一种新的民用飞机修理级别优化模型

修理级别分析(LORA)是飞机保障性分析的一个重要内容,是在设计阶段评估维修对飞机成本影响的一种系统分析方法。现有的LORA经济性分析模型对于系统寿命周期的维修费用考虑不够,且不适合民用飞机维修特点,基于此,提出了一个适合民用飞机维修规划三层三级的LORA经济性分析的数学模型,在此基础上,以最小化维修成本为目标建立了维修级别优化模型,并利用免疫粒子群算法超强的多变量、非线性及全局优化能力,对所提出的民机维修级别优化模型进行了求解。最后通过算例的比较分析,说明了本文LORA经济性分析模型的合理性以及模型求解算法的高效性。     

传声器校准用活塞发生器的数学模型

通过理论分析推导出活塞发生器的数学模型,并且通过实验验证了模型和实测声压的一致性。     

一种目标识别中基于关键点的仿射不变矩

针对常用仿射不变矩对轮廓边缘信息敏感的特点,提出了一种基于关键点的局部仿射不变矩,以分割出来的物体灰度图像为基础,首先计算出物体的质心,然后以质心为扩展点向周围引伸出多条射线,寻找每条射线方向上的最近灰度极值点,将所有灰度极值点当作关键点集合,并按照文中提出的计算仿射不变矩的方法提取出多阶不变矩,将其当作神经网络的输入向量,放入已经训练好的神经网络来达到识别物体的目的,将此方法应用在了飞机图像的识别上,实验结果证明此方法在物体轮廓分割不完整和有噪点污染的情况下能保持很好的稳健性,简单有效并具有较广的应用范围.     

卫星发射试验风险控制

文章回顾了为保证安全执行发射任务所进行的“风云3号”卫星发射试验风险控制工作:通过分解卫星发射试验流程中的主要事件并结合FMEA和FTA分析结果,辨识并确认了卫星在执行进场发射试验时可能导致任务失败的安全性关键项目,在此基础上从设计、生产、试验、工艺保证、质量控制等方面制定并落实了一系列安全性保障措施,从而确保卫星安全发射。     

基于常规多普勒滤波器组结构的合成宽带距离像性能分析

为了消除散布效应对宽带信号的不利影响,可以利用多个窄带信号合成宽带信号;同时,通过在子带内处理相干脉冲串,可进行杂波抑制、目标速度估计和区分不同速度的目标。然而,在子带内分别使用常规多普勒滤波器组,会引进多普勒散布效应所造成的输出失配误差,从而造成合成距离像的失真。分析了多普勒散布效应与子带常规多普勒滤波器组输出失配之间的关系,推导了运动目标通过滤波器组后所成高分辨距离像的距离走动公式,给出了适合子带常规多普勒滤波器组的目标速度临界值。仿真实验验证了以上结论。     

控制器故障下的软件定义机载网络数据平面迁移策略

软件定义网络（SDN）为突破机载网络功能僵化所造成的瓶颈提供了新的契机。针对软件定义机载网络控制域内控制节点故障问题,提出一种基于时延和负载均衡的传输节点迁移策略,将故障控制节点所管控数据平面中的传输节点迁移至其他正常控制节点,恢复网络的正常管控。以迁移时延和负载均衡率为目标,建立迁移优化模型,提出基于时延和负载均衡的传输节点迁移算法,指导数据平面中的迁移动作,最终对两个性能目标进行综合权衡。实验结果表明,与距离就近迁移（DCM）策略和分布式逐跳（DHA）策略相比,所提策略在迁移时延和负载均衡率2个方面进行了合适的折中,避免出现迁移时延过大或控制节点负载失衡的情况,有效解决机载网络故障恢复问题。     

电子战无人直升机抵近支援干扰仿真研究

随着雷达对抗技术的飞速发展,电子战无人直升机超低空突防实施抵近支援干扰呈现出特有的优越性。建立了无人直升机干扰雷达数学模型和抵近支援干扰时压制区计算模型,通过实例仿真分析,定量直观地得出电子战无人直升机抵近雷达时干扰效果和作战效能有相应提高的研究结论。     

Instrument study of the Lunar Dust eXplorer (LDX) for a lunar lander mission

One of the highest-priority issues for a future human or robotic lunar exploration is the lunar dust. This problem should be studied in depth in order to develop an environment model for a future lunar exploration. A future ESA lunar lander mission requires the measurement of dust transport phenomena above the lunar surface. Here, we describe an instrument design concept to measure slow and fast moving charged lunar dust which is based on the principle of charge induction. LDX has a low mass and measures the speed and trajectory of individual dust particles with sizes below one micrometer. Furthermore, LDX has an impact ionization target to monitor the interplanetary dust background. The sensor consists of three planes of segmented grid electrodes and each electrode is connected to an individual charge sensitive amplifier. Numerical signals were computed using the Coulomb software package. The LDX sensitive area is approximately 400 cm². Our simulations reveal trajectory uncertainties of better than 2° with an absolute position accuracy of better than 2 mm.     

基于遗传算法的测站资源优化分配方法研究

针对测站级测控资源分配中存在的测控任务冲突问题,根据资源分配的一般原则及最优准则,建立资源分配优化模型,利用遗传算法给出该模型的求解方法及步骤。通过实例分析,证明该方法用于解决资源分配问题行之有效。     

Simulation-based evaluation of a cold atom interferometry gradiometer concept for gravity field recovery

The prospects of future satellite gravimetry missions to sustain a continuous and improved observation of the gravitational field have stimulated studies of new concepts of space inertial sensors with potentially improved precision and stability. This is in particular the case for cold-atom interferometry (CAI) gradiometry which is the object of this paper. The performance of a specific CAI gradiometer design is studied here in terms of quality of the recovered gravity field through a closed-loop numerical simulation of the measurement and processing workflow. First we show that mapping the time-variable field on a monthly basis would require a noise level below $5\text{mE}/\sqrt{\text{Hz}}$. The mission scenarios are therefore focused on the static field, like GOCE. Second, the stringent requirement on the angular velocity of a one-arm gradiometer, which must not exceed ${10}^{-6}$?rad/s, leads to two possible modes of operation of the CAI gradiometer: the nadir and the quasi-inertial mode. In the nadir mode, which corresponds to the usual Earth-pointing satellite attitude, only the gradient $V_{\mathit{yy}}$, along the cross-track direction, is measured. In the quasi-inertial mode, the satellite attitude is approximately constant in the inertial reference frame and the 3 diagonal gradients $V_{\mathit{xx}}\text{,}{V}_{\mathit{yy}}$ and $V_{\mathit{zz}}$ are measured. Both modes are successively simulated for a 239?km altitude orbit and the error on the recovered gravity models eventually compared to GOCE solutions. We conclude that for the specific CAI gradiometer design assumed in this paper, only the quasi-inertial mode scenario would be able to significantly outperform GOCE results at the cost of technically challenging requirements on the orbit and attitude control.