云南高美古湍流大气中激光短曝光光斑统计特性研究 下载: 1063次
Objective When laser propagates in the atmosphere, it is inevitable that a random change of the refractive index of the atmospheric turbulence causes laser spot broken and distorted, which seriously damages the laser beam quality. Under the conditions of weak- and strong-fluctuating media, the turbulence model has relatively rich theoretical research results. In view of the complexity of the turbulence itself, the current theory of light transmission in random media is not enough to perfectly present the far-field transmission characteristics of the laser spot in the strong-fluctuation region. As for the high-altitude laser transmission engineering applications, the spatial distribution of laser energy on the target surface and its temporal variation characteristics are vital for photoelectric system evaluation. However, the research of long-range high-altitude laser atmospheric transmission faces the following problems: first, there are few transmission tests in high-altitude areas; second, short-exposure laser spot data at high frame-rate sampling are rare; third, two-dimensional imaging methods to obtain more comprehensive laser statistical parameters are still lacking. In this study, we use a high-speed camera to perform laser spot tests at different distances in high-altitude areas. Using the laser short-exposure image data at sampling rate of 4000 frame·s -1, the change characteristics of short-exposure far-field laser spots in the turbulent atmosphere are qualitatively described, and the optical turbulence parameters such as ideal point scintillation indexes, aperture average factors, and inner scales are calculated. Our aim is to use the external field laser transmission test data to understand the spatial laser energy distribution on the target surface as well as its temporal variation characteristics under high-altitude conditions and to simultaneously determine its impact on the laser photoelectric system.
Methods The experiment is performed at an altitude of 3200 m in Gaomeigu, far from the underlying surface. The transmitting terminal is a 532 nm solid-state laser, while the receiving system is consisted of a high-speed CCD camera and a diffuse reflection screen. The camera is placed obliquely in front of the screen at ~9m, the transmitter is located in the balcony in the third floor of the Gaomeigu radar station at ~10m above the ground, and the receiver is located on the open ground at distances of 1, 2, 3, 4, and 4.7km from the transmitting terminal. The selected geographical location can ensure that most of the transmission paths are longer than 15m from the underlying surface and the maximum height is larger than 100m. In addition, the test is also performed in the canyon, far away from the underlying surface. The straight-line distance between two ends is ~2km, and the maximum link distance from the underlying surface is ~300m. The sampling rate of laser spot images is unified to 4000 frame·s -1, each image is 320pixel×240pixel, and the image accuracy is 8bit. For the acquired two-dimensional data of the laser spot, using the imaging method, the scintillation index under different receiving apertures is calculated, and subsequently polynomial fitting is used to obtain the ideal point scintillation index. With the former results, the refractive index structural parameters such as the path, aperture average factor, and inner scale can also be derived. Moreover, this method based on the calculation of the gray value of the light spot can also display the correlation coefficient between the power spectrum and its scaling rate as well as the spatial light intensity distribution.
Results and Discussion At a distance of 1km, with the playback of a short-exposure spot image at a sampling rate of up to 4000 frame·s -1, the broken laser spot shows a clear mesh-like morphological structure, and a few sharp bright spots are scattered. There are irregular shapes and different size holes inside the spot (Fig. 3). As time elapses, these flowing mesh-like structures and scattered bright spots undergo deformation and reconstruction. There is a certain unclear relationship between size of the reticulated texture and the scale of the turbulence. Using the light spot gray value data, the aperture average factor of different receiving radii on the target surface is obtained. The variation of the laser spot aperture average factor with the receiving aperture shows that the aperture average factor is not only a function of the aperture but also is related to the turbulence state (Fig. 6). The daily variation of
Conclusion The main purpose of designing this experiment is to find the transmission conditions far away from the underlying surface in high-altitude areas and to obtain the characteristics of the laser spot transmitted in high-altitude free atmospheric turbulence. Judging from the fact that there is no obvious “Mexican hat” structure in the daily variation trend of most transmission paths, the underlying surface on the transmission path has little influence, but there are still big differences between the real test conditions and high-altitude “free” atmosphere. In the future, we hope that we will have the opportunity to conduct high frame-rate laser spot image testing on higher and farther platforms to promote the understanding of laser transmission in free turbulent atmospheres.
1 引言
激光在大气中传输时,由于大气湍流折射率的实时随机变化,光斑形状发生破碎和畸变,表现为接收靶面上的光强出现随机变化,该现象通常被称为激光闪烁效应[1]。Tatarski[2]在Kolmogorov理论的基础上, 基于微扰理论,采用湍流谱分析方法建立了弱起伏条件下的激光大气传输理论,该理论进而发展为公认的激光光斑仿真算法模型。之后,有学者利用Markov近似方法,通过建立光强的统计矩方程,得到了强起伏条件下的闪烁均方差的渐近结果。Kerr 等[3]通过实验发现,在强湍流效应下聚焦光束在焦平面上的短曝光光斑也会破碎成一系列小光斑。由于湍流本身的复杂性,目前有关随机介质中的光传输理论尚不足以完美呈现中等至强起伏区域的激光光斑远场传输特性。因此,开展真实湍流大气中的激光传输光斑测试实验,获得光斑统计特性,对于了解特定环境下的激光传输规律、验证仿真模型以及指导工程实践均具有重要参考意义。
在湍流大气中的激光光斑特征研究方面,1998年饶瑞中等[4]利用50frame/s采样速率下的1km传输激光光斑数据,对被大气湍流退化的光斑形变特征进行了探索性研究,并计算分析了光强起伏的概率密度分布、功率谱,激光光斑的有效半径、最大半径及其质心漂移等特征参数的时间变化特性。同年,饶瑞中等[5]还采用数值仿真的方法,获得了湍流大气中的长曝光激光光斑能量分布,阐述了光斑的尺度测量及其特征描述等问题,并进一步结合实测数据,探究了被湍流大气退化的短曝光光斑的形变特征。2002年,饶瑞中[6]又针对准直激光束,采用数值仿真方法,探讨了用稳定半径、等效半径和锐度半径表征光斑质量的适用性问题。2005年,刘建国等[7]同样在1km的传输距离上,利用250frame/s采样速率下的漫反射激光光斑数据,分析了光斑的尺度特征和孔径平滑效应。之后,靖旭等[8]以500frame/s采样速率开展了类似的研究。2008年,Jiang等[9]在11.8km传输链路上利用相机和光电探测器获取的数据,分别计算了弱起伏和强起伏条件下激光光斑的概率密度函数、衰落统计量和高频频谱,最后给出了闪烁指数和空间折射率结构参数的日变化曲线。2020年,Ke等[10]在10.2km的距离上开展了激光传输实验,采用CCD相机以75frame/s采样速率获得了分辨率为320pixel×256pixel的远场光斑图像。在数字图像处理方法的帮助下,学者们有效地获取了光斑斑点区域。
在国际上,1994年,Butts等[11-12]利用ABLEX实验平台开展了高空10~15km的远距离水平光斑测试,测试距离最远达到了200km,旨在核实自适应光学系统能否很好地补偿高空湍流效应,其相机的采样速率只有30frame/s。之后不久,Washburn[13]利用更高精度和更高分辨率的CCD相机,在ABLE ACE机载平台上,开展了与ABLEX类似的光斑测试实验,旨在探究在激光水平长程传输过程中大气传输数值仿真模型的适用情况。但文章并未注明采样速率。2003年, Vorontsov等[14]开展了近水平2.3km光学链路上的激光传输测试,使用Dalsa CAD6-0256W型CCD在256pixel×256pixel分辨率下以760frame/s 的采样速率获取了激光短曝光闪烁图像,同时利用大气湍流参数测量设备获得了激光传输链路上的相干长度和折射率结构常数。同年,Nakamura等[15]在室内(100m)和露天(750m)条件下,利用NAC Image Technology公司的MEMRECAM fx K3高速相机以2000frame/s采样速率记录了激光传输光斑图像。
上述研究普遍存在采样频率低等问题,所公布的激光光斑数据量有限,不足以呈现湍流大气中的短曝光激光光斑特性。本文利用高速相机,以漫反射幕布为合作靶面,在3200m的高海拔地区,选择尽可能远离下垫面的地势,开展了1km至近5km的激光大气传输光斑测试实验,获得了大量采样速率达4000frame/s的激光光斑数据。并利用该数据,通过二维成像的方法,计算并统计分析了目标靶面上的光强闪烁指数、光强闪烁的孔径平均因子A、闪烁频谱及其标度指数、光强的空间分布和空间相关系数,以及测试光路上的大气湍流折射率结构常数
2 测试设备及场景
实验测试设备及测试区域的地形地貌分别如
激光光斑数据的采集时间为2019年6月20日至28日,每日上午10时至下午17时,采集期间天气晴好,激光发射功率约为1W,每间隔20min采集一组视频图像,每组时长8s,采样速率统一设置为4000frame/s,每幅图像尺寸为320pixel×240pixel,图像位数为8bit,曝光时间为245μs。
3 短曝光光斑基本特征
为了直观地展现短曝光光斑图像的典型特征,同时为了帮助理解短/长曝光图像之间的差异,
图 4. 普通手机相机拍摄的长曝光光斑图像
Fig. 4. Long-term exposure spot patterns taken by normal cellphone camera
由于人眼的视觉暂留效应,通常人们所看到的激光光斑均可视为长曝光图像的动态变化。只有将连续拍摄的短曝光光斑图像进行慢速回放,人眼才能真正看到短曝光图像的变化过程:前后两幅图像的高度相近,光斑纹理随着横向风向一侧微小移动,光斑结构随着时间逐渐变化,直至数十帧后才会完全改变。这些流动的网纹结构、散落的亮斑及其形变与重构等现象在短曝光图像中格外清晰,而网纹的粗细和大小与湍流的强度和尺度存在某种目前尚未明确的关联。
4 光斑统计分析
基于云南高美古高海拔区域远离下垫面洁净大气中的短曝光激光光斑数据,我们对激光经高原大气湍流后的光强闪烁效应及其统计参数,如闪烁指数、闪烁频谱、孔径平均因子和激光光斑的空间相关系数等,进行了详细的计算分析。同时,为了便于理解,根据闪烁指数给出了传输路径上的大气折射率结构常数
4.1 闪烁指数及相关量
4.1.1 闪烁指数的计算方法
闪烁指数通常用归一化光强起伏方差
式中:I为光强;<·>为统计平均。根据激光在大气中的传播理论,假设光传播路径上湍流强度均匀,弱起伏条件下球面波的点闪烁指数
式中:k为波数;L为光传输距离。由 (2) 式可以得到大气折射率结构常数
闪烁指数随探测面积的增大而减小的现象被称为孔径平滑效应,常用孔径平均因子A来表征,定义为一定接收孔径r内的闪烁指数
式中:A<1。
为了得到理想的点闪烁指数
图 5. 不同孔径下四阶多项式拟合的闪烁指数
Fig. 5. Scintillation indexes fitted with fourth-order polynomial under different apertures
图 6. 不同闪烁指数下孔径平均因子随接收孔径半径的变化
Fig. 6. Aperture averaging factor versus receiving aperture radius under different scintillation indexes
Churnside[17]给出了弱起伏条件下,圆形接收孔径r内的平均因子A与大气湍流内尺度l0的关系:
式中:λ为波长;r0为大气相干长度。通过(4)、(5)式可以获得大气湍流内尺度l0。
4.1.2 基于闪烁指数的计算结果
本文分别计算了激光经过1,2,3,4,4.7km路径的闪烁指数、孔径平均因子、折射率结构常数及内尺度,其中2km路径包含平坦高原与下垫面为山林沟壑的峡谷。通过投影幕上网格大小为10cm的标准网格图来标定光斑图像每个像元对应的实际光斑尺寸。以2019年6月20日1km传播路径下的光斑数据为例,下午1:00至4:00左右的激光闪烁统计数据如
表 1. 2019年6月20日1km激光传输实验中的闪烁特征结果
Table 1. Scintillation characteristics in 1km laser transmission experiment on Jun 20, 2019
|
4.2 光强闪烁的功率谱
光强起伏的功率谱特征反映了湍流介质折射率场起伏的统计特征。前人研究[18]表明:起伏功率谱可以划分为高频段和低频段两个部分。在低频段区间,功率谱呈近似常数分布;而在高频段区间,功率谱与频率呈指数变化关系。在高频区间,功率谱满足标度率[19]:
式中:W为功率谱;T0为标度指数;f为频率。理论结果[4]认为T0=-8/3。大量研究表明,根据实测数据计算的功率谱密度分布比理论上的更为复杂:由低频段的常数区开始,首先经历常数区与无标度区的缓变区,然后正式进入无标度区,随后经历无标度区和噪声区的缓变区,最后止于高频噪声区。以上5段区间可视为5条斜率不同的直线段。直线段具体函数[19]为
式中: ai为直线斜率;bi为截距;i为直线段的编号; lnfi-1≤lnf≤lnfi,其中f0为最低频率,f5为最高频率,f1、f2、f3、f4 为5段直线相邻处的交接频率值。本文重点关注无标度区的标度率,使用分段拟合方法求得标度指数。
图 10. 与
图 11. 不同传输距离下的功率谱。(a) 1km; (b)2km; (c)3km; (d)4km; (e) 4.7km
Fig. 11. Power spectra under different transmission distances. (a) 1km; (b) 2km; (c) 3km; (d) 4km; (e) 4.7km
4.3 光强闪烁的空间相关性
以光斑中心为原点,距离原点a处的光强起伏与原点处的光强起伏的空间相关系数Rn(a)可表示为
式中:I(0)为原点处的实时光强;
式中:Bn(a)为归一化的空间相关系数;σn(0)为原点处的光强起伏方差;σn(a)为距离原点a处的光强起伏方差。
图 12. 2km传输距离下归一化相关系数随轴偏移量的变化
Fig. 12. Normalized correlation coefficient versus axis offset under 2 km transmission distance
5 结论
统计分析了云南高美古高海拔区域远离下垫面不同传输距离上的短曝光激光光斑测试数据。结果表明:短曝光光斑内部存在大小不一的不规则孔洞,这些孔洞构成了短曝光光斑的网纹状结构,网纹间同时夹杂着由大气随机透镜效应引起的零星亮点。动态播放的光斑图像可以展示出网纹结构和散落亮点的移动、消散和重构过程。在激光光斑中,不同环围半径内平均光强的闪烁指数随环围半径的增大而逐渐减小,表现出典型的孔径平滑效应,且孔径平均因子不仅仅是孔径r的函数,还与湍流状态参数特别是内尺度相关。在弱起伏条件下,可以利用孔径平均因子计算出大气湍流内尺度,结果表明,实测的湍流内尺度大多为0.01~0.1m,符合理论预期;高原大气湍流的闪烁频谱呈现经典的五段线结构,截止频率大多为200~1000Hz,测试条件下的湍流谱标度指数的均值约为-7.2/3,且分布较为离散,其标准偏差为0.85。以2km测试数据为例,计算的激光光斑的空间相关尺度不到10pixel,且空间相关尺度与湍流强度呈正相关,该结果与湍流场本身的空间相关尺度与湍流强度呈正相关这一现象相一致
设计该实验的主要目的是在高海拔地区寻找远离下垫面的传输条件,获取高空自由大气湍流中传输的激光光斑特性。从大多数传输光路上的
[1] 饶瑞中. 现代大气光学[M]. 北京: 科学出版社, 2012.
Rao RZ. Modern atmospheric optics[M]. Beijing: Science Press, 2012.
[2] 塔塔尔斯基. 湍流大气中波的传播理论[M]. 温景嵩, 宋正方, 曾宗泳,等, 译. 北京: 科学出版社, 1978.
Tatarski VI. Wave propagation in a turbulent medium [M]. Wen J S, Song Z F, Zeng Z Y, et al, Transl. Beijing: Science Press, 1978.
[4] 饶瑞中, 王世鹏, 刘晓春, 等. 激光在湍流大气中的光强起伏与光斑统计特征[J]. 量子电子学报, 1998, 15(2): 3-5.
[5] 饶瑞中, 王世鹏, 刘晓春. 被湍流大气退化的激光光斑: 尺度测量与形变特征描述[J]. 光学学报, 1998, 18(4): 3-5.
[6] 饶瑞中. 湍流大气中准直激光束的光斑特征Ⅰ.特征半径[J]. 中国激光, 2002, 29(10): 889-894.
[7] 刘建国, 黄印博, 王英俭. 强湍流效应下激光大气传输短曝光光斑统计分析[J]. 强激光与粒子束, 2005, 17(3): 321-324.
[8] 靖旭, 吴毅, 侯再红, 等. 湍流大气中激光传输光强起伏特征研究[J]. 光学学报, 2010, 30(11): 3110-3116.
[10] Ke X Z, Jing Y K. Far-field laser spot image detection for use under atmospheric turbulence[J]. Optical Engineering, 2020, 59(1): 016103.
[12] Weaver L D, Butts R R. ABLEX high-altitude laser propagation experiment[J]. ]. Proceedings of SPIE, 1994, 2120: 30-42.
[13] Washburn D C. ABLE ACE: a high-altitude propagation program[J]. Proceedings of SPIE, 1997, 3065: 296-306.
[16] Andrews L C, Phillips R L, Hopen C Y, et al. Theory of optical scintillation[J]. Journal of the Optical Society of America A, 1999, 16(6): 1417-1429.
[17] Churnside J H. Aperture averaging of optical scintillations in the turbulent atmosphere[J]. Applied Optics, 1991, 30(15): 1982-1994.
[18] Strohbehn JW. Laser beam propagation in the atmosphere[M]. Heidelberg: Springer, 1978.
[19] 饶瑞中, 王世鹏, 刘晓春, 等. 实际大气中激光闪烁的频谱特征[J]. 中国激光, 1999, 26(5): 411-414.
[20] 王倩, 梅海平, 钱仙妹, 等. 近地面大气光学湍流空间相关特性的实验研究[J]. 物理学报, 2015, 64(11): 114212.
Wang Q, Mei H P, Qian X M, et al. Spatial correlation experimental analysis of atmospheric optical turbulence in the near ground layer[J]. Acta Physica Sinica, 2015, 64(11): 114212.
张骏昕, 梅海平, 沈刘晶, 黄印博, 罗福, 吴小龑. 云南高美古湍流大气中激光短曝光光斑统计特性研究[J]. 中国激光, 2021, 48(1): 0105001. Junxin Zhang, Haiping Mei, Liujing Shen, Yinbo Huang, Fu Luo, Xiaoyan Wu. Statistical Properties of Short-Exposure Laser Patterns in Turbulent Atmosphere of Gaomeigu, Yunnan Province[J]. Chinese Journal of Lasers, 2021, 48(1): 0105001.