Monopropellant droplet decomposition for large activation energies

The quasi-steady decomposition of a monopropellant droplet in a quiescent atmosphere is analyzed, in the limit of large activation energy, for a gas phase decomposition reaction of the Arrhenius type.An analytical relation is obtained for the pre-exponential rate constant, or the Damköhler number, as a function of the droplet vaporization rate. The curve giving the vaporization rate in terms of the Damköhler number has, for low values of the temperature at infinity, an S shaped form which exhibits ignition-extinction characteristics.With increasing values of the droplet vaporization rate above the pure vaporization value we pass from a nearly frozen regime, under which ignition conditions occur due to the large temperature sensitivity of the reaction rate, to a complete decomposition regime, in which practically all the fuel is decomposed in a thin reactive-diffusion zone, separated from the droplet surface by a transport region of lower temperature. For sufficiently large Damköhler numbers, the thin reaction zone is located close to the droplet surface, so that the flame structure becomes planar: a linear relation between the droplet radius and time is obtained in this case. The non planar effects are responsible for the extinction of the flame at low Damköhler numbers and ambient temperatures below the adiabatic flame temperature. For sufficiently low values of the ambient temperature, an intermediate regime exists in which the enhancement of the vaporization rate is due only to the fraction of the vaporized fuel decomposed at a thin reaction zone, not far from the droplet, where the temperature reaches its maximum value; the remaining fuel is decomposed very far from the droplet.     

An approach to the autoignition of a turbulent mixture

This paper considers the turbulent homogeneous mixing of two reactants undergoing a one step, second order, irreversible, exothermic chemical reaction with a rate constant of the Arrhenius type. A statistically stationary turbulent velocity field is assumed given and unaffected by mass or heat production due to the chemical reaction. Relative density fluctuations are neglected. A Hopf-like functional formalism is presented, with application to both statistically inhomogeneous and statistically homogeneous flows. Single and double point probability density function differential equations are derived from those functional equations. The limit of very large activation energies is considered; a low degree of statistical correlation between temperature and concentration fields during the ignition period is hypothesized. After making use of the homogeneity assumption a closure problem is still present due to the nonlocalness of the molecular diffusion term. The problem is rendered closed by assuming a Gaussian conditional expected value for the temperature at a point given the temperature at a neighboring point. The closure is seen to preserve very important mathematical and physical properties. A linear first order hyperbolic differential equation with variable coefficients for the probability density function of the temperature field is obtained. A second Damköhler number based on Taylor's microscale turns out to be an important controlling parameter. A numerical integration for different values of the second Damköhler number and the initial stochastic parameters is carried out. The mixture is seen to evolve towards an eventual thermal runaway, the detailed behavior however being different for different systems. Some peculiarities during the ignition period evolution are uncovered.     

On gas-phase ignition of diffusion flame in the stagnation-point boundary layer

An analytical expression for gas-phase ignition is developed for a diffusion flame in the two-dimensional or axisymmetric stagnation-point boundary-layer flow of a hot oxidizing gas about a vaporizing condensed fuel surface. The analysis is based on the limit of large activation energy for a one-step, irreversible reaction describing the overall combustion process in the gas phase. The approach in this work, following that of our recent analysis on extinction for the same geometry, is to seek an exact correspondence of the parameters of the present problem with those of counterflow diffusion-flame problem of Liñán. Such a correspondence has been found in the frozen-flow regime and as a consequence, the asymptotic structure of the flame in the nearly frozen ignition regime is identical in both the problems. A particular result of this observation is the availability of an analytical criterion for ignition in the present problem. The analysis reveals that contrary to the case of extinction, fluid dynamic details do not have significant effect on the ignition criterion and that Liñán's results may be applied with good accuracy to the condensed fuel problem.     

基于节点改善策略的伪谱轨迹优化

基于Legendre Gauss Lobatto（LGL）伪谱法(PSM)求解最优控制问题的原理,研究了有限推力轨迹快速优化设计以及计算节点和效率改善策略。对远程变轨动力学模型进行了无量纲化处理,并设计了初值生成和串行优化求解策略,用来提高多变量多约束大规模稀疏非线性规划问题的收敛性。结合极小值原理推导了问题的最优性必要条件,基于协态映射原理设计了数值解的最优性验证方法。利用二阶微分方程技术消去部分状态量,建立了节点和计算效率改进模型,并讨论了该策略的相关数值处理技术。仿真结果表明,本文的算法可以快速地提供应用于实际飞行任务的最优解,节点改善策略在保证计算精度的同时,有效降低了计算收敛时间。     

Adaptive sparse grid quadrature filter for spacecraft relative navigation

This paper explores a novel adaptive sparse grid quadrature filter. The sparse grid quadrature approach has been recently developed for nonlinear estimation problems to alleviate the curse-of-dimensionality issue of the Gauss–Hermite quadrature filter. Accuracy level of the sparse grid quadrature filter is an important tuning factor that affects desired performance. The proposed filter autonomously adjusts the accuracy level of the sparse grid quadrature rule in both prediction and update steps by increasing the level gradually until an adaptation criterion is satisfied. The adaptation criterion is derived based on a quadrature error estimator. The nestedness property of sparse grid quadrature rule enables efficient computation in adaptation by reusing quadrature points of the previous level sparse grid quadrature. An application to spacecraft relative navigation has been made to demonstrate the adaptive spare grid quadrature filter outperforming the extended Kalman filter and the unscented Kalman filter.     

分析插值误差和斜率的轨迹优化网格细化方法

提出一种基于插值误差和斜率分析的轨迹优化自适应网格细化方法,包括节点插入算法和节点删除算法。节点插入算法分析各个离散节点的控制变量的插值误差。若插值误差较大,则在该节点周围增加节点细化网格;否则,不进行细化。节点删除算法分析各个离散节点处的控制变量斜率。若某个节点的左斜率和右斜率都为零,那么删除该节点;否则,保留该节点。采用三个典型的轨迹优化算例验证了所提出的方法的有效性和特色,并且与其它几种网格细化方法进行了对比。仿真结果表明,本文方法生成的网格规模较小,需要的网格迭代次数较少,能够快速、高精度求解非光滑轨迹优化问题。     

周边式对接机构的航天器首次接触撞击点的确定

对周边式对接机构的航天器进行了研究，在对接捕获过程中首次接触撞击点的确定问题，文中建立了三类可能接触区域的空间解析描述关系式，根据空间“点”投影的不变性原理，给出了确定可能接触点的数学方程，使原问题由三维空间问题化为二维平面问题，其难度和复杂性都大大简化。最后，文中还给出了多组算例。     

特征模型分散式自适应姿态控制在高超声速飞行器中的应用

 针对高超声速飞行器在再入过程中强耦合、大扰动和气动参数大范围变化的问题,基于特征建模的思想,把原非线性动力学方程用一个二阶时变差分方程组形式的特征模型描述,建立了攻角通道独立、 偏航/滚转通道耦合的特征模型,并设计了分散式自适应姿态控制器,给出了稳定性分析和数值仿真。由于基于特征模型设计的控制器组成的闭环系统是一个非常复杂的混合系统,稳定性分析在特征建模理论中一直是一个难点,提出的稳定性分析方法完全解决了此类不含内动态相对阶为二的多输入多输出系统的稳定性问题。     

Folding patterns of planar gossamer space structures consisting of membranes and booms

The combination of large membranes and light-weight deployable booms, often called a gossamer structure, has enabled innovative space missions, such as solar sailing, to become possible. Though many designs have been proposed and demonstrated, two problems remain regarding the folding patterns of the membranes. The first problem involves considering the thickness of a membrane to enable uniform and compact folding. The other involves membrane-folding patterns that allow for connecting the membrane to the booms at multiple points and deploying them together while minimizing the use of complex mechanisms. This study proposes three methods that consider the thickness, and two of them can keep the crease lines straight, in contrast to the conventional non-straight crease line solutions. In addition, this study derives one effective design to integrate a membrane with diagonal booms through the systematic classification of existing membrane folding patterns.     

An experimental study of flame turbulence

An experimental study of the turbulence generated by an enclosed, premixed, propane-air flame has been carried out in a combustion chamber of 25 × 20 cm cross section. Care was taken to reduce any effects of the axial pressure gradient. By suitable changes in the grid geometry, the turbulence intensity and scale of the approach flow were varied independently. The results of these experiments show that a strong link exists between the mechanisms of turbulent flame propagation and flame-generated turbulence. Thus three distinct regions may be identified, each having different characteristics in regard to the effects of turbulence scale on flame-generated turbulence. For each region, a physical mechanism for flame-generated turbulence is proposed. In particular, it is observed that over a wide range of intensity and scale of the approach turbulence, (a) the relative turbulence intensity in the flame zone varies in the range 1–2 times the relative turbulence intensity of the cold flow, (b) in the region of intermediate turbulence levels ( ) the flame-generated turbulence intensity reaches a minimum value which is equal to the approach stream turbulence intensity, (c) the flame-generated turbulence intensity reaches a maximum value when the rate of production of turbulent vorticity is equal to about half its rate of viscous dissipation.     

Design of satellite constellations with continuous coverage on elliptic orbits of <Emphasis Type="Italic">Molniya</Emphasis> type

New methods of choosing the structures of satellite constellations (SC) on elliptical orbits of the Molniya type are presented. The methods, using critical inclination and putting the orbit apogee in the Earth’s hemisphere with an area of continuous coverage, are based on geometrical analysis of two-dimensional representation of the coverage conditions and SC motion in the space of inertial longitude of the orbit ascending node and time. The coverage conditions are represented in the form of a certain region. Dynamics of all satellites in this space is represented by uniform motion along a straight line approximately parallel to the ordinate axis, while the satellite system forms a grid. The problem of choosing a minimal (as far as the number of satellites is concerned) SC configuration can be formulated as a search for the most sparse grid. The contemporary advanced methods of computational geometry serve as an algorithmic basis for the problem solution. Design of SC for continuous coverage of latitude belts with the use of kinematically regular systems is considered. A method of analyzing single-track systems for continuous coverage of arbitrary geographic regions is described, which makes a region at any time instant observable by at least one satellite of the system. As an example, SC on elliptical orbits are considered with periods of ～4, 12, and 24 hours.     

基于网格自适应的飞行器防热材料热传导系数辨识

为准确辨识高超声速飞行器防热材料热传导系数，构造了针对热传导偏微分方程系统的约束泛函极值问题，基于表征多项式逼近性能的Weierstrass定理，采用Lagrange多项式对热传导系数进行参数化建模，进而将无穷维的约束泛函极值问题转化为有限维的非线性规划问题，再利用基于动态优化方程的优化方法将此非线性规划问题转化为常微分方程初值问题进行求解。为从较有限的测量数据中准确地“学习”热传导系数，建立采用“过拟合”判别准则的网格自适应迭代算法改善辨识精度。研究表明本文采用的辨识策略与优化方法有效，辨识结果可以较准确地反映材料热传导系数的变化规律。     

NASA'S Robotic Lunar Lander Development Project

Over the last 5 years, NASA has invested in development and risk-reduction activities for a new generation of planetary landers capable of carrying instruments and technology demonstrations to the lunar surface and other airless bodies. The Robotic Lunar Lander Development Project (RLLDP) is jointly implemented by NASA Marshall Space Flight Center (MSFC) and the Johns Hopkins University Applied Physics Laboratory (APL). The RLLDP team has produced mission architecture designs for multiple airless body missions to meet both science and human precursor mission needs. The mission architecture concept studies encompass small, medium, and large landers, with payloads from a few tens of kilograms to over 1000 kg, to the Moon and other airless bodies. To mature these concepts, the project has made significant investments in technology risk reduction in focused subsystems. In addition, many lander technologies and algorithms have been tested and demonstrated in an integrated systems environment using free-flying test articles. These design and testing investments have significantly reduced development risk for airless body landers, thereby reducing overall risk and associated costs for future missions.     

一种自动优化的三角形有限元网格生成技术

提出了一种自动优化和三角形有限元网格生成方法，这种方法可以给出最佳的单元形态和节点分布方式，非常有利于提高有限元分析的精度和节约计算机内存及机时，该法已经应用于固体火箭发动机燃烧稳定性预估，并取得了良好的结果。由计算结果证明，该法正确，有效和实用。     

基于AE-UKF的航天器自主导航方法

针对自主导航过程中不确定的测量噪声和干扰的影响,提出了一种具有自适应能力的状态估计方法AE\|UKF方法。该方法通过对非线性的状态方程进行Unscented变换,对非线性的测量方程求取雅可比矩阵,同时在滤波迭代过程中引入自适应因子,使滤波器具有自适应能力。将AE\|UKF方法应用于自主导航系统导航信息的估计过程中,仿真结果表明：AE\|UKF方法在达到UKF方法精度的同时,能够克服自主导航过程中不确定的噪声和随机干扰的影响。     

增强高超声速化学反应流数值计算的稳定性方法

针对高超声速化学反应流数值模拟中包含源项对角化雅可比矩阵的数值格式稳定性很差的问题,在对采用上下对称的高斯-赛德尔(LU-SGS)格式的离散方程进行线性稳定性分析的基础上,提出一种依据流动状态对流场的块网格进行局部加密的方法,以增强流场数值计算稳定性。在高超声速流场中,对包含激波后亚声速高温区域的块网格给予加密,而其他区域中的块网格可以适当进行网格粗化或保持不变。算例结果表明,相比于初始网格,局部加密后的改进网格并不会产生过多的网格量,而且能够采用较大的柯朗-弗里德里奇-列维（CFL）数以达到数值求解过程中稳定计算和快速收敛的目的。     

磁悬浮控制力矩陀螺框架伺服系统扰动力矩分析与抑制

对磁悬浮控制力矩陀螺框架伺服系统，提出了一种基于自适应逆扰动消除控制的设计方法。该控制器采用将被控对象动态控制和对象扰动控制分离处理的方法研究了控制力矩陀螺框架伺服系统的非线性摩擦干扰力矩、陀螺房内部高速磁悬浮转子系统因框架变速转动对框架伺服系统产生的耦合力矩以及航天器姿态改变导致框架伺服系统本身的参数大范围变化等问题。设计的自适应逆扰动消除器大大改善了框架伺服系统的控制性能、提高了系统的速率输出精度。仿真结果表明，提出的扰动力矩抑制方法是可行的，具有很强的鲁棒性。     

小型固体运载器起飞段姿态控制方法研究

针对控制系统,采用侧喷流发动机和栅格舵的新型小型多级固体运载火箭开展了起飞段姿态控制方法研究。给出了在低马赫数条件下侧喷流发动机的力放大因子和栅格舵的控制力矩,建立了运载器姿态控制模型。根据侧喷流和栅格舵的特点,设计了PID控制器与智能开关控制器相结合的新型控制器,运用神经网络算法实现了开关控制器参数的在线选择。仿真结果表明,所设计的控制器满足控制精度、稳定度和鲁棒性的要求,算法简便,具有工程可实现性,以侧喷流发动机和栅格舵为执行机构的控制系统能满足运载火箭起飞段姿态控制要求,为陆基或空射小型固体运载火箭姿态控制方法提供了一种新思路。     

地基光学图像中本体-帆板结构航天器姿态估计方法

提出一种利用地基光学图像估计空间目标姿态的方法。针对本体-帆板结构航天器外形结构特点，定义了一种不易遮挡的两轴结构特征，设计了地基光学图像中该特征的提取方法；根据航天器成像过程建立了特征点二三维几何投影方程，实现了航天器三维姿态的解算。利用单站序列图像，可获得航天器姿态变化角速度信息。在轨航天器的试验结果表明，本方法能有效地从地基光学图像估计出航天器姿态。     

预警卫星对主动段弹道跟踪算法研究

针对预警卫星视线（LOS，Line-of-Sight）测量下对弹道导弹目标的跟踪问题，传统的方法（如EKF）存在易发散、不稳定的现象。引入一种更为稳健和有效的滤波估计算法UKF，其通过特别选取一些样点，给出随机变量经过非线性变化后的均值和方差，从而得到滤波器基于非线性方程的更新，避免了非线性方程的线性化。解析表达了导弹推力加速度值和导弹在重力作用下姿态的变化特性，提出了基于弹体旋转的主动段运动方程。通过Mortte-Carlo仿真，与其它算法比较，说明该算法跟踪性能更好，在空间预警系统应用中具有较好的适应能力。