CN-121328122-B - Satellite orbit forecasting method, system, equipment and medium based on gravitational field vectorization calculation model

CN121328122BCN 121328122 BCN121328122 BCN 121328122BCN-121328122-B

Abstract

The invention discloses a satellite orbit forecasting method, a system, equipment and a medium based on a gravity field vectorization calculation model, wherein the method comprises the steps of establishing a motion equation of a satellite under an inertial system; the method comprises the steps of constructing a gravity field rapid calculation model, dividing a gravity field gravity position-to-position gradient formula into the sum of a harmonic part and a field harmonic part, carrying out vectorization reconstruction on the sum to obtain a gravity field vectorization calculation model, obtaining the position and speed vector of a satellite at the initial moment, inputting the position and speed vector into a motion equation of the satellite under an inertia system, combining the gravity field vectorization calculation model, and solving by adopting a numerical integration method to obtain the satellite orbit. The gravity field vectorization calculation model provided by the invention is equivalent to the numerical value of the traditional circulation algorithm, so that the precision of the gravity field model is ensured to be free from loss, and the matrix operation is used for replacing the traditional double circulation structure, so that the calculation efficiency of the high-order gravity field is obviously improved, and the satellite orbit forecast calculation efficiency is improved on the basis of ensuring the precision.

Inventors

ZHANG LEI
LI KAI
Tang Chengpan
HU XIAOGONG
Shu Fengchun

Assignees

中国科学院上海天文台

Dates

Publication Date: 20260512
Application Date: 20251020

Claims (8)

1. The satellite orbit forecasting method based on the gravity field vectorization calculation model is characterized by comprising the following steps of: establishing a motion equation of the satellite under an inertial system; Constructing a gravity field rapid calculation model, namely dividing a gradient formula of gravity field gravity position versus position into a sum of a harmonic part and a field harmonic part, and carrying out vectorization reconstruction on the sum to obtain a gravity field vectorization calculation model; Acquiring the position and speed vector of a satellite at the initial moment, inputting the position and speed vector into a motion equation of the satellite under an inertial system, combining a gravity field vectorization calculation model, and solving by adopting a numerical integration method to obtain a satellite orbit; the construction process of the gravity field rapid calculation model is as follows: The gravity field gravity bit versus position gradient formula is written as the sum of the harmonic part and the field harmonic part and is expressed as follows: ; ; ; In the formula, Representation points The gravitational field gravitational force position at the position, And Respectively represent gravitational field gravitation positions A band harmonic part and a field harmonic part; And Respectively as dots Corresponding longitude, geocentric latitude and radius; 、 And Is a spherical harmonic coefficient; the normalized n-order Legend polynomial; An associated Legendre polynomial of n-order m-order after normalization; ; ; ; Vectorizing and reconstructing the harmonic part and Tian Xie parts of the gravity field gravity position-to-position gradient formula to obtain a gravity field vectorization calculation model, wherein vectorization of the harmonic part and the field harmonic part is expressed as follows: ; ; Wherein: ; ; ; ; ; ; ; ; ; ; ; ; In the formula of " "Hadamard product representing a matrix; Is that All 1 column vectors.
2. The method for predicting satellite orbits based on gravitational field vectorization calculation model as claimed in claim 1, wherein the equation of motion of the satellite under inertial system is expressed as follows: ; Wherein, the And The position and acceleration vectors of the mass center of the satellite under the inertial system are respectively represented, The radial direction of the satellite under the inertial system; Is the gravitational constant of celestial body; Is the earth mass; is the gravitational acceleration of the central celestial body, Is the gravitational field nonspherical gravitational acceleration under the inertial system, Is the sum of all perturbation accelerations except the gravitational field gravitational acceleration and gravitational field non-spherical gravitational acceleration.
3. The satellite orbit prediction method based on the gravity field vectorization calculation model according to claim 1, wherein the process of calculating the gravity field gravitational bit gradient by using the gravity field vectorization calculation model is as follows: Inputting a three-dimensional fixed system point, and introducing a global gravity field coefficient; Converting the point from a geodetic coordinate system to a geodetic spherical coordinate system to obtain the longitude, geodetic latitude and radial of the point ; Calculating Legend function and pair thereof according to recursive algorithm And calculating the partial derivatives of the components thereof 、、 And ; According to Matrix variables required in computing gravitational field vectorization computation models, including 、、、、、、 And ; Substituting the gravity field vectorization calculation model to calculate the gravity field gravity position gradient of the point.
4. The method for predicting satellite orbits based on gravitational field vectorization calculation model as claimed in claim 3, wherein in the process of solving to obtain satellite orbits by adopting numerical integration method: the gravitational field gravitational acceleration of the point is represented by the gravitational field gravitational acceleration of the point under the earth-fixed system; The gravitational field gravitational acceleration of the point is converted to inertial system through rotation matrix to obtain gravitational field non-spherical gravitational acceleration in inertial system.
5. A satellite orbit prediction system based on a gravity field vectorization calculation model, which is used for realizing the satellite orbit prediction method based on the gravity field vectorization calculation model as claimed in any one of claims 1-4, and the system comprises: The motion equation building module is used for building a motion equation of the satellite under an inertial system; The gravity field rapid calculation model construction module is used for constructing a gravity field rapid calculation model, namely dividing a gravity field gravity position-to-position gradient formula into a sum of a harmonic part and a field harmonic part, and carrying out vectorization reconstruction on the sum to obtain a gravity field vectorization calculation model; The satellite orbit determination module is used for acquiring the position and the speed vector of the satellite at the initial moment, inputting the position and the speed vector into a motion equation of the satellite under an inertial system, combining a gravity field vectorization calculation model, and solving by adopting a numerical integration method to obtain the satellite orbit.
6. An electronic device, comprising: A memory having a computer program stored thereon; a processor for loading and executing the computer program to implement a satellite orbit forecasting method based on a gravity field vectorized computing model as claimed in any one of claims 1-4.
7. A computer readable storage medium having stored thereon a computer program, which when executed by a processor implements a satellite orbit prediction method based on a gravity field vectorized computing model as claimed in any one of claims 1 to 4.
8. A computer program product having computer instructions stored therein, which when executed by a processor, implement the method of satellite orbit prediction based on a gravity field vectorized computing model as claimed in any one of claims 1 to 4.

Description

Satellite orbit forecasting method, system, equipment and medium based on gravitational field vectorization calculation model Technical Field The invention relates to the technical field of satellite orbit determination, in particular to a satellite orbit forecasting method, system, equipment and medium based on a gravity field vectorization calculation model. Background The provision of the satellite high-precision orbit forecast plays a key role in the fields of remote sensing, weather, satellite navigation systems and the like. Among satellite orbit dynamics models, the gravitational field model is one of the core components. It describes the gravitational potential field of the earth, the dominant force considered in orbit predictions. The earth's gravitational field is an objective reflection of the spatial distribution, movement and variation of earth materials, typically described in terms of spherical harmonic expansion, which contains thousands or even millions of terms. The accuracy of the gravitational field model directly affects the accuracy of the satellite orbit. With the progress of measurement technology and data processing technology, gravity detection tasks such as GRACE Follow-On and GOCE based On satellites become a mainstream means, and high-resolution and fine gravity field models are continuously and iteratively updated. Currently, the international mainstream organization has the capability of providing 2160 order gravity field models with even higher accuracy. In the precise orbit determination of the satellite, integral operation is carried out according to a given initial orbit, the space gravitation action is calculated based on an priori gravity field model, but the double nested circulation structure existing in the superposition calculation of the gravity field leads to the increase of the calculation complexity in the order of O (n 2), and the operation efficiency is obviously reduced. Because of the characteristics of satellite orbit prediction, the gravity field calculation model needs to be called repeatedly in the integration right function, and the traditional calculation scheme based on double circulation is serious in calculation time consumption, so that the performance influence of the extrapolation orbit of the logarithmic integration method is obvious. Single computation time is thus a key bottleneck affecting overall performance. Aiming at the technical bottleneck of the calculation of the gravitational field model in satellite orbit prediction, the prior research breaks through the performance limitation mainly through parallel calculation technology. The Showa east team provides a regional decomposition and aggregation communication strategy based on an MPI framework, and the Fourier transformation is combined to accelerate Legendre function calculation, li Gongwei adopts a Stokes integration method to construct a rapid calculation model, approximate calculation is realized through discretization expression, but slight deviation exists between the rapid calculation model and an original integration model, and Xiancai team designs a parallelization data processing flow based on OpenMP, so that the calculation speed is effectively improved. These parallel schemes all optimize the efficiency of the gravity field model calculation to a certain extent. However, the prior art solutions described above have the following structural drawbacks: 1. inherent in the algorithm level is the defect. Conventional loop computation involves a large number of redundant operations, such as order-wise double loop invocation of gravity field coefficient matrices. 2. Parallel computing achieves engineering limitations at the level of implementation. The existing parallel computing scheme has the problems of high adapting cost, strong architecture dependence and the like, and the application of heterogeneous accelerators such as GPU is limited by the research and development cost and energy efficiency ratio of a miniaturized parallel platform, so that the scheme has not been applied in large scale in the field of geodetic measurement. 3. Short plates of an algorithm optimization scheme. On one hand, the gravity field parameter inversion of the multi-focus high-dimensional equation set solving in the existing method is mainly used for improving the performance of a constructed normal equation and a solving process thereof, and an obvious short plate still exists for the efficiency optimization of single-time calculation of high-order gravitation. On the other hand, when the complexity of the gravity field calculation is simplified by using the approximate model, the loss of the precision of the gravity field model is also brought to a certain extent. In summary, the gravity field model of the prior art solution cannot give consideration to both calculation accuracy and efficiency, so that satellite orbit prediction cannot achieve the improvement of calculation efficiency and guarantee