Effects of entry conditions on cracked kerosene-fueled supersonic combustor performance

Supersonic combustion of thermally cracked kerosene was experimentally investigated in two model supersonic combustors with different entry cross-section areas. Effects of entry static pressure, entry Mach number, combustor entry geometry, and injection scheme on combustor performance were systematically investigated and discussed based on the measured static pressure distribution and specific thrust increment due to combustion. In addition, the methodology for characterizing flow rate and composition of cracked kerosene was detailed. Using a pulsed Schlieren system, the interaction of supercritical and cracked kerosene jet plumes with a Mach 2.5 crossflow was also visualized at different injection temperatures. The present experimental results suggest that the use of a higher combustor entry Mach number as well as a larger combustor duct height would suppress the boundary layer separation near the combustor entrance and avoid the problem of inlet un- start.

Waverider configurations derived from general conical flowfields

A method based on the computational fluid dynamics (CFD) is presented for a flexible waverider's design. The generating bodies of this method could be any cones. In addition, either the leading edge or the profile of the scramjet's inlet is used as the waverider's definition curve, parameterized by the quadric function, the sigmoid function or the B-spline function. Furthermore, several numerical examples are carried out to validate the method and the relevant codes. The CFD results of the configurations show that all the designs are successful. Moreover, primary suggestions are proposed for practical design by comparing the geometrical and aerodynamic performances of the cone-derived waveriders at Mach 6.

A novel oblique detonation structure and its stability

Oblique detonation structures induced by the wedge in the supersonic combustible gas mixtures are simulated numerically. The results show that the stationary oblique detonation structures are influenced by the gas flow Mach number, and a novel critical oblique detonation structure, which is characterized by a more complicated wave system, appears in the low Mach number cases. By introducing the inflow disturbance, its nonstationary evolution process is illustrated and its stability is verified.

利用图象分析法求湍流剪切层的光学密度分布曲线

本文介绍利用一个Microneye Bullet摄象头,一台ApplellPlus微计算机及一台EpsonRX-80打印机从干涉条纹读出信息并由此求出剪切层中光学密度的方法,讨论了在编写计算机程序时考虑过的问题,从一张由湍流不均匀剪切流的实验中取得的Much-Zehnder干涉照片画出流场中的光学密度分布以及剪切层的扩展情况。最后还用同样方法画出激光通过剪切层后相位角偏差的均方根变化曲线。

Numerical investigation of dynamic stall of an oscillating aerofoil

The flow structure around an NACA 0012 aerofoil oscillating in pitch around the quarter-chord is numerically investigated by solving the two-dimensional compressible N-S equations using a special matrix-splitting scheme. This scheme is of second-order accuracy in time and space and is computationally more efficient than the conventional flux-splitting scheme. A 'rigid' C-grid with 149 x 51 points is used for the computation of unsteady flow. The freestream Mach number varies from 0.2 to 0.6 and the Reynolds number from 5000 to 20,000. The reduced frequency equals 0.25-0.5. The basic flow structure of dynamic stall is described and the Reynolds number effect on dynamic stall is briefly discussed. The influence of the compressibility on dynamic stall is analysed in detail. Numerical results show that there is a significant influence of the compressibility on the formation and convection of the dynamic stall vortex. There is a certain influence of the Reynolds number on the flow structure. The average convection velocity of the dynamic stall vortex is approximately 0.348 times the freestream velocity.

Numerical-simulation of low supersonic-flow around a sphere

The controlled equations defined in a physical plane are changed into those in a computational plane with coordinate transformations suitable for different Mach number M(infinity). The computational area is limited in the body surface and in the vicinities of detached shock wave and sonic line. Thus the area can be greatly cut down when the shock wave moves away from the body surface as M(infinity) --> 1. Highly accurate, total variation diminishing (TVD) finite-difference schemes are used to calculate the low supersonic flowfield around a sphere. The stand-off distance, location of sonic line, etc. are well comparable with experimental data. The long pending problem concerning a flow passing a sphere at 1.3 greater-than-or-equal-to M(infinity) > 1 has been settled, and some new results on M(infinity) = 1.05 have been presented.

High-speed streams from coronal holes and the accelerating mechanism of the solar-wind

In this paper, the general Mach number equation is derived, and the influence of typical energy forms in the solar wind is analysed in detail. It shows that the accelerating process of the solar wind is influenced critically by the form of heating in the corona, and that the transonic mechanism is mainly the result of the adjustment of the variation of the crosssection of flowing tubes and the heat source term.The accelerating mechanism for both the high-speed stream from the coronal hole and the normal solar wind is similar. But, the temperature is low in the lower level of the coronal hole and more heat energy supply in the outside is required, hence the high speed of the solar wind; while the case with the ordinary coronal region is just the opposite, and the velocity of the solar wind is therefore lower. The accelerating process for various typical parameters is calculated, and it is found that the high-speed stream may reach 800 km/sec.

爆轰驱动激波管缝合激波马赫数计算

采用数值求解的方法得出了氢氧爆轰驱动激波管的缝合状态参数。以空气为试验气体时，缝合激波马赫数随着H2摩尔浓度的增加而增加，H_2摩尔浓度达到90％左右时达到最大。当缝合马赫数较高时，需要考虑高温真实气体效应的影响，缝合激波马赫数较理想气体的高。以氢空气混合气为试验气体时，缝合激波马赫数较以空气为试验气体的小。通过调整驱动气体与被驱动气体的初始参数，可以得到即能恰好消除Taylor波又能缝合的运行状态。

不同形状微管道中的气体流动

不同形状微尺度管道（圆形、六边形、半圆形、不同宽高比的矩形）中的气体流动特性是微机电系统设计最为关心的问题之一．文中利用信息保存（IP）方法和直接模拟Monte Carlo（DSMC）方法进行研究，给出两种方法的计算结果相互符合，并与其它研究者的BGK模型方程计算结果进行了比较．对于微尺度管道中关心的低Mach数流动，IP方法的统计收敛效率明显优于DSMC方法,通过拟合IP和DSMC结果，给出了圆形、六边形、半圆形、不同宽高比的矩形截面情况下无量纲质量流率与等效Knudsen数的关系．

可压缩均匀各向同性湍流的直接数值模拟

采用8阶精度的中心差分格式及7阶精度的迎风偏斜格式对Reλ=72-153,Mt=0.2-0.7的均匀各向同性湍流进行了直接数值模拟，建立了湍流数据库。与他人的计算结果吻合十分理想，说明方法的有效性。数值结果表明，采用适当的迎风型差分格式可以克服起动问题（start-up problem)对湍流Mach数的限制，提高可计算的湍流Mach数，是可压湍流直接数值模拟的有效方法。分析了压缩性效应对湍流统计量的影响，发现压缩性使得湍动能的衰减加快。探讨了可压湍流中微激波产生的机理，对流场进行了标度律分析。发现在本文的Reynolds数和湍流Mach数条件下，流场中扩展自相似性仍然成立，同时发现压缩性对标度指数影响不大。

流体运动稳定性方程组椭圆特性的分析与应用

利用抛物化稳定方程(PSE)特征分析得知，原始扰动量的线性和非线性PSE整体来说为抛物型．利用PSE的次特征分析证明，对速度Ⅳ，在亚音速和跨音速区，线性PSE分别为椭圆型和双曲一抛物型；对速度u+口，在亚音速和跨音速区，菲线性PSE分别为椭圆型和双曲．抛物型(其中， U和“分别为主流方向的扰动和未扰流速度分量)．结论表明，流体运动稳定性方程组的“抛物化”简化。仅把信息的对流扩散传播抛物化，而保留了信息的对流扰动传播特性，PSE实质上是扩散抛物化稳定性方程组．根据特征次特征理论提出了消除PSE剩余椭圆特性的方法，所得结论对线性PSE已有结论一致，并给出了Mach数的影响．同时，进一步给出了消除非线性PSE的剩余椭圆特性的方法．

微尺度气体流动

了解微尺度气体流动特点是微机电系统设计和优化的基础。有关的研究可以上溯到20世纪初Knudsen的平面槽道流动质量流量的测量和Millikan的小球阻力系数的测量，实验结果揭示了稀薄气体效应即尺度效应对气体运动的重要影响。由于流动特征长度很小，微尺度气流经常处到滑流区甚至过渡领域，流动的相似参数为Knudsen数和Mach数。因此可以考虑利用相似准则，通过增大几何尺寸、减小压力的途径，解决微机电系统实验观测遇到的困难。为解决直接模拟Monte Carlo方法分析微机电系统中低速稀薄气流遇到的统计涨落困难，我们提出了信息保存法（IP），该方法能够有效克服统计散布，并已成功用于多种微尺度气流。

可压衰减湍流中被动标量场的直接数值模拟及谱分析

运用高精度迎风差分方法及8阶精度群速度控制型差分格式(GVC8)对初始 Reynolds数为72，初始湍流Mach数0．2-43．9的可压衰减湍流的流场及被动标量场进行了直接数值模拟，被动标量场的Schmidt数为2~10．通过不同计算网格及不同方法的数值计算对本文结果进行了验证．指出了可压衰减湍流中被动标量能谱存在的一1律，压缩性效应使得高波数成分衰减加快．对被动标量场进行了标度律分析，发现可压湍流中被动标量场具有扩展自相似性，而压缩性效应对被动标量标度指数的影响较小．发现随着被动标量Schmidt数增加，其标度指数的奇异性增强．

可压缩钝楔边界层转捩到湍流的直接数值模拟

直接数值模拟了来流Mach数为6的钝楔边界层在特定扰动(音速点附近壁面吹吸)下的转捩到湍流的整个过程.分析了平均速度剖面,脉动速度均方根及剪切应力等统计量,并与不可压理论及实验结果进行了比较.展示了转捩过程中的涡结构并分析了压力梯度对转捩的影响.

Starting Nozzle Flow Simulation Using K-G Two-Equation Turbulence Model

The starting process of two-dimensional nozzle flows has been simulated with Euler, laminar and k - g two-equation turbulence Navier-Stokes equations. The flow solver is based on a combination of LUSGS subiteration implicit method and five spatial discretized schemes, which are Roe, HLLE, MHLLE upwind schemes and AUSM+, AUSMPW schemes. In the paper, special attention is for the flow differences of the nozzle starting process obtained from different governing equations and different schemes. Two nozzle flows, previously investigated experimentally and numerically by other researchers, are chosen as our examples. The calculated results indicate the carbuncle phenomenon and unphysical oscillations appear more or less near a wall or behind strong shock wave except using HLLE scheme, and these unphysical phenomena become more seriously with the increase of Mach number. Comparing the turbulence calculation, inviscid solution cannot simulate the wall flow separation and the laminar solution shows some different flow characteristics in the regions of flow separation and near wall.