-
文中通过仿真计算和实验观测两种方式分析高速流场对星点偏折的影响。仿真计算主要采用光线追迹数值计算实现,物理观测中星点的中心位置计算采用灰度加权质心法。
-
通过CFD计算、NPLS观测等方法,可以获得观测实验段内的流场的密度分布,并通过公式(1)的Gladstone-Dale方程将其进一步转化为折射率分布,用以进行光线追迹计算可得:
$$ n = 1 + {K_{GD}}\rho $$ (1) 式中:$n$表示折射率;$\;\rho $为以${{{\rm{m}}^3}/ {\rm{kg}}}$为单位的密度值;${K_{GD}}$为与波长弱相关的Gladstone-Dale常数;波长单位取$ \text{μm}$时可按公式(2)求取。进而可得:
$$ {K_{GD}}(\lambda ) = 2.23 \times {10^{ - 4}}\left( {1 + \frac{{7.52 \times {{10}^{ - 3}}}}{{{\lambda ^2}}}} \right) $$ (2) 对已知折射率分布的流场,可将其进行格网划分,将每个细分格网视作各向同性的密度均匀单元。此时,可按照几何光学的原则进行光线追迹。光线追迹不能描述衍射或干涉现象,但是对于光束偏折和像差计算是有效方法。
光线追迹的实现基于公式(3)的光线方程可得:
$$ \frac{{{d}}}{{{\text{d}}s}}\left[ {n(\vec r)\frac{{{\text{d}}\vec r}}{{{\text{d}}s}}} \right] = \nabla n(\vec r) $$ (3) 式中:$\vec r$为光线传播路径的位置矢量;$n(\vec r)$表示折射率分布;${\text{d}}s$为沿传播路径上的一个递进步长。
具体地,光线矢量$\vec T$按照公式(4)来定义:
$$ \vec T = n\frac{{{\text{d}}\vec r}}{{{\text{d}}s}} = \frac{{{\text{d}}\vec r}}{{{\text{d}}t}} $$ (4) 式中:参数$t = \displaystyle\int {{{{\text{d}}s} \mathord{\left/ {\vphantom {{{\text{d}}s} n}} \right. } n}} $,也可写为${\text{d}}t = {{{\text{d}}s} \mathord{\left/ {\vphantom {{{\text{d}}s} n}} \right. } n}$。将其代入公式(3)可得:
$$ \frac{{{{\text{d}}^2}\vec r}}{{{\text{d}}{t^2}}} = n\nabla n = \frac{{\nabla {n^2}}}{2} $$ (5) 对于CFD计算获得的二维密度-折射率分布数据,公式(5)可以写作二维分量形式:
$$ \frac{{{{\text{d}}^2}R}}{{{\text{d}}{t^2}}} = D(R) $$ (6) 式中:$R = {(x,y)^{\rm{T}}}$, ${ {T}} = {({{{T}}_x},{{{T}}_y})^{{\rm{T}}}} = n{({{{\text{d}}x} / {{\text{d}}s}},{{{\text{d}{y}}} / {{\text{d}}s}})^{\rm{T}}}$;$D = n{({{\partial n} / {\partial x}},{{\partial n} / {\partial y}})^{\rm{T}}} = {{{{({{\partial {n^2}} / {\partial x}},{{\partial {n^2}} / {\partial y}})}^{\rm{T}}}} / 2}$。公式(6)给出了非均匀介质与传播路径之间的变化关系,由于该微分方程无法直接解析求解,故采用四阶Runge-Kutta法获得数值解。
一旦通过求解获得了光线的传播路径$C$,就可通过积分路径$C$上折射率$n$来计算光线在非均匀介质中的传输路径长度,即光程(Optical Path Length, OPL),可表示为:
$$ OPL = \int_C {n{\text{d}}s} $$ (7) 工程实践中,光程差(Optical Path Difference, OPD)是更重要且更易获得的指标,其定义可表示为:
$$ OPD = \int_C {n{\rm{d}}s} - \left\langle {\int_C {n{\rm{d}}s} } \right\rangle $$ (8) 式中:$\left\langle {} \right\rangle $表示所有光束的空间平均值。
-
受到高速流场扰动后的星点成像发生变化,扰动造成的星点中心偏移是影响天文定姿精度的核心要素。因此,获取星点在像平面坐标系的位置是姿态解算的重要步骤。
星点提取的主要方法包括质心法和拟合法。其中,质心法通常包含经典质心法、带阈值的质心法和灰度加权质心法;拟合方法主要有圆拟合和高斯拟合法。在众多星点中心提取算法中,灰度加权质心法既能达到较好的精度,同时具有较高可靠性,故文中采用其作为星点中心计算方法。该方法依公式(9)提取星点中心坐标$({x_c},{y_c})$:
$$ \left\{ \begin{gathered} {x_c} = \frac{{\displaystyle\sum\limits_{i = 1}^m {\displaystyle\sum\limits_{y = 1}^n {g\left( {{x_i},{y_j}} \right){x_i}} } }}{{\displaystyle\sum\limits_{i = 1}^m {\displaystyle\sum\limits_{j = 1}^n {g\left( {{x_i},{y_j}} \right)} } }} \\ {y_c} = \frac{{\displaystyle\sum\limits_{i = 1}^m {\displaystyle\sum\limits_{y = 1}^n {g\left( {{x_i},{y_j}} \right){y_j}} } }}{{\displaystyle\sum\limits_{i = 1}^m {\displaystyle\sum\limits_{j = 1}^n {g\left( {{x_i},{y_j}} \right)} } }} \\ \end{gathered} \right. $$ (9) 式中:$g({x_i},{y_j})$表示像平面上点$({x_i},{y_j})$的灰度值。
Observation experiment on star-light deflection of star-points under high-speed mixing flow
-
摘要: 天文导航是一种重要的飞行器自主导航手段。在高速飞行器上进行的观测,不可避免地会受到窗口外侧高速流场的扰动,使得星敏感器捕获的星点图像出现偏移、模糊等退化现象,影响天文定位定姿精度。对星图退化的计算和校正的研究多基于计算机仿真结果。文中建成了一座可在实验段中生成马赫2.5/3.5混合层结构的小型静风洞,以直径10 m的室内穹顶上的仿真星点为观测对象,透过实验段中不同位置的流场进行了星点观测和中心点解算,获得了星点图像受到流场扰动的数据,并将其与计算机仿真结果进行对比。结果表明:导航星光偏折量高于计算机仿真的估计值。在喷口近端,高速混合流场对星光偏折的扰动较大,垂直流场方向的偏折均值小于0.5″,沿流场方向偏离均值为3.85″,最大接近4.89″;在喷口远端,垂直方向星光偏折均值为−1.36″,沿流场方向偏折均值约−0.49″,最高达−2.69″。近端星光偏折变化幅度较小,稳定性较远端更强,有利于建模校正。该实验对校正仿真模型、优化高速流场下的天文定姿精度有着重要的意义。
-
关键词:
- 高速流场 /
- 马赫2.5/3.5混合层 /
- 小型风洞 /
- 室内穹顶 /
- 星光偏折
Abstract:Objective Celestial navigation is an important method of autonomous navigation. Astronomical observation of high-speed aircraft will be disturbed inevitably by the high-speed flow nearby the observation window, which causes the star maps degradation like displacement and blurring. And this will lead to a decrease in the accuracy of the stars center, which will have a direct effect on the accuracy of astronomical attitude determination. At present, most studies on the calculation and correction of star map degradation are based on computer simulation, whose results are greatly affected by the configuration of model parameters and may not be consistent with the real physical process. Therefore, it is necessary to construct the physical experimental observation conditions of the influence of high-speed flow on star-light deflection and to carry out experimental research. Methods A small static wind tunnel is built, which can generate a Mach 2.5/3.5 mixing layer structure in the test section. The calibrated simulated star-points on the indoor dome with a diameter of 10 m are measured through the high-speed flow, and the star centroids are extracted to collect the data of imaging displacements by the real flow. The data of star image disturbed by the flow field are obtained and compared with the computer simulation results. Results and Discussions The deflection by flow is greater than the estimated value of computer simulations. At the near end of the tunnel nozzle, the high-speed mixing layer makes a large star-light deflection. The mean deflection in the vertical direction of the flow field is less than 0.5″, and that in the direction of the flow field is 3.85″, and the maximum is close to 4.89″ (Fig.8). At the far end, the mean deflection in these two direction is −1.36″ and −0.49″ respectively (Fig.9). The variation of starlignt deflection at the near end is smaller and more stable than that at the far end, which is conducive for modeling correction (Fig.10). Conclusions A star-points observation system under the high-speed flow was constructed based on the indoor dome, and a Mach 2.5/3.5 mixed high-speed flow field was generated in the experimental observation section. The target star-points were observed from different observation positions, and the quantitative conclusion of the high-speed flow on star-points imaging disturbance was obtained for the first time by physical observation experiment. The results show that: 1) Star-light deflection is mainly concentrated in streamwise. This result is consistent with the expectation of the theoretical analysis; 2) The star-light deflection caused by the flow field at the near end of the nozzle is larger than that at the far end, but the variation range is smaller and more stable than that at the far end, which is conducive for modeling correction; 3) The absolute value of target starlight deflection caused by high-speed mixed flow is greater than the simulation result at both near and far end of the tunnel nozzle. The current work has proved the stability and effectiveness of the experimental system, which can provide an experimental basis to form a systematic understanding of the influence of flow structure on navigation starlight acquisition by the subsequent systematic observation under different altitude angles and azimuth angles, and provide experimental data of physical observation for simulation modeling. Then, a modified model of the influence of high-speed flow fields with different structures on starlight could be established, which may provide theoretical support for the suppression of aerodynamic influence and the deflection correction of air-cooled film in the astronomical observation of hypersonic vehicles. -
Key words:
- high-speed flow /
- Mach 2.5/3.5 mixing layer /
- small wind tunnel /
- indoor doom /
- star-light deflection
-
-
[1] Chen K, Pei S, Zeng C, et al. SINS/BDS tightly coupled integrated navigation algorithm for hyper-sonic vehicle [J]. Scientific Reports, 2022, 12(1): 10063-10069. doi: 10.1038/s41598-022-10063-9 [2] 陈冰, 郑勇, 陈张雷, et al. 临近空间高超声速飞行器天文导航系统综述[J]. 航空学报, 2020, 41 (08): 32–43. doi: 10.7527/S1000-6893.2019.23686 Chen Bing, Zheng Yong, Chen Zhanglei, et al. A review of celestial navigation system on near space hypersonic vehicle [J]. Acta Aeronautica et Astronautica Sinica, 2020, 41(8): 32-43. (in Chinese) doi: 10.7527/S1000-6893.2019.23686 [3] 张丽琴, 费锦东. 高速飞行器成像探测气动光学效应研究(特约)[J]. 红外与激光工程, 2020, 49(06): 228-232. doi: 10.3788/IRLA20201016 Zhang Liqin, Fei Jindong. Study on aero-optical effect of the imaging detection system of high speed flight vehicle (Invited) [J]. Infrared and Laser Engineering, 2020, 49(6): 20201016. (in Chinese) doi: 10.3788/IRLA20201016 [4] 陈冰, 郑勇, 陈张雷, 等. 基于校正后星图的高超声速飞行器天文定姿性能评估[J]. 测绘科学技术学报, 2021, 38(06): 585–591. Chen Bing, Zheng Yong, Chen Zhanglei, et al. Evaluation method of astronomical attitude determination performance on hypersonic vehicles based on corrected star map [J]. Journal of Geomatics Science and Technology, 2021, 38(6): 585-591. (in Chinese) [5] Wang M, Mani A, Gordeyev S. Physics and computation of aero-optics [J]. Annual Review of Fluid Mechanics, 2012, 44(1): 299-321. doi: 10.1146/annurev-fluid-120710-101152 [6] 史可天, 马汉东. 计算气动光学研究进展[J]. 空气动力学学报, 2019, 37(02): 186-192. doi: 10.7638/kqdlxxb-2017.0180 Shi Ketian, Ma Handong. Process in computational aero-optics [J]. Acta Aerodynamica Sinica, 2019, 37(2): 186-192. (in Chinese) doi: 10.7638/kqdlxxb-2017.0180 [7] Yang Bo, Fan Zichen, Yu He. Aero-optical effects simulation technique for starlight transmission in boundary layer under high-speed conditions [J]. Chinese Journal of Aeronautics, 2020, 33(7): 1929-1941. doi: 10.1016/j.cja.2020.02.015 [8] Hopkins K J, Porat H, Mcintyre T J, et al. Measurements and analysis of hypersonic tripped boundary layer turbulence [J]. Experiments in Fluids, 2021, 62(8): 03254. doi: 10.1007/s00348-021-03254-z [9] Lee S, Lee B J, Jeung I-S. Wavefront distortion due to the shock wave and boundary layer in the supersonic flow over a compression ramp [J]. Aerospace Science and Technology, 2021, 110: 6489. doi: 10.1016/j.ast.2021.106489 [10] Wilcox C C, Healey K P, Agena B D, et al. Air force research laboratory aero-effects laboratory optical metrology system and performance[C]//SPIE, 2020,11490: 11490A. [11] Chen H, Gao H, Zhang H. Integrated navigation approaches of vehicle aided by the strapdown celestial angles[C]//2019 IEEE 4th International Conference on Advanced Robotics and Mechatronics (ICARM), 2019: 911-917. [12] 杨博, 刘文东, 李旻珺. 盲目反卷积算法在高超流场星图复原中的应用[J]. 红外与激光工程, 2013, 42(8): 2231-2237. Yang Bo, Liu Wendong, Li Minjun. Application of blind deconvolution algorithm in hypersonic flow field star map restoration [J]. Infrared and Laser Engineering, 2013, 42(8): 2231-2237. (in Chinese) [13] Yang Bo, Fan Zichen, Yu He, et al. A new method for analyzing aero-optical effects with transient simulation [J]. Sensors (Basel, Switzerland), 2021, 21(6): 2199. doi: 10.3390/s21062199 [14] Chen Bing, Zheng Yong, Xu Bin, et al. Beam deflection correction model of wedge-shaped shock waves over hypersonic vehicles [J]. Infrared and Laser Engineering, 2021, 50(12): 20210182. (in Chinese) doi: 10.3788/IRLA20210182 [15] Zhao X, Yi S, Ding H. Influence of cooling film pressure on the imaging quality of a hypersonic optical dome [J]. Optical Engineering, 2020, 59(1): 013104. doi: 10.1117/1.OE.59.1.013104