An empirical solar radiation pressure model for satellites moving in the orbit-normal mode

We present a family of empirical solar radiation pressure (SRP) models suited for satellites orbiting the Earth in the orbit normal (ON) mode. The proposed ECOM-TB model describes the SRP accelerations in the so-called terminator coordinate system. The choice of the coordinate system and the SRP parametrization is based on theoretical assumptions and on simulation results with a QZS-1-like box-wing model, where the SRP accelerations acting on the solar panels and on the box are assessed separately. The new SRP model takes into account that in ON-mode the incident angle of the solar radiation on the solar panels is not constant like in the yaw-steering (YS) attitude mode. It depends on the elevation angle of the Sun above the satellite’s orbital plane. The resulting SRP vector acts, therefore, not only in the Sun-satellite direction, but has also a component normal to it. Both components are changing as a function of the incident angle. ECOM-TB has been used for precise orbit determination (POD) for QZS-1 and BeiDou2 (BDS2) satellites in medium (MEO) and inclined geosynchronous Earth orbits (IGSO) based on IGS MGEX data from 2014 and 2015. The resulting orbits have been validated with SLR, long-arc orbit fits, orbit misclosures, and by the satellite clock corrections based on the orbits. The validation results confirm that—compared to ECOM2—ECOM-TB significantly (factor 3–4) improves the POD of QZS-1 in ON-mode for orbits with different arc lengths (one, three, and five days). Moderate orbit improvements are achieved for BDS2 MEO satellites—especially if ECOM-TB is supported by pseudo-stochastic pulses (the model is then called ECOM-TBP). For BDS2 IGSOs, ECOM-TB with its 9 SRP parameters appears to be over-parameterized. For use with BDS2 IGSO spacecraft we therefore developed a minimized model version called ECOM-TBMP, which is based on the same axis decomposition as ECOM-TB, but has only 2 SRP parameters and is supported by pseudo-stochastic parameters, as well. This model shows a similar performance as ECOM-TB with short arcs, but an improved performance with (3-day) long-arcs. The new SRP models have been activated in CODE’s IGS MGEX solution in Summer 2018. Like the other ECOM models the ECOM-TB derivatives might be used together with an a priori model.     

云制造环境下机床装备资源选择方法

针对云制造环境下机床装备资源具有数量大、异质异构、服务制约因素多等问题，提出了基于多准则决策的机床装备资源选择方法。首先，利用本体论对机床装备资源建模，将物理资源虚拟化封装，实现虚拟资源与物理资源的逻辑映射。进一步地，使用语言变量量化决策者评估信息，采用模糊决策与试验评价实验室（FDEMATEL）和熵值法（EW）相结合的组合赋权法，引入模糊VIKOR（FVIKOR）多属性决策方法对机床装备资源进行综合评估，建立FDEMATEL-EW-FVIKOR的多准则混合决策模型。最后，将该方法应用于某生产制造企业，实验对比决策结果验证了多准则混合决策模型的可行性和有效性。     

哨兵2卫星综合水质指标的河流水质遥感监测方法

上海市河流水污染监测是当前研究的热点问题。针对上海市河流水质监测的需求,提出了一种基于因子分析的河流综合水质遥感反演的方法。以欧洲航空局发射的哨兵-2A(Sentinel-2A)卫星搭载的多光谱成像仪获取的遥感影像,以及准同步获取的河流断面实测水质参数数据为例,研究对象为上海市青浦区和松江区部分河道。选取高锰酸盐指数、氨氮等5项主要水质参数为监测指标,在因子分析的基础上建立水质遥感反演的模型,得到综合水质指标用于确定水质类别。实验结果表明:综合水质指标值越小,内陆河流水质状况越良好,水体更清洁。该方法能够应用于内陆河流水质遥感监测,为水环境管理部门提供参考信息。     

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。      

增广误差模型算法在目标跟踪中的应用

针对目标机动运行过程中,滤波模型与机动状态模型失配的问题,提出了一种新的增广状态误差滤波模型。不同于现有增广方案,该模型从模型失配所致状态滤波误差的角度出发,将状态估计误差增广为一状态量,通过滤波估计后用其校正原状态量。算法分析表明,该增广滤波模型具有自适应调节多重渐消因子的等效特性,增强了对目标的跟踪能力。基于该增广状态误差滤波模型,给出了滤波算法设计并进行了仿真实验。实验结果表明,基于该模型的滤波算法在对机动目标进行跟踪时具有更强的鲁棒性。     

The adjusted optical properties for Galileo/BeiDou-2/QZS-1 satellites and initial results on BeiDou-3e and QZS-2 satellites

Solar Radiation Pressure (SRP) is the dominant non-gravitational perturbation for GNSS (Global Navigation Satellite System) satellites. In the absence of precise surface models, the Empirical CODE Orbit Models (ECOM, ECOM2) are widely used in GNSS satellite orbit determination. Based on previous studies, the use of an a priori box-wing model enhances the ECOM model, especially if the spacecraft is a stretched body satellite. However, so far not all the GNSS system providers have published their metadata. To ensure a precise use of the a priori box-wing model, we estimate the optical parameters of all the Galileo, BeiDou-2, and QZS-1 (Quasi Zenith Satellite System) satellites based on the physical processes from SRP to acceleration. Validation using orbit prediction proves that the adjusted parameters of Galileo and QZS-1 satellites exhibit almost the same performance as the corresponding published and “best guess” values. Whereas, the estimated parameters of BeiDou-2 satellites demonstrate an improvement of more than 60% over the initial “guess” values. The resulting optical parameters of all the satellites are introduced into an a priori box-wing model, which is jointly used with ECOM and ECOM2 model in the orbit determination. Results show that the pure ECOM2 model exhibits better performance than the pure ECOM model for Galileo, BeiDou-2 GEO and QZS-1 orbits. Combined with the a priori box-wing model the ECOM model (ECOM+BW) results in the best Galileo, BeiDou-2 GEO and QZS-1 orbits. The standard deviation (STD) of satellite laser ranging residuals reduce by about 20% and 5% with respect to the pure ECOM2 model for Galileo and BeiDou-2 GEO orbits, while the reductions are about 40% and 60% for QZS-1 orbits in yaw-steering and orbit-normal mode respectively. BeiDou-2 IGSO and MEO satellite orbits do not benefit much from the a priori box-wing model. In summary, we suggest setting up a unified SRP model of ECOM+BW for Galileo, QZS-1, and BeiDou-2 orbits based on the adjusted metadata. In addition, we estimate the optical parameters of BeiDou-3e and QZS-2 satellites using a limited number of tracking stations. Results regarding the unified SRP model indicate the same advantages, the STD of satellite laser ranging residuals reduces by about 30% and 20% for QZS-2 and BeiDou-3e orbits respectively over orbit products without a priori model. The estimation procedure is effective and easy to apply to the new emerging satellites in the future.     

无刷双馈电机静止坐标系数学模型的研究

由于没有电刷和滑环,提高了运行的可靠性,降低了维护成本,因此无刷双馈电机(BDFM)在变速恒频发电领域有着广阔的应用前景。由于特殊的结构和工作原理,BDFM数学模型十分复杂。从BDFM多回路空间矢量数学模型出发,推导出了两个静止坐标系之间的坐标变换关系以及功率绕组静止坐标系数学模型和控制绕组静止坐标系数学模型,推导过程简洁明了。所得结果为BDFM理论分析和不同控制策略的研究提供了理论基础。MATLAB/Simulink环境下的仿真结果验证了数学模型及坐标变换关系的正确性。     

C/SiC材料主被动氧化烧蚀机理及计算方法研究

基于热化学平衡方法建立了任意比例C/SiC材料的主被动氧化烧蚀模型，开展了C/SiC材料氧化烧蚀机理的计算研究，并基于典型材料烧蚀试验结果进行了充分验证。计算结果表明，C/SiC材料的氧化烧蚀特性取决于表面温度、氧分压以及组分等因素，可能会出现主动氧化和被动氧化两种破坏机制，目前的烧蚀模型能够预测出任意比例C/SiC材料两种氧化烧蚀机制的转换过程；SiC含量对C/SiC材料的氧化烧蚀特性有明显的影响，随着SiC含量的提升，主/被动氧化转换临界分压会减小，材料的抗氧化性能越好；但当材料均处于主动氧化阶段时，SiC含量越高材料的无量纲烧蚀速率越大，材料的抗烧蚀性能减弱。     

谱估计理论在弹道数据参数化建模中的应用

针对高超声速飞行器防御作战中目标运动规律的描述问题，提出一种基于经典和现代谱估计理论的弹道数据参数化分析和建模方法。该方法将弹道数据视作非平稳随机信号，研究其变化规律：首先，通过线性消势法消除信号的线性趋势，将其转变为平稳信号，以便于进行谱估计；然后，采用Welch算法给出大致的谱图，结合该谱图和改进的协方差法确定自回归参数模型的阶数，以避免模型阶数选择准则引起的不确定性问题；最终，给出弹道数据的参数化模型。仿真结果表明，使用本文算法建立的弹道数据参数化模型与动力学模型具有较高的一 致性。     

基于Probit模型的航空差旅管理市场分析

研究分析企业和差旅管理公司的哪些因素会对差旅管理业务的选择产生影响,帮助差旅管理公司明确企业的差旅管理偏好,从而推动中国航空差旅管理市场的发展.针对航空差旅管理市场的现状,在对50家企业进行问卷调查后,通过Probit模型得出10个因素对企业选择差旅管理业务的影响程度,其中“现金流量”、“资金结算手段”、“增值服务种类”和“政策与系统的匹配度”4个因素具有显著性.最后,根据实证结果对企业和差旅管理公司提出相应的建议.