首页> 外文学位 >Application of singular value decomposition to gravity field model development using satellite data.
【24h】

Application of singular value decomposition to gravity field model development using satellite data.

机译:奇异值分解在卫星数据重力场模型开发中的应用。

获取原文
获取原文并翻译 | 示例

摘要

Models of the Earth's gravity fields are necessary for satellite orbit determination. Obtaining a gravity field model involves solving a set of linear equations, where the unknowns are coefficients in a spherical harmonics approximation of the gravity field. Standard solution techniques require the linear system of equations to be full rank. In order to obtain more accurate gravity field models, more terms are added to the spherical harmonic series. These higher order terms can introduce observability problems and make the system of linear equations to be rank deficient. To prevent rank deficiency, a priori information based on Kaula's rule can be used, or surface gravity anomaly data can be added.;An alternative method to overcome the rank deficiency problem is to use the singular value decomposition (SVD). In this study, SVD was applied to solutions of various gravity fields of degree and order 70. The SVD routine from EISPACK was modified to accommodate the orthogonalized observation matrix, 27 Mw in size, in the available memory of a Cray YMP-864.;Three cases are studied. The first case uses only satellite tracking data, which results in a highly singular system of equations. The second case is satellite tracking and altimeter data, which is full rank but has problem caused by uneven data distribution. The third case is full rank system of the equations. The SVD solutions, which were obtained by zeroing 30 to 40 small singular values, fit the observation better than the gravity fields obtained by full rank solutions.;The orbits obtained by processing actual satellite observation data were compared with the orbits obtained using the SVD gravity fields. Using the SVD, a few gravity fields having the comparable total RMS fit to that of JGM-3 were produced. Some of these gravity fields showed better orbit fit in total RMS than JGM-3.;The SVD gravity fields were also evaluated by their degree variances and covariances. The geoid undulation differences of the SVD gravity fields and the JGM-3 were investigated. The geographically correlated orbit differences were generated to check the orbit difference due to gravity field difference of the SVD gravity fields and the JGM-3.;This study demonstrated that the SVD can be used to solve large gravity field problems regardless of the singularity of the system and the SVD solution provides an opportunity for better gravity fields when compared to the full rank solution.
机译:为了确定卫星的轨道,必须使用地球重力场的模型。获得重力场模型涉及求解一组线性方程,其中未知数是重力场的球谐函数近似中的系数。标准解决方案技术要求方程的线性系统必须完整。为了获得更准确的重力场模型,将更多项添加到球谐序列中。这些高阶项会引入可观察性问题,并使线性方程组的秩不足。为了防止秩不足,可以使用基于Kaula规则的先验信息,或者可以添加表面重力异常数据。克服秩不足问题的另一种方法是使用奇异值分解(SVD)。在这项研究中,将SVD应用于各种度数和阶数为70的重力场的解决方案。对EISPACK的SVD例程进行了修改,以在Cray YMP-864的可用内存中容纳大小为27 Mw的正交观测矩阵。研究了三种情况。第一种情况仅使用卫星跟踪数据,这导致方程组高度单一。第二种情况是卫星跟踪和高度计数据,虽然是全等级的,但由于数据分布不均而存在问题。第三种情况是方程的全秩系统。通过将30至40个小奇异值归零而获得的SVD解决方案比通过全秩解获得的重力场更适合于观测。;将通过处理实际卫星观测数据获得的轨道与使用SVD重力获得的轨道进行比较领域。使用SVD,产生了一些具有与JGM-3相当的总RMS的重力场。这些重力场中的一些在总RMS上显示出比JGM-3更好的轨道拟合。; SVD重力场也通过它们的度方差和协方差来评估。研究了SVD重力场和JGM-3的大地水准面波动差异。产生了与地理相关的轨道差异,以检查由于SVD重力场和JGM-3的重力场差异引起的轨道差异;这项研究表明,无论SVD的奇异性如何,SVD均可用于解决大型重力场问题。与全等级解决方案相比,该系统和SVD解决方案为更好的重力场提供了机会。

著录项

  • 作者

    Ahn, Kyu Suk.;

  • 作者单位

    The University of Texas at Austin.;

  • 授予单位 The University of Texas at Austin.;
  • 学科 Geodesy.;Engineering Aerospace.;Remote Sensing.
  • 学位 Ph.D.
  • 年度 1996
  • 页码 159 p.
  • 总页数 159
  • 原文格式 PDF
  • 正文语种 eng
  • 中图分类
  • 关键词

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号