Hierarchical scheduling for real-time agile satellite task scheduling in a dynamic environment

The imaging processes of optical satellites can be easily affected by unexpected environmental changes, such as changes in cloud coverage. Given the difficulty to predict environmental uncertainties, traditional offline scheduling methods need a follow-up re-scheduling process that responds to real-time environmental information. This repetitive scheduling processes make the offline fine scheduling process a waste of computational resources. Additionally, the offline scheduling method is quite complex owing to its lack of a hierarchy mechanism. To solve these problems, we propose a hierarchical scheduling method for the real-time scheduling problem. This method divides the scheduling process into three steps: pre-assignment, rough scheduling, and fine scheduling. A hierarchical scheduling algorithm based on ant colony algorithm is proposed. Tests with 36 scenarios show that the calculation time is efficiently reduced with this new mechanism. With this consideration of the dynamic environment, the re-scheduling process becomes unnecessary, meaning the wasting of computational resources is avoided and the solution profit is improved.     

Simultaneous observations of spread F irregularities over Southeast Asia longitude using Sanya VHF radar,GPS and C/NOFS

Through concurrently measurements by Communication/Navigation Outage Forecasting System (C/NOFS), Sanya VHF radar and GPS ionospheric scintillation receiver on 12 March 2010, five plasma bubbles were found and three of them were observed by all those instruments. Two well-developed plumes with strong backscatter echoes were measured by Sanya radar and their corresponding depletions were observed by C/NOFS in Orbit 10317, 10318 and 10319. Broad plasma depletions resulting from merging process were found in orbit of 10318. The occurrence time and geophysical positions of scintillations correlate well with observations implemented by Sanya VHF radar and C/NOFS. Observations from three types of instrument indicate that the spread F irregularities have distinct scale. There were longitudinal differences between Sanya VHF radar and C/NOFS as irregularities measured, and the eastward drift of developed bubbles are responsible for these differences.     

基于热喷涂技术制备的热障涂层

着眼于热喷涂技术在热障涂层制备方面的研究现状,在分析热障涂层失效原因的基础上,介绍了等离子喷涂技术及冷喷涂技术和在热障涂层结合层和陶瓷层制备方面的研究进展。     

有色噪声作用下的INS/地磁组合导航算法研究

为了克服基本扩展卡尔曼滤波对噪声统计特性的约束,针对地磁测量噪声为有色噪声的特性,基于扩维卡尔曼滤波算法实现了惯性/地磁组合导航算法。分析了分布式组合导航方案的基本原理,在子滤波器观测信息为实测地磁场总强度信息条件下,基于巡航导弹巡航段飞行过程采用基本扩展卡尔曼滤波和扩维卡尔曼滤波算法进行了仿真分析。仿真结果表明,扩维卡尔曼滤波具有较好的稳定性和收敛性,解决了扩展卡尔曼滤波算法的发散问题。     

缝道参数对多段翼型气动性能的影响

数值模拟研究缝道参数对着陆构型多段翼型气动性能的影响,采用有限体积法求解雷诺平均N-S(Navier-Stokes)方程,湍流模型采用S-A(Spalart-Allmaras)模型,给出了多段翼型升力系数与表面压力系数随各缝道参数的变化规律.计算结果表明:采用的计算模型可用于二维多段翼型的计算;缝道参数对多段翼型气动性能的影响显著,在各个缝道参数的变化范围内升力系数均存在最优值,适当的缝道参数可以抑制后翼上表面逆压梯度,消除后翼分离,并保持前翼尾迹不与后翼边界层相掺混,提高多段翼型的气动性能.      

基于多圆交汇的天文定位与组合导航方法

惯性/天文组合导航系统具有全自主、抗干扰能力强等特点,在一些特殊的导航领域受到了人们的高度重视。本文研究了一种天文多圆交汇迭代定位算法,具有数值计算稳定,适合任意多颗导航恒星参与计算的优点,并能同时计算出对应的定位误差协方差阵;在此基础上,将捷联惯性导航系统与天文导航系统组合,构成了扬长避短的组合导航系统,采用扩展卡尔曼滤波算法实现捷联惯导与天文定位两者的信息融合。最后进行了仿真实验,其结果表明,该天文定位算法简单有效,定位误差模型准确,组合后的系统具有较高的精度。     

基于STK的GPS/LEO无线电掩星技术仿真研究

GPS/LEO无线电掩星技术反演地球大气参数剖面已经具有较高的精度. 国外开展了多个GPS/LEO掩星项目, 但中国还尚未深入进行相关的实验, 这制约了中国掩星技术的发展. 本文提出基于STK进行GPS/LEO掩星技术研究的方法; 根据GPS/LEO掩星的原理, 推导出掩星事件发生的条件和掩星切点的计算公式; 利用STK对掩星过程进行模拟, 得到掩星数据. 在大气球对称假设和大气模型已知的条件下, 反演得到中性大气折射指数. 通过比较模型和反演数据, 表明反演数据精度较高, 验证了利用STK模拟GPS/LEO掩星实验方法的可行性.      

考虑耦合损伤的燃气轮机叶片材料高低周复合疲劳寿命研究

燃气轮机叶片在实际工作过程中,受到高低周载荷共同作用而易发生疲劳失效.针对高低周复合疲劳损伤演化和寿命预测比较复杂这一问题,考虑高低周载荷叠加形成的耦合损伤,建立了一种新的高低周复合疲劳寿命预测模型,在此基础之上,引入弱化函数描述某些低幅高周载荷对复合损伤的弱化作用,建立了具有低载强化寿命效应的高低周复合疲劳寿命预测模...     

铁木辛柯梁固有振动频率的边界元解法

根据傅里叶变换推导具有两个广义位移的铁木辛柯梁固有振动的基本解.利用加权残量方法,从控制微分方程出发建立边界积分方程,进而根据边界条件得到频率方程,采用代数特征值方法和影响系数方法求解频率,并分析了两种方法的特点.以杆为例证明了对于一维均匀结构,对不同的边界条件利用边界元方法(BEM,Boundary Element Method)都可以得到精确频率.将铁木辛柯梁的BEM结果与有限元结果和精确解进行了比较.     

基于动态Bayesian网络的叶片加工质量监控与溯源

 针对叶片加工过程中质量精度不高的问题,提出了基于动态Bayesian网络的叶片加工质量监控与溯源方法.利用动态Bayesian网络建立起叶片加工工序间的相互联系,实现对整个加工过程的控制.基于Bayesian网络对影响加工工序的因素集建立因果联系,采用多元统计过程控制中的T²控制图完成对各工序影响因素集的监控.进行误差溯源时,根据Bayesian网络建立的因果关系对失控样本的T²统计量依据原因变量进行误差分解,并构建各分解变量的控制限,将其作为误差源判定的条件.通过对某叶片加工过程的仿真,验证了所提方法的有效性.