VRS改正数算法研究及精度分析

发布时间：2018-04-28 06:51

本文选题：网络RTK + 误差改正数　；参考：《解放军信息工程大学》2013年硕士论文

【摘要】：本文针对VRS定位过程中各种误差改正数算法及精度展开研究，介绍了几种常规的空间内插模型，总结了多路径误差、接收机观测噪声、天线载波相位中心偏差三种非空间相关误差产生的原因及削弱方法。对轨道误差、电离层延迟误差、对流层延迟误差三种空间相关误差的改正数算法进行了研究。重点对电离层延迟误差和对流层延迟误差两种改正模型的可行性进行了实验分析。论文的主要研究内容及创新点可总结如下： 1、详细阐述了近十年来主要的四种空间相关误差内插模型：线性组合模型（LCM）、距离相关模型（DIM）、低阶表面拟合模型（LSM）和线性内插模型（LIM），并对其内插效果以及应用特点进行了总结。 2、对CORS(Continuous Operational Reference System)网络数据观测和处理过程中存在的多路径误差、接收机观测噪声、天线载波相位中心偏差三种非空间相关误差产生的原因及其削弱、消除的方法进行了介绍。 3、总结分析了常规内插方法在卫星轨道误差修正过程中存在的不足，介绍了ARS/VRS改正数计算模型，并对其进行了理论分析。 4、研究了VRS定位中电离层延迟改正数算法，针对电离层延迟双差残差与基线长度以及卫星高度角之间存在的变化关系进行了研究，，并将几种常规空间内插模型应用于电离层延迟改正数计算中，结合实验对各模型的稳定性和修正精度进行了比对分析，结果表明LIM模型效果较优。 5、针对中小尺度电离层行扰（MSTIDs）对CORS定位产生的影响进行了研究，提出一种附加了时域信息的电离层延迟误差预报模型，并通过实验数据对其进行了验证分析。实验结果表明：在MSTIDs活跃时段，预报模型与常规模型相比，其电离层延迟最大改正量可从0.2m提高到0.05m，大大降低了MSTIDs的影响。研究了模型预报周期的长短对模型精度的影响，结果表明：对于60km以上距离的中长尺度CORS网络，采用90s预报周期的电离层延迟预报模型可以将改正数计算精度由0.05~0.07m提高到大约0.02m。 6、针对对流层双差残差与基线长度、测站高差以及卫星高度角之间存在的变化关系进行了算例分析，结合实测数据将几种常规空间内插模型应用于对流层改正数计算中，并对各内插模型的实验结果进行了比对分析。 7、在对流层延迟改正数计算过程中，针对测站间高程差异对常规内插模型造成的高程方向偏差进行了研究和实验分析，提出一种基于常规空间内插模型的对流层延迟高程方向偏差修正模型，并结合实验数据对其可行性进行了研究。实验结果表明，对于高度角大于25°的目标卫星，在流动站与参考站高程差异接近于800m时，其内插精度能够保持在7~8cm以内，较之常规的LIM模型精度提高了大约1~2cm，能够准确的计算出流动站处对流层延迟双差值，在应用上是可行的。
[Abstract]:In this paper, based on the research of various error correction algorithms and accuracy in VRS positioning process, several conventional spatial interpolation models are introduced, and the multipath error and receiver observation noise are summarized. The causes and weakening methods of three kinds of non-spatial correlation errors of antenna carrier phase center deviation. The correction algorithms of three spatial correlation errors, orbit error, ionospheric delay error and tropospheric delay error, are studied. The feasibility of two correction models of ionospheric delay error and tropospheric delay error is analyzed experimentally. The main contents and innovations of this thesis can be summarized as follows: 1. Four main spatial correlation error interpolation models in recent ten years are described in detail: linear combination model, distance correlation model, low order surface fitting model and linear interpolation model. The interpolation effect and application characteristics are summarized. 2. The causes of non-spatial correlation errors, such as multipath error, receiver observation noise and antenna carrier phase center deviation, in the course of data observation and processing in CORS(Continuous Operational Reference System) network, are introduced, and the methods to eliminate them are also introduced. 3. The shortcomings of the conventional interpolation method in the correction of satellite orbit error are summarized and analyzed. The calculation model of ARS/VRS correction is introduced, and the theoretical analysis is carried out. 4. The algorithm of ionospheric delay correction in VRS positioning is studied, and the relationship between the residual of ionospheric delay and the length of the baseline and the satellite altitude angle is studied. Several conventional spatial interpolation models are applied to the calculation of ionospheric delay corrections. The stability and correction accuracy of each model are compared and analyzed by experiments. The results show that the LIM model is more effective. 5. The influence of mesoscale ionospheric disturbances (MSTIDs) on CORS location is studied, and a prediction model of ionospheric delay error with time-domain information is proposed, which is verified and analyzed by experimental data. The experimental results show that the maximum correction of ionospheric delay can be increased from 0.2 m to 0.05 m in the active period of MSTIDs, which greatly reduces the influence of MSTIDs. The effect of the length of the model prediction period on the accuracy of the model is studied. The results show that the ionospheric delay prediction model with 90 s prediction period can improve the accuracy of correction calculation from 0.05 ~ 0.07m to about 0.02m for the meso-long scale CORS network with the distance above 60km. 6. An example is given to analyze the relationship between tropospheric double difference residuals and baseline length, station height difference and satellite altitude angle. Several conventional spatial interpolation models are applied to the calculation of tropospheric corrections in combination with measured data. The experimental results of each interpolation model are compared and analyzed. 7. In the course of calculating tropospheric delay correction, the deviation of elevation direction caused by the conventional interpolation model is studied and experimentally analyzed according to the height difference between stations. A tropospheric delayed elevation deviation correction model based on conventional spatial interpolation model is proposed, and the feasibility of the model is studied in combination with experimental data. The experimental results show that when the height difference between mobile station and reference station is close to 800m, the interpolation accuracy of target satellite with altitude greater than 25 掳can be kept within 7~8cm. Compared with the conventional LIM model, the accuracy of the model is improved by about 1 ~ 2 cm, and the double difference of tropospheric delay at the mobile station can be calculated accurately, which is feasible in application.
【学位授予单位】：解放军信息工程大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：P228.4

【参考文献】