GPS实时相对定位及其在地震预警中的应用

发布时间：2018-03-30 07:37

本文选题：GPS　切入点：实时　出处：《中国地震局地震研究所》2013年硕士论文

【摘要】：作为现代大地测量的主要观测手段之一，GNSS技术能获取地震时期高精度的地表形变信息，而利用高采样率GNSS数据能获得地震时期瞬时同震形变，从而间接获得地震波信号，为进一步研究地震震源破裂过程、地震波特性、地震震中反演等地震学问题提供新的数据来源。高精度GPS数据处理一般采用用后处理方式对RINEX文件进行处理。后处理方式可以提供高精度的定位结果，但却有时滞性、低时间分辨率等缺点。（1）采用IGS最终星历解算可保证较高的精度，，但至少有12d的时间延迟；即使采用IGS超快速星历也需要从GPS接收机取回足够历元数的RINEX文件后才能解算。（2）后处理软件通常一次性读入所有观测数据，计算机的性能将制约其处理的数据量，如Bernese和GAMIT中的后处理动态模块每次均只能处理3h左右的高采样率数据。（3）常规处理软件大多只能处理最高1Hz采样率数据，获取高频位移信号能力较弱。因此GPS后处理方式的滞后性不能满足地震等自然灾害短临预警对实时性、高时间分辨率的要求。 GAMIT/GLOBK的实时运动学模块trackRT采用逐历元相对定位模式，解算过程中消除了大部分的星历误差、卫星钟差和电离层延迟误差，精度大幅提高，可以实时处理常规采样率(30S至1Hz）和高采样率(1Hz以上)数据。针对GPS后处理方式存在的问题，本文研究了进行了如下研究：（1）GPS相对定位的理论基础，主要包括时间与坐标系统、相对定位的观测模型和参数估计方法，并分析了相对定位中的各种误差影响；（2）阐述了trackRT实时相对定位的原理，用trackRT对连续站数据进行了实时处理，并用其RIENX模拟版本模拟计算了2011年日本宫城Mw9.0级地震和2013年四川雅安Mw7.0级地震的GPS资料，获得了与后处理方式较吻合的动态定位结果，证明了trackRT能够实时捕获地震中的动态位移；（3）研究了利用GPS定位结果提取地震波信号的数学模型，包括形变波波初至时刻的提取、地震震中的反演以及地震发生时刻的确定。
[Abstract]:As one of the main observation methods of modern geodesy, GNSS technology can obtain high precision ground deformation information during earthquake period, while using GNSS data with high sampling rate can obtain instantaneous coseismic deformation during earthquake period, and thus indirectly obtain seismic wave signal. It provides a new source of data for further study of earthquake source rupture process, seismic wave characteristics, seismic epicenter inversion and other seismological problems. High-precision GPS data processing usually uses post-processing to process RINEX files. The post-processing method can provide high-precision positioning results, but it has time-delay. The IGS final ephemeris solution can guarantee high accuracy, but it has at least 12 days time delay. Even if the IGS superfast ephemeris is used, it is necessary to retrieve enough RINEX files from the GPS receiver to solve the problem.) the post-processing software usually reads all the observation data at one time, and the performance of the computer will restrict the amount of data it processes. For example, the post-processing dynamic module in Bernese and GAMIT can only deal with the high sampling rate data of about 3 hours each time.) most of the conventional processing software can only handle the highest 1Hz sampling rate data. The ability to obtain high-frequency displacement signals is weak, so the lag of GPS post-processing can not meet the requirement of real-time and high-time resolution for short-term and impending early warning of natural disasters such as earthquakes. The real time kinematics module trackRT of GAMIT/GLOBK adopts epoch-by-epochal relative positioning mode, which eliminates most of the ephemeris error, satellite clock error and ionospheric delay error, and improves the accuracy greatly. The data can be processed from 30 s to 1 Hz in real time and over 1 Hz at high sampling rate. In view of the problems existing in the post-processing mode of GPS, this paper studies the theoretical basis of GPS relative positioning as follows, mainly including time and coordinate system, observation model and parameter estimation method of relative positioning. The principle of real time relative positioning of trackRT is expounded, and the data of continuous station is processed in real time by trackRT. The GPS data of the Miyagi Mw9.0 earthquake in 2011 and the Ya'an Mw7.0 earthquake in Sichuan in 2013 have been simulated and calculated by using its RIENX simulation version. The dynamic location results are in good agreement with the post-processing method. It is proved that trackRT can capture the dynamic displacement in earthquake in real time.) the mathematical model of extracting seismic wave signal by using GPS positioning results is studied, including the first arrival time of deformation wave, the inversion of seismic epicenter and the determination of earthquake occurrence time.
【学位授予单位】：中国地震局地震研究所
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：P228.4;P315.7

【参考文献】