推力架测试系统误差的研究

针对卫星推进系统小推力测量的特点,主要是分析推力架系统所带来的推力误差,对引起推力误差的因素进行了分析研究。     

基于Lyapunov最小-最大方法的卫星编队飞行队形保持

提出一种卫星编队飞行队形保持的鲁棒控制方法.将环境摄动、推力误差等因素作为干扰项加到相对动力学方程中.然后对于该不确定系统,采用Lyapunov最小一最大方法设计相应的控制律,在存在干扰的情况下,采用该控制律仍然能保证系统渐近稳定.同时针对经典的Lyapunov最小一最大方法导致的控制颤振问题,进行了改进,证明了改进后的控制律能够实现系统的一致终极有界,同时消除颤振.通过仿真,验证了采用Lyapunov最小-最大方法在系统存在干扰的情况下,比一般控制方法具有更强的鲁棒性.     

Sporadic structures in equatorial ionosphere as revealed from GPS occultation data

A method for monitoring of sporadic formations in the lower ionosphere by use of the amplitude and phase variations of decimeter radio waves in the occultation trans-ionospheric link GPS satellite — LEO CHAMP satellite is described. Typical variations of the amplitude and phase of the occultation signal, caused by layered formations in the lower ionosphere, are considered. Parameters of sporadic structures measured during period of especially strong solar flashes from October 25 till November 9, 2003, are described. Results of statistical analysis of the occurrence frequency of sporadic layers, their altitude distribution, and thickness are presented. The electron density distribution in the lower ionosphere in the equatorial zone is estimated.     

空间电源数字孪生系统

数字孪生系统是一种详细描述现实系统的数字化模拟系统,基于虚拟空间中的模型理解、预测、优化和控制现实系统,与现实系统具有实时性、同步进化性和交互性.为满足未来空间飞行器智能化、高机动性、多目标任务的特点,需要提升空间电源系统智能化监测管理能力.本文分析了空间电源数字孪生系统的构成要素,从数字孪生模型构建、新型传感技术以及...     

软件帧格式同步器设计及查找表的运用

帧格式同步是卫星原始数据地面处理前端的必经步骤。软件帧格式同步具有很好的灵活性,易于修改以适应各种情况的变化,缺点是处理速度较硬件慢。文章介绍了在软件帧格式同步程序设计中,应用各种查找表来提高处理速度,快速锁定同步字和还原帧数据的方法。经过实际应用验证,效果很好。     

碳纤维复合材料卫星天线反射面型面精度稳定性分析

文章简述了卫星天线的组成、制造工艺过程,着重对影响天线反射面型面精度和尺寸稳定性有关内容,从材料特性、铺层分析、成型模具材料和产品质量控制等内容进行了分析。     

产品寿命预测

对基于失效或故障时间数据的产品寿命预测的基本概念、重要意义、方法以及与其它学科的关系等进行了系统评述.     

一种基于迁移学习的遥测数据异常检测方法

为解决卫星遥测数据异常检测面临的数据不平衡且缺乏有标签样本的问题,提出一种基于一维卷积神经网络(1dCNN)迁移学习的异常检测方法.首先利用源域卫星的遥测数据对1dCNN进行预训练,使得模型的卷积层具有卫星状态特征的提取能力;然后将训练好的模型迁移到缺乏标签数据的目标域卫星中;利用目标域有标签样本对预训练模型进行微调,...     

Development cost prediction of general aviation aircraft using combined estimation technique

Obtaining accurate development cost estimation results of general aviation aircraft is crucial for companies to adopt the best strategy in the development process. To address this problem, this paper proposes a combination of three commonly used single prediction methods. The optimal weight values of the three single prediction methods are determined by utilizing the shortest ideal point method. Ten cost datasets collected from literature are utilized for fitting and testing the combined prediction method, and the weight coefficients of the three individual prediction methods are calculated as 0.6859, 0.0035 and 0.3106, respectively. The results of this study indicate that the developed method has better fitting and estimation accuracy than that of the three individual methods, with average fitting and predicting error values of 2.60% and 6.43%, respectively. Additionally, the cost data of military and civil aircraft development from literature are collected for verification. The results further confirm that the proposed method is not only superior to the single prediction methods in terms of high precision but has wider applications. More importantly, this research can provide important reference for general aviation aircraft companies in term of product cost planning and corporate sales strategies.     

应用BP神经网络针对高通量质子事件的通量数值预报

质子事件的爆发与太阳软X射线辐射有着很强的相关性,利用GOES卫星的1～8 (A)波段和0.5～4 (A)波段的软X射线数据,选取一些特征参量验证该相关性并应用到质子事件短期预报中.在当前质子事件传输物理机制不完全明确的情况下,在现有的预报质子事件有无的模型基础上,利用BP神经网络,根据软X射线通量水平等预测事件质子峰值通量水平,再对训练后的网络进行检验,检验预测所得结果与实际探测值误差小于一个量级,具备一定实用意义.