基于多源数据和广义回归神经网络的ZWD预报模型

doi:10.11947/j.AGCS.2023.20220084

测绘学报 ›› 2023, Vol. 52 ›› Issue (9): 1492-1503.doi: 10.11947/j.AGCS.2023.20220084

基于多源数据和广义回归神经网络的ZWD预报模型

黎峻宇^1,2, 姚宜斌³, 刘立龙^1,2, 张豹³, 黄良珂^1,2, 曹利颖³

1. 桂林理工大学测绘地理信息学院, 广西桂林 541006;
2. 广西空间信息与测绘重点实验室, 广西桂林 541006;
3. 武汉大学测绘学院, 湖北武汉 430079

收稿日期:2022-02-23 修回日期:2023-05-16 发布日期:2023-10-12
作者简介:黎峻宇(1989-),男,博士,讲师,研究方向为GNSS近地空间环境监测。E-mail:yl_lijunyu@163.com
基金资助:
广西科技计划项目(2020GXNSFBA297145;GuikeAD23026177);国家自然科学基金(42064002;42074035);广西空间信息与测绘重点实验室基金(21-238-21-05);桂林理工大学科研启动基金(GUTQDJJ6616032)

A predicting ZWD model based on multi-source data and generalized regression neural network

LI Junyu^1,2, YAO Yibin³, LIU Lilong^1,2, ZHANG Bao³, HUANG Liangke^1,2, CAO Liying³

1. College of Geomatics and Geoinformation, Guilin University of Technology, Guilin 541006, China;
2. Guangxi Key Laboratory of Spatial Information and Geomatics, Guilin 541006, China;
3. School of Geodesy and Geomatics, Wuhan University, Wuhan 430079, China

Received:2022-02-23 Revised:2023-05-16 Published:2023-10-12
Supported by:
Guangxi Natural Science Foundation of China (Nos. 2020GXNSFBA297145;GuikeAD23026177);The National Natural Science Foundation of China (Nos. 42064002;42074035);Guangxi Key Laboratory of Spatial Information and Geomatics (No. 21-238-21-05);The Foundation of Guilin University of Technology (No. GUTQDJJ6616032)

摘要/Abstract

摘要： 对流层湿延迟是GNSS误差源较难改正的部分。主流的天顶湿延迟(ZWD)经验模型大多基于单源数据,如探空数据或再分析资料的一种,且通过预设模型函数表征ZWD在不同尺度上的变化,难以准确描述ZWD在不同尺度上的非线性复杂变化,精度有待进一步提高。针对此问题,基于多源数据和具有强大非线性逼近能力的广义回归神经网络(GRNN),构建了一种ZWD预报模型。首先,采用GRNN模型优化和降采样两种不同数据源的格网ZWD,并将其与探空ZWD合并,获取高质量的ZWD数据集;然后,根据ZWD受时空影响较大的特点和机器学习的特点,设计了GRNN训练模型的输入变量和输出变量;最后,采用后验寻优的方法确定模型参数,进而获得最优的预报模型。经探空ZWD验证,相比国际典范经验模型GPT2w,模型的精度提高了25.3%(以RMS计);相对同方法单源(探空)数据模型,精度改善了11.1%;且模型的预报精度具有良好的时空稳定性。此外,计算效率和PPP应用试验结果表明,模型的计算效率可满足GNSS实时应用的需求,且应用于PPP的改进效果优于GPT2w。本文所提方法无须预设模型函数即可获得较高的ZWD预报精度,为ZWD模型化提供了一种思路。

关键词: 对流层湿延迟, 多源数据, GRNN, 非线性逼近, 预报

Abstract: Tropospheric wet delay is a more difficult part of GNSS error sources to be corrected. Most of the approved empirical models of zenith wet delay (ZWD) are based on single-source data (i.e. radiosonde data or reanalysis data), and the variation patterns of ZWD on different scales are characterized by preset model functions, so it is difficult to accurately describe the nonlinearly complex variations of ZWD on different scales, and the accuracy needs to be further improved. To address this issue, a predicting ZWD model is constructed based on multi-source data with higher spatiotemporal resolution and a generalized regression neural network (GRNN) with strong nonlinear approximation capability. Firstly, grid ZWD of two different data sources is optimized and downsampled by a GRNN model, and merged with radiosonde ZWD to obtain high-quality ZWD dataset. Then, the input and the output vectors of the GRNN training model is designed according to the characteristics that ZWD is greatly affected by time and space and the characteristics of machine learning. Finally, a posteriori optimization method is used to determine the model parameters, and then the optimal forecasting model is obtained. Validated by the radiosonde ZWD, in comparison with the approved empirical GPT2w model and the single-source (i.e. radiosonde) data model with the same method, the accuracy of the proposed model is improved by 25.3% and 11.1% respectively in terms of RMS. And the accuracy of the proposed model has good spatiotemporal stability. In addition, the computational efficiency and PPP application experimental results show that the computational efficiency of the proposed model can meet the needs of GNSS real-time applications, and the improvement effect of PPP is better than that of GPT2w.The proposed model obtains high ZWD forecasting accuracy without setting the model function, which provides an idea for ZWD modeling.

Key words: tropospheric wet delay, multi-source data, GRNN, nonlinear approximation, prediction

中图分类号:

P228

黎峻宇, 姚宜斌, 刘立龙, 张豹, 黄良珂, 曹利颖. 基于多源数据和广义回归神经网络的ZWD预报模型[J]. 测绘学报, 2023, 52(9): 1492-1503.

LI Junyu, YAO Yibin, LIU Lilong, ZHANG Bao, HUANG Liangke, CAO Liying. A predicting ZWD model based on multi-source data and generalized regression neural network[J]. Acta Geodaetica et Cartographica Sinica, 2023, 52(9): 1492-1503.

参考文献

[1] SAASTAMOINEN J H. Atmospheric correction for the troposphere and the stratosphere in radio ranging satellites[J]. The Use of Artificial Satellites for Geodesy, 1972, 15: 247-251.
[2] 李博峰, 王苗苗, 沈云中, 等. 不同全球对流层天顶延迟产品在中国区域的比较[J]. 同济大学学报(自然科学版), 2014, 42(8): 1267-1272. LI Bofeng, WANG Miaomiao, SHEN Yunzhong, et al. Comparison of different global troposphere zenith path delay products in China[J]. Journal of Tongji University (Natural Science), 2014, 42(8): 1267-1272.
[3] YANG Fei, GUO Jiming, MENG Xiaolin, et al. Establishment and assessment of a zenith wet delay (ZWD) augmentation model[J]. GPS Solutions, 2021, 25(4): 148.
[4] LU Cuixian, LI Xingxing, ZUS F, et al. Improving BeiDou real-time precise point positioning with numerical weather models[J]. Journal of Geodesy, 2017, 91(9): 1019-1029.
[5] 伍冠滨, 陈俊平, 伍晓勐, 等. 基于非差非组合PPP-RTK的大气改正模型及其性能验证[J]. 测绘学报, 2020, 49(11): 1407-1418. WU Guanbin, CHEN Junping, WU Xiaomeng, et al. Modeling and assessment of regional atmospheric corrections based on undifferenced and uncombined PPP-RTK[J]. Acta Geodaetica et Cartographica Sinica, 2020, 49(11): 1407-1418.
[6] HOPFIELD H S. Two-quartic tropospheric refractivity profile for correcting satellite data[J]. Journal of Geophysical Research, 1969, 74(18): 4487-4499.
[7] COLLINS J P, LANGLEY R B. A tropospheric delay model for the user of the wide area augmentation system[M].Fredericton: Department of Geodesy and Geomatics Engineering, University of New Brunswick, 1997.
[8] LEANDRO R, SANTOS M, LANGLEY R. UNB neutral atmosphere models: development and performance[C]//Proceedings of 2006 National Technical Meeting of the Institute of Navigation. Monterey: ION, 2006: 564-573.
[9] MÖLLER G R, WEBER R, BÖHM J. Improved troposphere blind models based on numerical weather data[C]//Proceedings of the 26th International Technical Meeting of the Satellite Division of the Institute of Navigation. Nashville: ION, 2013: 2489-2495.
[10] KRUEGER E, SCHUELER T, HEIN G W, et al. Galileo tropospheric correction approaches developed within GSTB-V1[C]//Proceedings of ENC-GNSS2004. Rotterdam:[s.n.], 2004.
[11] KRUEGER E, SCHVLER T, ARBESSER-RASTBURG B. The standard tropospheric correction model for the European satellite navigation system Galileo[C]//Proceedings of 2005 General Assembly of International Union of Radio Science (URSI). New Delhi: URSI, 2005.
[12] SCHVLER T. The TropGrid2 standard tropospheric correction model[J]. GPS Solutions, 2014, 18(1): 123-131.
[13] BÖHM J, MÖLLER G, SCHINDELEGGER M, et al. Development of an improved empirical model for slant delays in the troposphere (GPT2w)[J]. GPS Solutions, 2015, 19(3): 433-441.
[14] YAO Yibin, HU Yufeng. An empirical zenith wet delay correction model using piecewise height functions[J]. Annales Geophysicae, 2018, 36(6): 1507-1519.
[15] MÖLLER G, WEBER R, BÖHM J. Improved troposphere blind models based on numerical weather data[J]. Navigation, 2014, 61(3): 203-211.
[16] ZHENG Fu, LOU Yidong, GU Shengfeng, et al. Modeling tropospheric wet delays with national GNSS reference network in China for BeiDou precise point positioning[J]. Journal of Geodesy, 2018, 92(5): 545-560.
[17] YAO Yibin, SUN Zhangyu, XU Chaoqian. Establishment and evaluation of a new meteorological observation-based grid model for estimating zenith wet delay in ground-based global navigation satellite system (GNSS)[J]. Remote Sensing, 2018, 10(11): 1718.
[18] SUN Zhangyu, ZHANG Bao, YAO Yibin. An ERA5-based model for estimating tropospheric delay and weighted mean temperature over China with improved spatiotemporal resolutions[J]. Earth and Space Science, 2019, 6(10): 1926-1941.
[19] 黄良珂, 朱葛, 彭华, 等. 全球ZWD垂直剖面滑动窗口的格网模型[J]. 测绘学报, 2021, 50(5): 685-694. DOI: 10.11947/j.AGCS.2021.20200515. HUANG Liangke, ZHU Ge, PENG Hua, et al. A global grid model for the vertical correction of zenith wet delay based on the sliding window algorithm[J]. Acta Geodaetica et Cartographica Sinica, 2021, 50(5): 685-694. DOI: 10.11947/j.AGCS.2021.20200515.
[20] LI Lei, XU Ying, YAN Lizi, et al. A regional NWP tropospheric delay inversion method based on a general regression neural network model[J]. Sensors, 2020, 20(11): 3167.
[21] DING Maohua. A second generation of the neural network model for predicting weighted mean temperature[J]. GPS Solutions, 2020, 24(2): 61.
[22] ZHENG D Y, HU W S, WANG J, et al. Research on regional zenith tropospheric delay based on neural network technology[J]. Survey Review, 2015, 47(343): 286-295.
[23] JIN Shuanggen, LI Z, CHO J. Integrated water vapor field and multiscale variations over China from GPS measurements[J]. Journal of Applied Meteorology and Climatology, 2008, 47(11): 3008-3015.
[24] THAYER G D. An improved equation for the radio refractive index of air[J]. Radio Science, 1974, 9(10): 803-807.
[25] HAGEMANN S, BENGTSSON L, GENDT G. On the determination of atmospheric water vapor from GPS measurements[J]. Journal of Geophysical Research: Atmospheres, 2003, 108(D21): 4678.
[26] WARE R, CARPENTER R, GVLDNER J, et al. A multichannel radiometric profiler of temperature, humidity, and cloud liquid[J]. Radio Science, 2003, 38(4): 1-13.
[27] CHEN Qinming, SONG Shuli, HEISE S, et al. Assessment of ZTD derived from ECMWF/NCEP data with GPS ZTD over China[J]. GPS Solutions, 2011, 15(4): 415-425.
[28] YUAN Qiangqiang, XU Hongzhang, LI Tongwen, et al. Estimating surface soil moisture from satellite observations using a generalized regression neural network trained on sparse ground-based measurements in the continental U.S.[J]. Journal of Hydrology, 2020, 580: 124351.
[29] ZHANG Bao, YAO Yibin. Precipitable water vapor fusion based on a generalized regression neural network[J]. Journal of Geodesy, 2021, 95(3): 36.
[30] LI Wei, YUAN Yunbin, OU Jikun, et al. A new global zenith tropospheric delay model IGGtrop for GNSS applications[J]. Chinese Science Bulletin, 2012, 57(17): 2132-2139.
[31] ASKNE J, NORDIUS H. Estimation of tropospheric delay for microwaves from surface weather data[J]. Radio Science, 1987, 22(3): 379-386.
[32] 胡羽丰. 全球高精度对流层延迟建模及其在地基GNSS技术中的应用研究[J]. 测绘学报, 2020, 49(4): 535. DOI: 10.11947/j.AGCS.2020.20190099. HU Yufeng. Research on the establishment of global tropospheric delay model with high precision and its applications in ground-based GNSS techniques[J]. Acta Geodaetica et Cartographica Sinica, 2020, 49(4): 535. DOI: 10.11947/j.AGCS.2020.20190099.
[33] RODRÍGUEZ J D, PÉREZ A, LOZANO J A. Sensitivity analysis of Kappa-fold cross validation in prediction error estimation[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2010, 32(3): 569-575.
[34] 姚宜斌, 余琛, 胡羽丰, 等. 利用非气象参数对流层延迟估计模型加速PPP收敛[J]. 武汉大学学报(信息科学版), 2015, 40(2): 188-192, 221. YAO Yibin, YU Chen, HU Yufeng, et al. Usingnon-meteorological parameters tropospheric delay estimation model for accelerating convergence of PPP[J]. Geomatics and Information Science of Wuhan University, 2015, 40(2): 188-192, 221.

基于多源数据和广义回归神经网络的ZWD预报模型

A predicting ZWD model based on multi-source data and generalized regression neural network

RichHTML

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

本文评价

[1]	孙根云, 王鑫, 安娜, 张爱竹. 基于多源遥感的大尺度高分辨率不透水面深度学习提取方法[J]. 测绘学报, 2023, 52(2): 272-282.
[2]	林珲, 吴贤宇, 潘家祎, 邹海波. 中国城市洪涝实时预报研究:现状与挑战[J]. 测绘学报, 2022, 51(7): 1306-1316.
[3]	袁俊军, 李凯, 唐成盼, 周善石, 胡小工, 曹建峰. 面向精密位置服务的低轨卫星轨道预报精度分析[J]. 测绘学报, 2022, 51(5): 640-647.
[4]	毛文婧, 王卫林, 焦利民, 刘安宝. 基于深度学习的中国连续空间覆盖PM_2.5浓度预报[J]. 测绘学报, 2022, 51(3): 361-372.
[5]	郭漩, 钱海忠, 王骁, 刘俊楠, 任琰, 赵钰哲, 陈国庆. 多源道路智能选取的本体知识推理方法[J]. 测绘学报, 2022, 51(2): 279-289.
[6]	黄良珂, 朱葛, 彭华, 陈华, 刘立龙, 姜卫平. 全球ZWD垂直剖面滑动窗口的格网模型[J]. 测绘学报, 2021, 50(5): 685-694.
[7]	王旭, 柴洪洲, 王昶, 种洋. 优选小波函数的小波神经网络预报GPS卫星钟差[J]. 测绘学报, 2020, 49(8): 983-992.
[8]	吕栋, 欧吉坤, 于胜文. 基于MEA-BP神经网络的卫星钟差预报[J]. 测绘学报, 2020, 49(8): 993-1003.
[9]	王旭, 柴洪洲, 王昶. 卫星钟差预报的T-S模糊神经网络法[J]. 测绘学报, 2020, 49(5): 580-588.
[10]	杨宇飞, 杨元喜, 胡小工, 唐成盼, 赵立谦, 徐君毅. 北斗三号卫星两种定轨模式精度比较分析[J]. 测绘学报, 2019, 48(7): 831-839.
[11]	范千, 方绪华, 许承权, 杨荣华. 变形监测数据预报的动态贝叶斯ELM方法[J]. 测绘学报, 2019, 48(7): 919-925.
[12]	唐炉亮, 戴领, 任畅, 张霞. 现实与赛博空间数据相结合的城市活动事件时空建模[J]. 测绘学报, 2019, 48(5): 618-629.
[13]	范昊鹏, 孙中苗, 张丽萍, 刘晓刚. 顾及映射函数误差的对流层延迟两步估计法[J]. 测绘学报, 2019, 48(3): 286-294.
[14]	韩松辉, 张国超, 张宁, 朱建青. AR模型中AO类异常值探测及其在GPS卫星钟差预报中的应用[J]. 测绘学报, 2019, 48(10): 1225-1235.
[15]	王冬霞, 郭睿, 谢金石, 胡小工, 袁运斌. 北斗新一代导航卫星的组合钟差预报模型及精度评估[J]. 测绘学报, 2018, 47(S0): 61-70.