区域增强PPP技术及其在矿区变形监测应用研究

发布时间：2019-07-04 20:05

【摘要】：精密单点单位技术（PPP）不需要坐标已知的固定参考站，不受基线长度的制约，单台接收机即可在全球范围内进行静态或动态定位，并且能直接得到全球同精度的ITRF框架内的精密三维坐标。本文以大面积矿区高精度变形监测为背景，结合PPP关键技术展开相关研究，结合实例分析了基于区域增强PPP在矿区变形监测中的应用。本文主要研究成果和贡献如下： 1）提出质心拉格朗日插值法对PPP钟差和轨道数据进行预处理，该方法只需在拉格朗日插值的形式上稍作改变，简单方便。实验结果表明插值阶数为13时，稳定在毫米级。无论是最大误差还是RMS，Z方向插值效果均优于X、Y方向。 2）利用区域CORS站点联合周围IGS跟踪站，研究了基于IGU预报轨道实时估算卫星钟差的方法，并探讨了非差模糊度固定的精密单点定位方法。该方法对于提高PPP精度和可靠性具有一定的参考价值。 3）针对PPP的随机模型和函数模型，分别提出了先验信息定权法和EMD修正非差观测值法。该方法能够准确地确定观测值的权重和有效地减弱非建模误差的影响。 4）提出PSO优化BP神经网络，改变BP算法依赖梯度信息指导网络的方法来调整权重，利用PSO算法全局性搜索的特点，寻找最为合适的网络连接权值和阈值，提高BP神经网络的泛化能力。BP神经网络具有很强的函数逼近和模式识别能力，通过PSO优化BP神经网络能够提高修正动力学模型和进行异常检测的能力。 5）分析PPP历元间差分模型的实时解算精度，，当考虑对流延迟的影响，并顾及对流层水平梯度引起的信号延迟。结果表明可以显著消除系统误差，一定程度上提高定位精度。 6）分析了精密单点定位用于大面积矿区变形监测的精度，实例计算结果表明经过4小时的连续观测即可获得毫米级的定位精度。该技术因无需建立变形监测基准点，在大区域变形监测应用中有广泛的前景。为了确定监测点的瞬时历元的精确坐标，提出了基于速率信息和实测结果的抗差卡尔曼滤波模型。
文内图片：

图片说明：滑动式算法原理图
[Abstract]:Precision single point unit technology (PPP) does not need fixed reference station with known coordinates, and is not restricted by the length of baseline. a single receiver can locate static or dynamic around the world, and can directly obtain precise 3D coordinates in the framework of ITRF with the same precision in the world. In this paper, based on the background of high precision deformation monitoring in large area mining area, combined with the key technology of PPP, the application of regional enhanced PPP in deformation monitoring in mining area is analyzed with an example. The main research results and contributions of this paper are as follows: 1) the centrocentric Lagrangian interpolation method is proposed to preprocess PPP clock error and orbit data, which only needs to be changed slightly in the form of Lagrangian interpolation, which is simple and convenient. The experimental results show that the interpolation order is 13:00 and stable at millimeter level. Both the maximum error and the RMS,Z direction interpolation effect are better than X, Y direction. 2) using the regional CORS station and the surrounding IGS tracking station, the method of real-time estimation of satellite clock error based on IGU prediction orbit is studied, and the precise single point positioning method with fixed ambiguity is discussed. This method has certain reference value for improving the accuracy and reliability of PPP. 3) aiming at the stochastic model and function model of PPP, the prior information determination method and EMD modified non-difference observation method are proposed respectively. This method can accurately determine the weight of observation values and effectively reduce the influence of non-modeling errors. 4) PSO is proposed to optimize BP neural network, change the method that BP algorithm relies on gradient information to guide the network to adjust the weight, make use of the global search characteristics of PSO algorithm, find the most suitable network connection weight and threshold, and improve the generalization ability of BP neural network. BP neural network has strong function approximation and pattern recognition ability. Optimizing BP neural network by PSO can improve the ability of modifying dynamic model and abnormal detection. 5) the real-time solution accuracy of PPP calendar difference model is analyzed, when the influence of convective delay is taken into account, and the signal delay caused by tropospheric horizontal gradient is taken into account. The results show that the system error can be eliminated and the positioning accuracy can be improved to a certain extent. 6) the accuracy of precise single point positioning for deformation monitoring in large area mining area is analyzed. The calculation results show that the positioning accuracy of millimeter level can be obtained after 4 hours of continuous observation. This technology has a wide prospect in the application of deformation monitoring in large areas because it does not need to establish deformation monitoring reference points. In order to determine the exact coordinates of the instantaneous calendar of the monitoring points, a robust Kalman filter model based on rate information and measured results is proposed.
【学位授予单位】：中国矿业大学
【学位级别】：博士
【学位授予年份】：2013
【分类号】：TD325;P228.4

【参考文献】