The earth’s gravity field is a comprehensive reflection of the earth’s mass distribution and complex internal movement,and its high-precision time-varying information is used to study groundwater change,earthquake,glacier melting and so on.With the launch of gravity satellite in China,its precision and resolution will be greatly improved,which is of great significance to the development of earth related science and human life and production.Gravity satellite is an important means to measure time-varying gravity field.Since its launch in 2002,Gravity Recovery and Climate Experiment(GRACE)Gravity satellite has greatly improved the accuracy of people’s acquisition of time-varying information of Gravity field,and made up for the limitations of traditional methods such as limited scale and single means.The methods of gravity field inversion based on gravity satellite data mainly include dynamic method,short arc method and energy method,and each method has its own characteristics.The research content of this paper is the inversion of time-varying gravity field based on acceleration method.Acceleration method is a kind of inversion method which does not need to solve the complicated variational equation and its iterative operation,and does not lose the satellite vector information like the energy conservation method.Acceleration method basic principle is based on the satellite orbit,the interpolation velocity can be obtained on satellite total acceleration,after deducting the accelerometer observed after the conservative force,has been caused by a conservative force such as gravity acceleration,conservative forces,such as modeling the tide,tidal residual gravity acceleration,then you can estimate the coefficient of earth gravity field model are obtained.The accuracy of gravity field inversion depends on the accuracy of acceleration,which is affected by two factors: one is the measurement accuracy of satellite orbit or inter-satellite ranging data,and the other is the rationality of interpolation method.The main research contents of this paper are:(1)This paper introduces the main inversion methods of the earth’s gravitational field and the present situation of the earth’s gravitational field model at home and abroad,studies the time system and coordinate system involved in the acceleration method,gives the satellite force model,and derives the expression of the earth’s gravitational potential spherical harmonic function and the rectangular coordinate form of the Earth’s gravitational potential in detail.The principle of earth gravity field inversion by acceleration method is studied and its formula is derived.(2)The acceleration inversion method of gravity satellite requires the derivation of satellite motion acceleration data in the solution of gravity field,while gravity satellite can only get orbit data and velocity data by orbit data or velocity data,so the acceleration data need to be derived from orbit data or velocity data.In the process of deriving acceleration,high quality interpolation algorithm is needed.The accuracy of satellite motion acceleration is closely related to interpolation algorithm,so the study of interpolation method of satellite acceleration is a key technology of gravity satellite inversion.In this paper,Newton interpolation method and Chebyshev fitting method are used to derive satellite acceleration.The gravity satellite orbit is interpolated by downsampling method.The accuracy and adaptability of the two methods are compared and analyzed.The experimental results show that the orbital interpolation accuracy is related to the number of interpolation points.For The Newton interpolation method,the accuracy of 5-point interpolation is millimeter-level,and the interpolation effect is better.The9-point interpolation precision of Chebyshev polynomial fitting is millimeter level,and the interpolation effect is the best.In the comparison analysis of acceleration interpolation results,the accuracy of 5-point Newton interpolation is 0.66 mm /s2,and that of 9-point Chebyshev polynomial fitting is 0.48 mm /s2.It is concluded that chebyshev polynomial fitting method is the best interpolation method for gravity satellite acceleration,and the interpolation quality is better.(3)Build the acceleration method,the key research to the related algorithm,to solve the coefficient of acceleration and the construction of a right of observation equation in matrix problem,handle the actual gravity satellite data,get the time-varying gravity field model,and momentum integral method with the desires of the gravity field coefficient comparison,using the inside precision and outside precision evaluation way coefficient accuracy of gravity field are analyzed.The high order potential coefficient signal is smaller,while the low order potential coefficient signal is larger.In inner coincidence accuracy,the cumulative order error of acceleration method is 2E-4 m and that of dynamic integration method is 1E-3 m.In outer coincidence accuracy,the cumulative order error of acceleration method is 7E-4 m and that of dynamic integration method is 6E-3 m. |