无人战斗机缩比训练机气动特性计算分析

针对无人战斗机系统维护和飞行训练耗费巨大的问题,借鉴缩比模型验证机飞行试验的思路,研制具有相同飞行性能和操纵特性的缩比训练机。应用相似性原理,通过相似参数分析获得了缩比机与全尺寸原型机的飞行范围。在缩比机几何外形的基础上,应用Gridgen软件创建机体周围流场的非结构网格。通过Fluent软件,计算了缩比机纵向气动特性。根据俯仰力矩系数估算出全机气动中心位置,为缩比机的操纵性稳定性研究和控制率设计奠定了基础。     

Advanced analytical model for orbital aerodynamic prediction in LEO

The growing interest in low earth orbit (LEO) applications demands for accurate modeling of orbital aerodynamics. But classical analytical models of aerodynamic coefficients in free molecule flow, such as the Sentman’s model, Schamberg’s model and Schaaf-Chambre model, were built upon over simplistic gas-surface interaction models, which degrade the fidelity of aerodynamic prediction. This work presents a new analytical model of orbital aerodynamic coefficients based on the state-of-the-art Cercignani–Lampis–Lord (CLL) gas-surface interaction model, where lobular quasi-specular scattering pattern and separate accommodation degree for different velocity components can be well captured. A key component of the new model is a rigorous function approximation solution of the reflected normal momentum flux based on the CLL model which is derived for the first time and is validated within 1% for any hypothermal flow and surface accommodation conditions. Closed-form analytical solutions of aerodynamic coefficients for simple convex geometries are obtained and exhibit high accuracy (within 0.1%) in typical LEO scenarios. The new analytical model surpasses the classical models in some important aspects, such as overcoming the diffuse scattering hypothesis constraint, considering the variation of normal momentum exchange with the surface incidence angle and being applicable in any hypothermal flow situation. In virtue of the advanced CLL model and feasibility of coupling with the panel method technique, the new analytical model is promising to provide more accurate predictions on the orbital aerodynamic coefficients for LEO applications.     

基于天基边缘计算的在轨智能技术

针对低轨互联网星座的发展以及天基应用不断拓展对计算能力的需求,开展基于天基边缘计算的在轨智能技术研究,这对提升星地带宽利用率和实时决策效率具有重要意义.提出了以低轨卫星为核心的天基边缘计算架构,在此基础上,进一步提出了一种基于深度模仿学习的智能计算卸载模型,以及该模型的在轨分布式联邦学习训练方案.基于空天地海一体化网络...