一种基于Deep-GBM的航空发动机中介轴承故障诊断方法

针对航空发动机中介轴承故障信号难于识别的特点，提出了一种深度梯度提升模型（Deep-GBM）对振动信号特征进行逐层学习以提高分类模型的准确率。开展某型航空发动机中介轴承故障模拟实验，并采用经验模式分解(EMD)方法对采集的振动信号进行分解，提取内蕴模式函数(IMF)分量非线性动力学参数样本熵作为原始故障特征。采用Deep-GBM对中介轴承内环故障、内环和滚动体综合故障、正常、滚棒剥落、滚棒划伤五种不同状态进行识别。实验结果表明，所提出的Deep-GBM故障诊断准确率达到87%，相对于传统的机器学习模型准确率最高提升了28%，并具有良好的泛化能力。      

基于YOLO智能网络的红外弱小多目标检测技术

复杂背景下的红外弱小多目标检测是红外目标检测的难点,现有算法的稳健性难以满足实际弹载应用需求。针对弹载环境下的红外场景图像开展了基于典型深度学习网络模型的目标检测应用研究,提出一种基于YOLO(you only look once)网络的智能目标检测方法,通过高维特征学习表征和推理实现红外弱小多目标检测。采用传统的模板匹配算法和YOLO深度学习算法进行识别性能对比分析,验证了YOLO网络在红外弱小多目标检测方面的良好性能。实验结果表明:YOLO算法的检测概率可达92.2%,平均检测精度为0.844,与传统的模板匹配方法相比,YOLO具有明显的优势。     

可靠性设计优化的球空间设计法

可靠性设计优化（RBDO）在航空航天等领域已经被广泛应用，具有显著提高可靠性的效果；然而它包含两个迭代循环-总体优化循环和可靠性分析循环，双循环求解方法效率低下甚至难以承受。本文提出一种可以提高可靠性设计优化求解效率的方法-球空间设计法，该方法可以方便地表达搜索区域，最大可能失效点可以从极值点处精确获取；同时，该方法可以将总体优化以及可靠性分析同步进行。本文所提出的球空间法的有效性将通过2个算例进行测试，测试结果表明该方法不仅精度与同类方法持平，求解效率有明显提升，它将会对提高航空航天装备如卫星、火箭、飞行器等产品的可靠性提供更为有效的方法和技术支持。     

空间目标监视雷达信号频率的优选准则

随着在轨微小卫星及空间碎片数量不断增加，为保证在轨卫星的运行安全，需要对微小目标进行有效跟踪探测及稳定编目。雷达作为近地空间目标监视的主用设备，在对厘米级小目标进行探测跟踪时，其工作频率是决定其能否可靠探测关键因素，高频有利于探测小目标，并可获得相对稳定的RCS，但高频信号的波束小、搜索能力较低，实现难度大，也使得成本过高。为适于实际工程应用，需从效费比出发，对频率进行优选。本文从目标RCS的角度提出一种空间目标监视雷达信号频率的优选准则，给出该准则下周长波长比ka的取值范围。为了更加清楚地说明该优选准则的应用，给出了直径5 cm和10 cm的目标在该准则下信号频率的取值范围，并与国外工作做了比较，证明该准则可行。     

基于G-M制冷机的大尺寸冷屏设计方法

介绍了基于Gifford-Mcmahon (G-M)制冷机的大尺寸冷屏的设计方法。该低温冷屏配合真空环境模拟室共同使用，为遥感卫星的星载遥感器等部件的红外定标试验提供20K的低温冷背景。该冷屏采用5台G-M制冷机进行制冷，外形尺寸达到1m×1m，是目前国内同类型冷屏中尺寸最大的。详细介绍了冷屏在容器内的安装布置、冷缩时的补偿等结构设计方法、冷屏的表面温度分布以及降温时间等计算分析。成功解决了大尺寸冷屏在低温环境下的冷缩补偿难题，试验结果表明:冷屏主要测点温度为15K，温度均匀性误差为±1K，达到并超过设计要求。      

ZSGA?2型重力敏感器关键技术分析与试验验证

与国外主流海洋重力仪相比，国产海洋重力仪的分辨率低一个数量级，这制约了国产海洋重力仪的应用。针对国产海洋重力仪亟需提高分辨技术的需求，根据ZSGA-2型重力敏感器的工作原理，建立了弹性系统的力学模型，并开展了样机的静态性能测试。测试结果表明，经零位漂移和大地固体潮汐修正后的静态精度达到了0.03mGal，样机随重力仪进行楼层高差测量的重复性为0.15mGal。该型重力敏感器的分辨率已经达到了国外同类产品的技术水平，并具备了一定的动态适应性，为该型高分辨率零长弹簧重力敏感器开展高动态环境适应性技术研究奠定了坚实基础。     

Global ionospheric scintillations revealed by GPS radio occultation data with FY3C satellite before midnight during the March 2015 storm

Global observations of S4 amplitude scintillation index by the GPS Occultation Sounder (GNOS) on FengYun-3 C (FY3C) satellite reveal global dynamic patterns of a strong pre-midnight scintillations in F-region of the ionosphere during the St. Patrick’s Day geomagnetic super storm of 17–19 March 2015. The observed strong scintillations mainly occurred in the low latitudes, caused by equatorial plasma bubbles. During the main storm phase (March 17), the scintillations were first triggered in the New Zealand sector near 160°E longitudes, extending beyond 40°S dip latitude. They were also enhanced in the Indian sector, but significantly suppressed in East Asia near 120°E longitude and in Africa around 30°E longitude. During the initial recovery phase (March 18–19), the global scintillations were seldom observed in GNOS data. During the later recovery phase (after March 19), the scintillations recovered to the pre-storm level in Indian, African, and American sectors, but not in East Asian and any of Pacific sectors. These results closely correlate with observations of the density depletion structures by the Communication/Navigation Outage Forecasting System (C/NOFS) satellite, and ground-based instruments. Such consistency indicates reliability of our scintillation sensing approach even in a case-by-case comparison study. The prompt penetration electric field and disturbance dynamo electric field are suggested as the main factors that control the enhancement and inhibition of the scintillations during the storm, respectively.     

MIDACO software performance on interplanetary trajectory benchmarks

A numerical study of the MIDACO optimization software on the well known GTOP benchmark set, published by the European Space Agency (ESA), is presented. The GTOP database provides trajectory models of real-world interplanetary space missions such as Cassini, Messenger or Rosetta. The trajectory models are formulated as constrained nonlinear optimization problems and are known to be difficult to solve.     

Orbit determination and prediction of GEO satellite of BeiDou during repositioning maneuver

In order to establish a continuous GEO satellite orbit during repositioning maneuvers, a suitable maneuver force model has been established associated with an optimal orbit determination method and strategy. A continuous increasing acceleration is established by constructing a constant force that is equivalent to the pulse force, with the mass of the satellite decreasing throughout maneuver. This acceleration can be added to other accelerations, such as solar radiation, to obtain the continuous acceleration of the satellite. The orbit determination method and strategy are illuminated, with subsequent assessment of the orbit being determined and predicted accordingly. The orbit of the GEO satellite during repositioning maneuver can be determined and predicted by using C-Band pseudo-range observations of the BeiDou GEO satellite with COSPAR ID 2010-001A in 2011 and 2012. The results indicate that observations before maneuver do affect orbit determination and prediction, and should therefore be selected appropriately. A more precise orbit and prediction can be obtained compared to common short arc methods when observations starting 1 day prior the maneuver and 2 h after the maneuver are adopted in POD (Precise Orbit Determination). The achieved URE (User Range Error) under non-consideration of satellite clock errors is better than 2 m within the first 2 h after maneuver, and less than 3 m for further 2 h of orbit prediction.     

平台式重力仪航空重力测量数据后处理技术研究

针对航空重力测量对数据后处理的要求,结合平台式重力仪系统的组成与工作原理,在重力异常信号的频域特征基础上,设计出一套有严格频域规范的航空重力数据后处理流程,并对某海域实测数据进行了处理。结果表明,重力异常值内符合精度为12条东西测线达到0.43mGal/140s和0.84mGal/100s,4条南北测线达到0.39mGal/140s和0.79mGal/100s;以同测线GT-2A单次测量结果为标准值统计标准差,12条东西测线为0.72mGal/140s和0.98mGal/100s,4条南北测线为1.41mGal/140s和1.53mGal/100s。结果表明,设计的航空重力测量数据后处理方法可以实现重力异常信号的高精度提取。