留言板

尊敬的读者、作者、审稿人, 关于本刊的投稿、审稿、编辑和出版的任何问题, 您可以本页添加留言。我们将尽快给您答复。谢谢您的支持!

姓名
邮箱
手机号码
标题
留言内容
验证码

基于期望最大算法的空间事件及异常值探测

刘劲宏 吴晨韵 徐劲 杜建丽 雷祥旭

刘劲宏, 吴晨韵, 徐劲, 杜建丽, 雷祥旭. 基于期望最大算法的空间事件及异常值探测[J]. 空间科学学报. doi: 10.11728/cjss2022.06.211124123
引用本文: 刘劲宏, 吴晨韵, 徐劲, 杜建丽, 雷祥旭. 基于期望最大算法的空间事件及异常值探测[J]. 空间科学学报. doi: 10.11728/cjss2022.06.211124123
LIU Jinghong, WU Chenyun, XU Jin, DU Jianli, LEI Xiangxu. Space Event and Outlier Detection Based on Expectation Maximization Algorithm (in Chinese). Chinese Journal of Space Science, xxxx, x(x): x-xx doi: 10.11728/cjss2022.06.211124123
Citation: LIU Jinghong, WU Chenyun, XU Jin, DU Jianli, LEI Xiangxu. Space Event and Outlier Detection Based on Expectation Maximization Algorithm (in Chinese). Chinese Journal of Space Science, xxxx, x(x): x-xx doi: 10.11728/cjss2022.06.211124123

基于期望最大算法的空间事件及异常值探测

doi: 10.11728/cjss2022.06.211124123
基金项目: 重庆市自然科学基金面上项目(CSTB2022 NSCQ-MSX1093)、重庆市教委科学技术研究项目(KJQN202200701)和中国博士后科学基金项目(2021 M703487)共同资助
详细信息
    作者简介:

    刘劲宏:E-mail:liu-jh@whu.edu.cn

  • 中图分类号: P171

Space Event and Outlier Detection Based on Expectation Maximization Algorithm

  • 摘要: 美国ELSET数据库提供的TLE数据是目前使用最广泛的数据,在热层大气密度反演、弹道系数估计、碰撞预警等领域具有重要作用。受空间环境扰动、空间事件以及TLE产生过程等共同影响,ELSET数据库包含大量亟待清理的异常值和识别的空间事件,例如发布错误的TLE、轨道根数异常和Bstar异常。现有方法在清理异常轨道根数时缺乏统一性,需要使用不同的技术,清理流程较为繁杂,并且仅适用于特定轨道区域的少数目标。为克服现有方法的弊端,提出了一种基于期望最大算法的滑动窗口–多项式拟合预报方法,对含有轨道机动的碎片以及受空间环境影响的碎片进行异常值与空间事件探测。研究表明,该方法能够灵活处理不同空间环境下的异常值与空间事件探测,具有普适性,适用于所有轨道碎片。

     

  • 图  1  基于EM算法的异常值及空间事件探测流(黑色圆点代表TLE观测值,红色圆点代表预报失败值)

    Figure  1.  Flowchart of outlier detection based on EM algorithm (Black dot represents the TLE observation value, and the red dot represents the prediction failure value)

    图  2  采用三种阈值策略清理轨道倾角,碎片NORAD ID13025(Ariane 1 R/B)

    Figure  2.  Inclination outlier detection by three strategies for NORAD ID13025 (Ariane 1 R/B)

    图  3  平运动序列中的异常值清理结果

    Figure  3.  Outlier detection in mean motion

    图  4  偏心率序列中的异常值初步清理结果

    Figure  4.  Preliminary results of outlier detection in eccentricity series

    图  5  各个子序列的异常值清理结果

    Figure  5.  Outlier detection results of each subsequence

    图  6  偏心率序列中的异常值最终清理结果

    Figure  6.  Final results of outlier detection in eccentricity series

    图  7  轨道倾角序列中的异常值清理结果

    Figure  7.  the Results of outlier detection in inclination

    图  8  Bstar序列中的异常值清理结果

    Figure  8.  the Results of outlier detection in Bstar

    图  9  空间碎片12446,23358,27386和40351轨道倾角异常值清理结果

    Figure  9.  Outlier detection of inclination for debris 12446, 23358, 27386 and 40351

  • [1] MULDOON A, ELKAIM G, WEEDEN B, et al. Improved orbital debris trajectory estimation based on sequential TLE Processing[C]//60 th International Astronautical Congress. Daejeon, South Korea, 2009
    [2] VALLADO D A. Fundamentals of Astrodynamics and Applications[M]. 4 th ed. California: Microcosm Press, 2013
    [3] LIDTKE A A, GONDELACH D J, ARMELLIN R. Optimising filtering of two-line element sets to increase re-entry prediction accuracy for GTO objects[J]. Advances in Space Research, 2019, 63(3): 1289-1317 doi: 10.1016/j.asr.2018.10.018
    [4] OSWEILER V P. Covariance Estimation and Autocorrelation of NORAD Two-Line Element Sets[D]. Wright-Patterso: Air Force Institute of Technology, 2006
    [5] LEGENDRE P, DEGUINE B, GARMIER R, et al. Two line element accuracy assessment based on a mixture of Gaussian laws[C]//AIAA/AAS Astrodynamics Specialist Conference and Exhibit. Keystone, Colorado: AIAA, 2006: 6518
    [6] VALLADO D A, CEFOLA P J. Two-line element sets - Practice and use[C]//63 rd International Astronautical Congress. Naples, Italy: International Astronautical Federation, 2012: 29-44
    [7] SAUNDERS A, SWINERD G G, LEWIS H G. Deriving accurate satellite ballistic coefficients from two-line element data[J]. Journal of Spacecraft and Rockets, 2012, 49(1): 175-184 doi: 10.2514/1.A32023
    [8] GEUL J, MOOIJ E, NOOMEN R. GOCE statistical re-entry predictions[C]//Proceedings of 7 th European Conference on Space Debris. Darmstadt, Germany: The ESA Space Debris Office, 2017
    [9] SONG W D, WANG R L, WANG J. A simple and valid analysis method for orbit anomaly detection[J]. Advances in Space Research, 2012, 49(2): 386-391 doi: 10.1016/j.asr.2011.10.007
    [10] GONDELACH D, LIDTKE A A, ARMELLIN R, et al. Re-entry prediction of spent rocket bodies in GTO[C]//26 th AAS/AIAA Space Flight Mechanics Meeting. 2016
    [11] ÁGUEDA A, AIVAR L, TIRADO J, et al. In-orbit lifetime prediction for LEO and HEO based on orbit determination from TLE data[C]//6 th European Conference on Space Debris. Darmstadt, Germany, 2013: 22-25
    [12] DOLADO J C, AIVAR-GARCIA L, MATÉ A A, et al. OPERA: a tool for lifetime prediction based on orbit determination from TLE data[C]//24 th International Symposium on Space Flight Dynamics. Laurel, MD, USA, 2004
    [13] PATERA R P. Space event detection method[J]. Journal of Spacecraft and Rockets, 2008, 45(3): 554-559 doi: 10.2514/1.30348
    [14] LIU J H, LIU L, DU J L, et al. TLE outlier detection based on expectation maximization algorithm[J]. Advances in Space Research, 2021, 68(7): 2695-2712 doi: 10.1016/j.asr.2021.07.013
    [15] 崔希璋, 於宗俦, 陶本藻, 等. 广义测量平差[M]. 2版. 武汉: 武汉大学出版社, 2009

    CUI Xizhang, YU Zongchou, TAO Benzao, et al. Generalized Survey Adjustment[M]. 2 nd ed. Wuhan: Wuhan University Press, 2009
    [16] 刘劲宏. 基于TLE数据的空间目标大气再入预报[D]. 武汉: 武汉大学, 2020

    LIU Jinghong. Prediction of Atmospheric Reentry of Space Objects Based on TLE Data[D]. Wuhan: Wuhan University, 2020
  • 加载中
图(9)
计量
  • 文章访问数:  11
  • HTML全文浏览量:  14
  • PDF下载量:  2
  • 被引次数: 0
出版历程
  • 收稿日期:  2021-11-19
  • 录用日期:  2022-04-11
  • 修回日期:  2022-05-12
  • 网络出版日期:  2022-11-05

目录

    /

    返回文章
    返回