深空通信中提示否定确认型CFDP协议延时估算

为确定空间数据咨询委员会（CCSDS）建议高级在轨系统中CCSDS文件传输协议（CFDP）在深空通信环境中的适应性,研究了CFDP中提示否定确认型文件传输协议（Prompt NAK CFDP）的传输机理。给出了此传输方式的文件传输时延估算方法。分析了误码率、传输速率、协议数据单元的大小与数量,以及提示信息数等对传输时延的影响,获得了深空通信对误码率的要求。研究表明：CFDP可用于深空通信,为其进一步应用和改进提供了参考。     

某卫星挠性附件姿态动力学研究

根据导出的某有大型太阳帆板的挠性卫星数学模型,分析了模型中挠性振动对质心和姿态运动的影响,给出了卫星系统的动力学方程,并对帆板展开、卫星稳态运行的模型作工程化处理。分析和仿真结果表明：该方法能降低模型的复杂度,满足大型卫星的姿态指向精度和稳定度要求。     

舰载机起飞时机辅助决策系统建模

 由于航母扰动会影响舰载机起飞安全,而仅依靠个人经验决策起飞时机存在较大误差,因此提出了舰载机起飞时机辅助决策系统的概念并建立模型。通过分析航母运动对舰载机起飞的影响机理,选择飞机离舰爬升率作为评价起飞安全性的关键参数;结合舰载机起飞作业程序,确立起飞时机决策方法;在此基础上,利用人工神经网络的预测能力并结合有利/不利起飞时机的特征,建立舰载机起飞时机辅助决策系统模型。仿真结果表明,系统模型能够有效地为起飞指挥官提供辅助决策信息,提高舰载机起飞安全性。     

基于离散伴随的流场反演在湍流模拟中的应用

精确模拟湍流流动是学术界和工业界均普遍关注的问题。采用数据驱动湍流建模的思路,建立了基于离散伴随方法的流场反演框架。通过为SA模型涡黏性输运方程的生成项乘以非均匀分布的系数,并利用有限的观测数据对该系数进行推断,实现对SA模型的修正。为了提高带有物理约束的离散伴随优化的效率,发展了约束增广的伴随方法,其高效性在本文得到了验证。选取了结冰翼型和周期山2个分离算例进行分析,所得结果在2个算例中均能以很高的精度拟合观测数据,并能借助湍流模型的修正将有限的观测信息泛化到整个流场。分析表明,流场反演所推断出的修正区域具有较为明确的物理意义,能够指导湍流模型的进一步改进。     

流体力学深度学习建模技术研究进展

深度学习技术在图像处理、语言翻译、疾病诊断、游戏竞赛等领域已带来了颠覆性的变化。流体力学问题由于维度高、非线性强、数据量大等特点,恰恰是深度学习擅长并可以带来研究范式创新的重要领域。目前,深度学习技术已在流体力学领域得到了初步应用,其应用潜力逐渐得到证实。以流体力学深度学习技术为背景,结合课题组近期研究结果,探讨了流体力学深度学习建模技术及其最新进展。首先,对深度学习技术所涉及的基本理论做了介绍,阐释流场建模中常用深度学习方法背后的数学原理。其次,分别对流体力学控制方程、流场重构、特征量建模和应用等几个典型的人工智能与流体力学交叉问题应用场景所涉及的深度学习技术研究进展进行了介绍。最后,探讨了流体力学深度学习建模技术所面临的挑战与未来发展趋势。     

星敏感器低频误差分析

介绍星敏感器的误差分类、低频误差定义、误差产生原因及其测试标定方法等.星敏感器输出姿态周期性误差的主要原因为视场空间误差和热弹性变形误差,典型星敏感器采取相应的解决措施后,低频误差可控制在0.8″以内.     

应用灰度特征的行星陨石坑自主检测方法与着陆导航研究

针对陨石坑阴影不明显的行星着陆导航图像,提出了基于图像灰度值特征的陨石坑自主检测方法,通过兴趣区快速确定陨石坑边缘分布,解决了一般陨石坑检测方法对图像太阳高度角的依赖问题。利用检测出的陨石坑视线信息,采用非线性预测滤波方法估计着陆器着陆阶段的位置和姿态;鉴于视觉着陆导航对照了场景状况的高度依赖性,提出了一种利用分形理论和相邻点分级方法的天体随机地景建模方法。最后验证了所提方法的有效性。     

隔离信号调理模块自动校准系统设计

针对隔离信号调理模块型号多、参量多、使用量大,采用手动计量校准耗时耗力、容易出错、工作效率低等问题,在VEE Pro软件开发平台上,设计开发出一套隔离信号调理模块自动校准系统。该系统包括仪器信息记录、校准及打印输出三个主模块,测试过程采用实时测试显示,自动进行数据处理并存档。使用结果表明：该系统操作简便,提高了工作效率及准确度,降低了人为出错几率,确保了测试数据准确可靠。     

Storm-time atmospheric density modeling using neural networks and its application in orbit propagation

Upper atmospheric densities during geomagnetic storms are usually poorly estimated due to a lack of clear understanding of coupling mechanisms between the thermosphere and magnetosphere. Consequently, the orbit determination and propagation for low-Earth-orbit objects during geomagnetic storms have large uncertainties. Artificial neural networks are often used to identify nonlinear systems in the absence of rigorous theory. In the present study, an attempt has been made to model the storm-time atmospheric density using neural networks. Considering the debate over the representative of geomagnetic storm effect, i.e. the geomagnetic indices ap and Dst, three neural network models (NNM) are developed with ap, Dst and a combination of ap and Dst respectively. The density data used for training the NNMs are derived from the measurements of the satellites CHAMP and GRACE. The NNMs are evaluated by looking at: (a) the mean residuals and the standard deviations with respect to the density data that are not used in training process, and (b) the accuracy of reconstructing the orbits of selected objects during storms employing each model. This empirical modeling technique and the comparisons with the models NRLMSIS-00 and Jacchia-Bowman 2008 reveal (1) the capability of neural networks to model the relationship between solar and geomagnetic activities, and density variations; and (2) the merits and demerits of ap and Dst when it comes to characterizing density variations during storms.     

GRACE accelerometer calibration by high precision non-gravitational force modeling

The precise modeling and knowledge of non-gravitational forces acting on satellites is of big interest to many scientific tasks and missions. Since 2002, the twin GRACE satellites have measured these forces in a low Earth orbit with highly precise accelerometers, for about 15?years. Besides the significance for the GRACE mission, these measurement data allow the evaluation of modeling approaches and the improvement of force models. Unfortunately, before any scientific usage, the accelerometer measurements need to be calibrated, namely scale factor and bias have to be regularly estimated.In this study we demonstrate an accelerometer calibration approach, solely based on high precision non-gravitational force modeling without any use of empirically or stochastically estimated parameters, using our in-house developed satellite simulation tool XHPS. The aim of this work is twofold, first we use the accelerometer data and the residuals resulting from the calibration to quantitatively analyze and validate different non-gravitational force model approaches. In a second step, we compare the calibration results to three different calibration methods from different authors, based on gravity field recovery, GPS-based precise orbit determination, and based on modeled accelerations.We consider atmospheric drag forces and winds, as well as radiation forces due to solar radiation pressure, albedo, Earth infrared and thermal radiation (TRP) of the satellite itself. For TRP, we investigate different transient temperature calculation approaches for the satellite surfaces with absorbed power from the aforementioned radiation sources. A detailed finite element model of the satellite is utilized for every force, considering orientation, material properties and shadowing conditions for each element.For cross-track and radial direction, which are mainly affected by the radiative forces, our calibration residuals are quite small when drag is not super dominant (1–3?$\text{nm}/{\text{s}}^2$ for total accelerations around $\pm$50?$\text{nm}/{\text{s}}^2$). For these directions the calibration seems to perform better than the other compared methods, where some bigger differences were found. For the drag dominated along-track direction it is vice versa, here our method is not sensitive enough because the difference between modeled and measured drag is bigger (e.g. residuals around 10?$\text{nm}/{\text{s}}^2$ for total accelerations around $\pm$70?$\text{nm}/{\text{s}}^2$ for low solar activity). In along-track direction the orbit determination based methods are more sensitive and produce more reliable results. Results for the complete GRACE mission time span from 2003 to 2017 are shown, covering different seasonal environmental conditions.