利用GNSS获取动态可降水量的理论与方法研究

发布时间：2018-06-21 06:14

  本文选题：天顶总延迟 + 可降水量　； 参考：《西南交通大学》2014年博士论文

【摘要】：目前GNSS水汽反演的研究方向主要集中于静态分段估计对流层,然后根据所得对流层转化为可降水量,并将其应用于降雨等气象方面的分析。由于天气每时每刻存在变化,静态分段估计可能不足以满足实时可降水量监测的要求,故本文主要针对如何提高动态对流层的估计精度进行了深入的研究,研究工作主要包括以下几个方面：首先研究了提高差分GPS近实时获取动态可降水量精度的方法。差分GPS解算对流层的过程中仅需要采用预报轨道就可以获取实时对流层,进而转换为实时可降水量。为提高解算精度,我们对基站和流动站(静止台站)给予以较高精度的先验坐标,并对该坐标进行1 cm左右的精度约束。利用PBO观测网络中的17个测站采用上述方法计算表明：如果基站与流动站的测站高差较小,可以获取2 mm精度的实时可降水量,可以满足气象预报的要求；反之则获取的实时可降水量精度较差,这是由于测站上空对流层的差值与测站的高差强相关。针对传统PPP分析了两种改进的PPP获取动态对流层延迟的精度。由于差分GNSS获取对流层的精度受到测站高差的约束,PPP的出现则克服了这个问题,同时摆脱了差分GNSS需要多台接收机同时观测的负担。由于观测量没有经过差分,传统PPP中的模糊度为浮点解,因此我们尝试固定模糊度,结果表明该方法给对流层带来1-2mm的影响,而且可以加速待估参数的收敛速度；此外,我们还尝试利用GPS与GLONASS组合PPP技术解算对流层,结果显示双星座解算对流层与单星座解算对流层在局部差值可达几厘米,显然在观测质量不佳时多星座可以增加解算对流层的可靠性。基于传统PPP提出了顾及内部和外部误差源对对流层解算影响的理论模型。分析了高阶电离层误差、卫星钟差和接收机钟差对对流层参数估计精度的影响,结果表明电离层非活跃期低纬度地区二阶电离层对对流层估计的影响可达2 mm,三阶电离层的影响不超过0.5mm；单一观测量发生钟跳可能会导致对流层在局部发生几厘米的跳变,针对这一现象本文给出了钟跳探测的方法,可应用于高精度实时对流层的解算；实时卫星钟差与最终卫星钟差差别可达数百米,故需要合理设置接收机钟差的过程噪声。采用动态PPP技术,首次分析了基于海、陆、空载体的移动GNSS水汽获取精度。研究表明利用船载、车载和机载GNSS的数据可以获取中误差约为lcm左右的动态对流层延迟,转换为动态可降水量的精度在2-3 mm左右。提出了附加约束条件的PPP算法。传统的动态PPP精度通常在分米级,为提高动态PPP精度,可以利用生产实践中存在的已知信息,这些信息包括内部解算参数之间的联系和外部的已知数据。基于这些已知信息本文提出了附加约束条件的PPP算法,通过实验数据分析表明该算法在一定的条件下可以提高参数的收敛速度,大大提高PPP的定位精度,同时也极大改善了对流层参数估计的精度。最终研究了基于PPP技术的实时动态可降水量获取方法,分析了利用GNSS对暴雨进行预警的可行性。利用预报轨道与实时卫星钟差对2014年3月底香港地区的12个CORS站的对流层进行了计算和分析,研究结果显示GPS测定的暴雨警告与香港天文台发布的警告高度吻合。
[Abstract]:At present, the research direction of GNSS water vapor inversion is mainly focused on static sectional estimation of troposphere, and then based on the conversion of the troposphere to precipitable water and applying it to the meteorological aspects of rainfall. Because of the change in the weather every moment, the static sectional estimation may not meet the requirements of real-time precipitation monitoring. This paper mainly focuses on how to improve the estimation accuracy of dynamic troposphere. The research work mainly includes the following aspects: first, the method of improving the precision of dynamic precipitable water in the near real time of differential GPS is studied. In the process of calculating the troposphere, the real time troposphere can be obtained only by using the forecast orbit in the process of the differential GPS solution. In order to improve the real-time precipitable water, we give higher precision prior coordinates to the base station and the mobile station (stationary station) and carry out the precision constraints of about 1 cm of the coordinate. 17 stations in the PBO observation network are used to calculate the above methods: if the height difference between the base station and the station is smaller, The real-time precipitable water of 2 mm precision can be obtained to meet the requirements of the weather forecast. On the other hand, the accuracy of the real-time precipitable water is poor, because the difference between the troposphere over the station and the height difference of the station is strong. For the traditional PPP analysis, the precision of the dynamic tropospheric delay obtained by the two improved PPP is analyzed. The difference GNSS is obtained. The accuracy of the troposphere is constrained by the height of the station. The emergence of the PPP overcomes this problem and gets rid of the burden that the differential GNSS needs to observe at the same time. Since the observational measurement does not pass the difference, the fuzzy degree in the traditional PPP is floating point solution, so we try to fix the fuzziness, and the result shows that the method brings 1 to the troposphere. The effect of -2mm can also accelerate the convergence rate of the parameters to be estimated. In addition, we also try to solve the troposphere using the combination of GPS and GLONASS, and the results show that the local difference can reach several centimeters in the troposphere and the single constellation solution troposphere. It is obvious that the multi constellation can increase the solution troposphere when the observed mass is poor. Reliability. Based on the traditional PPP, a theoretical model considering the influence of the internal and external error sources on the troposphere is proposed. The influence of high order ionospheric error, satellite clock difference and receiver clock difference on the estimation accuracy of troposphere parameters is analyzed. The results show that the influence of the two order ionosphere on the troposphere estimation in the ionosphere inactive phase of the low latitude area can be found. Up to 2 mm, the influence of the three step ionosphere is not more than 0.5mm; the occurrence of the clock jump in a single measurement may cause the troposphere to occur in a few centimeters of the troposphere. In this paper, a method of detecting the clock jump is given, which can be applied to the calculation of high precision real-time troposphere, and the difference of the clock difference between the real time satellite and the final satellite clock can reach hundreds of meters. It is necessary to reasonably set the process noise of the receiver clock difference. Using the dynamic PPP technology, the accuracy of the moving GNSS water vapor acquisition based on the sea, land and air carrier is analyzed for the first time. The study shows that the data from the ship load, the vehicle and airborne GNSS can obtain the dynamic opposite flow layer delay of about LCM, and the accuracy of the conversion to the dynamic precipitable water is in 2- About 3 mm. A PPP algorithm with additional constraints is proposed. The traditional dynamic PPP precision is usually at the decimeter level to improve the dynamic PPP accuracy. The known information in the production practice can be used. These information includes the connections between the internal calculation parameters and the external known data. Based on these known information, the additional constraint bars are proposed. The PPP algorithm of the part shows that the algorithm can improve the convergence speed of the parameters under certain conditions, greatly improve the positioning accuracy of PPP, and greatly improve the accuracy of the estimation of the troposphere parameters. Finally, the real-time dynamic water reduction method based on PPP technology is studied. The application of GNSS to the rainstorm is analyzed. The feasibility of the early warning is made. The troposphere of 12 CORS stations in Hongkong area at the end of March 2014 is calculated and analyzed using the forecast orbit and the real-time satellite clock difference. The results show that the rainstorm warning measured by GPS is in accordance with the warning height issued by the Hongkong observatory.
【学位授予单位】：西南交通大学
【学位级别】：博士
【学位授予年份】：2014
【分类号】：P228.4;P412.2

【参考文献】

中国期刊全文数据库 前10条

1 周命端;郭际明;郑勇波;许承权;;卫星天线相位中心偏移对GPS精密单点定位精度的影响研究[J];测绘通报;2008年10期

2 蒋虎;空基GPS遥感地球大气参数方法研究[J];测绘学报;2001年03期

3 陈俊勇;地基GPS遥感大气水汽含量的误差分析[J];测绘学报;1998年02期

4 许承权;范千;杜刚;;GPS天线相位转绕误差及其对GPS精密单点定位的精度影响分析[J];测绘与空间地理信息;2011年05期

5 宋淑丽,朱文耀,丁金才,廖新浩,程宗颐,叶其欣;上海GPS综合应用网对可降水汽量的实时监测及其改进数值预报初始场的试验[J];地球物理学报;2004年04期

6 程晓,徐冠华,周春霞,王清华,鄂栋臣;应用GPS资料反演南极大气可降水量的试验分析[J];极地研究;2002年02期

7 李薇;袁运斌;欧吉坤;李慧;李子申;;全球对流层天顶延迟模型IGGtrop的建立与分析[J];科学通报;2012年15期

8 楚艳丽;郭英华;张朝林;王迎春;;地基GPS水汽资料在北京“7·10”暴雨过程研究中的应用[J];气象;2007年12期

9 严豪健,张贵霞,郭鹏,刘敏,洪振杰;CHAMP观测资料的振幅反演初步结果[J];天文学报;2005年01期

10 刘敏;郭鹏;叶其欣;张洁;朱雪松;;上海地区地基GPS水汽三维层析技术和初步应用[J];天文学报;2010年03期

中国重要会议论文全文数据库 前1条

1 章红平;吕海霞;李敏;施闯;;电离层延迟二阶项改正模型化及其对精密定位影响[A];第二届中国卫星导航学术年会电子文集[C];2011年

，

本文编号：2047548