测绘学报 ›› 2020, Vol. 49 ›› Issue (12): 1543-1553.doi: 10.11947/j.AGCS.2020.20200185

• 大地测量学与导航 • 上一篇    下一篇

精密重力测量中相对重力仪格值系数的贝叶斯估计方法

王林海1, 陈石1, 庄建仓2, 卢红艳1, 张贝1, 杨锦玲1,3   

  1. 1. 中国地震局地球物理研究所, 北京 100081;
    2. 情报与系统研究机构统计数理研究所, 东京 106-8569;
    3. 福建省地震局, 福建 福州 350003
  • 收稿日期:2020-05-15 修回日期:2020-08-26 发布日期:2020-12-25
  • 通讯作者: 陈石 E-mail:chenshi@cea-igp.ac.cn
  • 作者简介:王林海(1992-),男,博士生,研究方向为精密重力观测数据的处理方法。E-mail:wlh@cea-igp.ac.cn
  • 基金资助:
    国家重点研发计划(2018YFC0603502);国家自然科学基金(41774090;U1939205);中国地震科学实验场专项(2019CSES0105)

Bayesian estimation of the scale factor of relative gravimeter in precise gravity survey

WANG Linhai1, CHEN Shi1, ZHUANG Jiancang2, LU Hongyan1, ZHANG Bei1, YANG Jinling1,3   

  1. 1. Institute of Geophysics, China Earthquake Administration, Beijing 100081, China;
    2. The Institute of Statistical Mathematics, Research Organization of Information and Systems, Tokyo 106-8569, Japan;
    3. Fujian earthquake agency, China Earthquake Administration, Fuzhou 350003, China
  • Received:2020-05-15 Revised:2020-08-26 Published:2020-12-25
  • Supported by:
    The National Key Research and Development Program of China (No. 2018YFC0603502);The National Natural Science Foundation of China (Nos. 41774090;U1939205);The China Seismic Experiment Site (No. 2019CSES0105)

摘要: 相对重力仪的格值系数随时间会发生微小的变化,是影响精密重力测量精度的重要因素。通常需定期对相对重力仪进行专门的基线标定来评估仪器格值系数的变化。本文提出了一种利用重力观测数据进行格值系数评估的新方法,原理是利用测网中已知的多个绝对重力基准点作为先验约束,同时考虑仪器的非线性漂移变化,将格值系数作为超参数,基于贝叶斯原理和赤池贝叶斯信息准则(ABIC)估计最优值。通过对模拟数据的测试,该方法在高斯噪声和仪器非线性漂移等不确定性存在的情况下,可以获得格值系数的准确估计结果。对实测重力数据的测试表明:估计的格值系数与测量前在基线场标定的格值系数差值在5×10-5以内;而且相较于采用标定不准确的格值系数,该方法可以获得与绝对重力测量结果差异更小的平差重力值。本文研究结果为有效提高精密重力测量的效率和精度提供了方法保障。

关键词: 精密重力测量, 相对重力仪, 贝叶斯原理, 格值系数估计

Abstract: The scale factor of the relative gravimeter changes slightly with time, which is an important factor affecting the accuracy of precise gravity survey. It is necessary to regularly perform a special baseline calibration on the relative gravimeter to evaluate the change of the instrument’s scale factor. This study presents a new method that can be used to evaluate the scale factor based on the gravity observation data only. The principle is to use multiple absolute gravity datum stations known in the survey network as constraints, and to take into account the nonlinear drift of the instrument, and then to estimate scale factor as one of the hyper-parameters by Bayesian theory and Akaike’s Bayesian information criterion (ABIC). Through simulation data testing, this method can obtain the accurate estimation of scale factor in the presence of uncertainties such as Gaussian noise and instrument nonlinear drift. The test of the measured gravity data shows that: the differences between the estimated scale factors and the calibration results of baseline field before measurement are within 5×10-5, and compared with using the inaccurate calibrated scale factors, this method can obtain the better estimation of gravity values which are less different from the results of absolute gravimetry. The results of this study provide method guarantee for effectively improving the efficiency and accuracy of precise gravity survey.

Key words: precise gravity survey, relative gravimeter, Bayesian theory, scale factor estimation

中图分类号: