Research On The Error Compensation Model Of Gravity Datum Network Data Processing

As one of the main tasks of geodesy,the ground gravity measurement not only provides basic data for obtaining the information of the earth ’s gravity field,but also strengthens people accurate cognition of the internal and external information of the earth on the basis of explaining the mechanism of the earth ’s movement.It provides accurate geological information for space exploration,military science and technology,geological disasters and national economic construction.The national gravity basic network is an important part of China ’s modern surveying and mapping system.Its realization is reflected by high-precision gravity control points distributed on the surface and with a certain density.It is impossible to complete largescale measurement tasks only by absolute gravity measurement.Relative gravity measurement has become an important means to establish a gravity control network.Different observation methods contain different errors.How to reasonably process gravity measurement data to obtain high-precision gravity values of control points is a key technology for establishing a gravity control network.This paper focuses on the solution method and principle of gravity basic network,and writes the program of data preprocessing,parameter estimation method and gravity network adjustment algorithm.The main work includes:(1)Based on the analysis and comparison of the three generations of gravity basic network established in China,the characteristics of the national gravity basic network in each period are analyzed and summarized,including the composition of the control network,the measurement method,the data processing method and the accuracy of the results.(2)According to the characteristics of the error source characteristics of the measuring instrument,the corrections of the ground gravity measurement observation data preprocessing are discussed in detail,including earth tide correction,pressure correction,polar motion correction,instrument effective height correction,light speed limited correction,groundwater influence correction and gravity vertical conversion height correction,and the correction formula is given.Considering the objective design of the observation instrument with systematic error,the adjustment model considering the influence of instrument parameters is discussed,and the parameter test method is given.(3)Considering the inevitable existence of abnormal observations in gravity observation data,different gross error control strategies are discussed based on data detection and robust estimation,and their applicability and reliability in different application scenarios are discussed.Based on the consideration of algorithm time,Laida criterion,RFP method and robust estimation are the optimal choices.Based on the positioning point of view,the Laida criterion,the adjacent variance ratio test and the RFP method are more suitable.Based on the fixed value angle,the adjacent variance ratio test and RFP method are preferred.From the control effect,the Laida criterion is the best,and the other schemes are relatively close.Considering comprehensively,in the adjustment calculation of large-scale measured gravity network,the robust estimation method is preferred for gross error processing.(4)In order to suppress the influence of abnormal error of observation data and consider the influence of different instrument types on the processing results of gravity observation data,Helmert robust variance component estimation is used to balance the contribution of different instrument observations.The experimental results of the 2000 gravity basic network measurement data show that using robust estimation compared to least squares estimation increases the average accuracy of gravity points from 7.64 to 6.77,an increase of about11%;The average accuracy of gravity points estimated by the variance component of absolute observation fixed weight and relative joint observation is 6.77 μGal,while the average accuracy of gravity points classified by gravity observation data of different instruments is 5.78μGal,which is about 15 % higher.