以及模拟诊断结果表明,该方法能隔离多故障,同时降低虚警率,是一种十分有效的实时故障诊断方法。     

用于音频过采样D/A转换器的∑-△调制器设计研究

文章针对采样频率为44.1kHz的16位数字音频信号，采用CookBook方法，研究设计了用于过采样率为64倍的音频数模转换器的五阶3比特输出sigma-deha（∑-△）调制器。该调制器通带内信噪比（SNR）的matlab仿真实验结果达到了120dB以上，能够很好的抑制通带内噪声。该调制器设计结构采用前反馈和负反馈分支的∑△型．大大降低了电路的复杂性，使硬件实现十分方便，具有重要的应用价值。     

基于图卷积的视网膜血管轮廓及高不确定度区域细化框架

针对传统卷积神经网络(convolutional neural network, CNN)受感受野大小的限制,无法直接有效地获取空间结构及全局语义等关键信息,导致宽血管边界及毛细血管区域特征提取困难,造成视网膜血管分割表现不佳的问题,提出一种基于图卷积的视网膜血管分割细化框架。该框架通过轮廓提取及不确定分析方法,选取CNN粗分割结果中潜在的误分割区域,并结合其提取的特征信息构造出合适的图数据,送入残差图卷积网络(residual graph convolutional network, Res-GCN)二次分类,得到视网膜血管细化分割结果。该框架可以作为一个即插即用模块接入任意视网膜血管分割网络的末端,具有高移植性和易用性的特点。实验分别选用U型网络(U-neural network, U-Net)及其代表性改进网络DenseU-Net和AttU-Net作为基准网络,在DRIVE、STARE和CHASEDB1数据集上进行测试,本文框架的Sp分别为98.28%、99.10%和99.04%,Pr分别为87.97%、88.87%和90.25%,证明其具有提升基准网络分割效果的细化能力。     

A Fullerene Seeded Strategy for Facile Construction of Nitrogen-Doped Carbon Nano-Onions as Robust Electrocatalysts

Carbon nano-onions (CNOs) as a novel form of carbon materials hold peculiar structural features but their electrocatalytic applications are largely discouraged by the demanding synthesis conditions (e.g., ≥1500 °C and vacuum). Using C₆₀ fullerene molecules as the sacrificial seeds and melamine as the main feedstock, herein, a novel strategy for the facile construction of CNOs nanoparticles is presented with ultrafine sizes (≈5 nm) at relatively low temperatures (≤900 °C) and atmospheric pressure. During the calcination, in-depth characterizations reveal that C₆₀ can retain the melamine-derived graphitic carbon nitride from complete sublimation at high temperatures (≥700 °C). Owing to the N removal and subsequent pentagon generation, severely deformed graphitic fragments together with the disintegrated C₆₀ molecules merge into larger sized nanosheets with high curvature, eventually leading to the formation of N-doped defect-rich CNOs. Owing to the integration of multiple favorable structural features of pentagons, edges, and N dopants, the CNOs obtained at 900 °C present superior oxygen reduction half-wave potential (0.853 V_RHE) and zinc–air cathode performance to the commercial Pt/C (0.838 V_RHE). Density functional theory calculation further uncovers that the carbon atoms adjacent to the N-doped edged pentagons are turned into the ORR-active sites with O₂ protonation as the rate-determining step.     

移动P2P流媒体中基于有效容量的邻居节点选择机制

For Peer-to-Peer (P2P) streaming services in mobile networks, the selection of appropriate neighbour peers from candidate peers with demanding data is an important approach to improve Quality-of-Service (QoS). This paper proposes a novel Effective Capacity Peer Se-lection (ECPS) scheme based on effective capacity. In the ECPS scheme, the neighbour peer selection problem was modeled using the Multiple Attribute Decision Making (MADM) theory, which considered multiple factors of candidate peers, including Signal to Interference and Noise Ratio (SINR), residen-cy time, power level, security, moving speed, and effective capacity. This model could increase the suitability of ECPS for wireless mobile environments. Then, the Technique for Order Preference by Similarity to Ideal Solu-tion (TOPSIS) was used to solve the MADM problem and identify the preferred neighbour peers. Simulation results show that the ECPS scheme can improve the network throughput, reduce packet delay by about 82%, and al-most double the packet delivery ratio of the mobile P2P streaming service.     

C 波段全固态T / R 组件设计

C波段星载合成孔径雷达(SAR)系统是应用于海洋目标监视的有力手段之一.基于T/R组件的有源相控天线可靠性高,成为现阶段星载SAR雷达的研究热点,其对T/R组件的需求特点包括:高效率、低功耗、重量轻、费用低、高幅度相位稳定性和高可靠性.文中介绍了最新研制的四通道C波段星载T/R组件,采用微波多芯片组件技术,T/R组件通道间距为1/4波长,其输出功率大于22W,噪声系数2.3dB,接收增益大于23dB,功率附加效率大于23％,重量小于200g.