Ultra-low Orbit Guarantee Method, System and Equipment for Low Earth Orbit Satellites

Abstract

The present application discloses an ultra-low orbit guarantee method, system and device for a near-earth satellite. The method includes: determining the satellite orbit elements at the current moment based on the atmospheric drag perturbation term, the conventional perturbation term and the small thrust perturbation term, and inputting the satellite orbit elements at the current moment into the orbit mechanics model; determining the first velocity position parameters of the satellite in the ICRS coordinate system at the prediction moment based on the orbit mechanics model and the satellite orbit elements, and converting the first velocity position parameters into the ECEF coordinate system to obtain the second velocity position parameters at the prediction moment; inputting the time feature, satellite state feature, space environment feature at the current moment, and the second velocity position parameters into the orbit correction model to obtain the target error prediction value; adjusting the second velocity position parameters based on the target error prediction value to obtain the target velocity position parameters of the satellite at the prediction moment; and performing an orbit decay risk warning based on the target velocity position parameters of the satellite at the prediction moment.

Description

Technical Field

[0001] This application relates to the field of space environment detection technologies, particularly to the field of satellite detection technologies, and especially to a method, system, and device for ensuring the ultra-low orbit of a near-earth satellite. Background Art

[0002] Ultra-low orbit satellites (also known as near-earth satellites) operate in orbits between 150 km and 300 km from the Earth's surface. Due to factors such as a large number of satellites and an increased impact of the atmospheric environment, ultra-low orbit satellites now face challenges such as rapid orbital decay, increased collision risks, and increased difficulty in attitude control.

[0003] Generally, due to the resistance of the Earth's atmosphere, the orbit of an ultra-low orbit satellite will gradually decay, eventually causing the satellite to re-enter the Earth's atmosphere and burn up. This not only results in the loss of expensive satellite equipment but may also pose a threat to ground facilities and personnel safety. Currently, the orbit management of most satellites adopts a "post-response" strategy, that is, orbit raising operations are carried out after the satellite orbit decays to a certain extent.

[0004] However, when encountering extreme space weather such as solar flares and geomagnetic storms, the atmospheric density will change significantly. In such scenarios, when the satellite is at a lower position, it will be unable to keep the thruster on for a long time due to high air resistance, small ion thruster thrust, or insufficient power on the satellite, which will then lead to a greater risk of satellite crash. Therefore, if the atmospheric density and the speed and position of the satellite cannot be accurately predicted in advance, in extreme space weather, after the satellite orbit decays to a certain extent, it will be unable to smoothly raise the orbit due to large air resistance and will eventually crash. Based on this, it is crucial to build a perfect ultra-low orbit guarantee scheme and system to ensure that the satellite can operate stably in orbit and reduce the occurrence of crashes. Summary of the Invention

[0005] In view of this, the embodiments of this application provide a method, system, device, medium, and product for ensuring the ultra-low orbit of a near-earth satellite, which can accurately predict the atmospheric density and the speed and position of the satellite in advance, and perform an early warning of the orbit decay risk based on the target speed and position parameters of the satellite at the predicted moment, so as to prevent the satellite orbit from decaying significantly and then descending to a lower position.

[0006] In a first aspect, an embodiment of the present application provides a method for ensuring an ultra-low orbit of a near-earth satellite. The method includes: inputting historical environmental parameters corresponding to a target prediction period and M target prediction values into a first meta-model, and allocating M first weight values to the M target prediction values through the first meta-model, where the M target prediction values are output by M basic prediction models based on the historical environmental parameters corresponding to the target prediction period; using the M first weight values to perform weighted summation on the M target prediction values to determine the atmospheric density of the target prediction period; based on the real-time windward mass ratio, drag coefficient, atmospheric density of the satellite at the current moment, as well as the speed scalar and speed vector of the satellite relative to the atmosphere, determining the atmospheric drag perturbation term of the satellite at the current moment, where the current moment is within the target prediction period; based on the atmospheric drag perturbation term, conventional perturbation term, and small thrust perturbation term, determining the satellite orbit elements at the current moment, and inputting the satellite orbit elements at the current moment into an orbit mechanics model; based on the orbit mechanics model and the satellite orbit elements, determining the velocity position parameters of the satellite at the prediction moment in the geocentric inertial rectangular ICRS coordinate system to obtain first velocity position parameters, and converting the first velocity position parameters into the earth-fixed rectangular ECEF coordinate system to obtain second velocity position parameters at the prediction moment; inputting the time feature, satellite state feature, space environment feature, and second velocity position parameters at the current moment into an orbit correction model so that the orbit correction model evaluates the error of the second velocity position parameters to obtain a target error prediction value; adjusting the second velocity position parameters based on the target error prediction value to obtain the target velocity position parameters of the satellite at the prediction moment; performing an orbit decay risk warning based on the target velocity position parameters of the satellite at the prediction moment.

[0007] Second aspect, an ultra-low orbit guarantee system for a near-earth satellite provided by an embodiment of the present application includes: a weight allocation module, configured to input historical environment parameters corresponding to a target prediction period and M target prediction values into a first meta-model, and allocate M first weight values to the M target prediction values through the first meta-model, where the M target prediction values are output by M basic prediction models based on the historical environment parameters corresponding to the target prediction period; a determination module, configured to perform weighted summation on the M target prediction values by using the M first weight values to determine the atmospheric density of the target prediction period; the determination module is further configured to determine the atmospheric drag perturbation term of the satellite at the current moment based on the real-time windward mass ratio, drag coefficient, atmospheric density of the satellite at the current moment, and the speed scalar and speed vector of the satellite relative to the atmosphere, where the current moment is within the target prediction period; the determination module is further configured to determine the satellite orbit elements at the current moment based on the atmospheric drag perturbation term, the conventional perturbation term, and the small thrust perturbation term, and input the satellite orbit elements at the current moment into an orbit mechanics model; a coordinate system conversion module, configured to determine the velocity position parameters of the satellite at the prediction moment in the geocentric inertial rectangular ICRS coordinate system based on the orbit mechanics model, obtain the first velocity position parameters, and convert the first velocity position parameters into the earth-fixed rectangular ECEF coordinate system to obtain the second velocity position parameters at the prediction moment; an error evaluation module, configured to input the time characteristics, satellite state characteristics, space environment characteristics, and the second velocity position parameters at the current moment into an orbit correction model, so that the orbit correction model evaluates the error of the second velocity position parameters to obtain a target error prediction value; an orbit correction module, configured to adjust the second velocity position parameters based on the target error prediction value to obtain the target velocity position parameters of the satellite at the prediction moment; a risk warning module, configured to perform an orbit decay risk warning based on the target velocity position parameters of the satellite at the prediction moment.

[0008] Third aspect, an electronic device provided by an embodiment of the present application includes: a processor and a memory storing computer program instructions; when the processor executes the computer program instructions, the steps of the ultra-low orbit guarantee method for a near-earth satellite as in the first aspect are implemented.

[0009] Fourth aspect, a computer-readable storage medium provided by an embodiment of the present application has computer program instructions stored thereon, and when the computer program instructions are executed by a processor, the steps of the ultra-low orbit guarantee method for a near-earth satellite as in the first aspect are implemented.

[0010] Fifth aspect, an embodiment of the present application provides a computer program product, which is stored in a non-volatile storage medium, and when the computer program product is executed by a processor, the steps of the ultra-low orbit guarantee method for a near-earth satellite as in the first aspect are implemented.

[0011] In a sixth aspect, an embodiment of the present application provides a chip, which includes a processor and a communication interface. The communication interface is coupled to the processor, and the processor is configured to run programs or instructions to implement the steps of the method for ensuring the ultra-low orbit of a near-earth satellite as described in the first aspect.

[0012] The present application provides a method, system, device, medium and product for ensuring the ultra-low orbit of a near-earth satellite. Accurate atmospheric density is a prerequisite for accurately predicting satellite velocity and position parameters. Therefore, the present application can combine the prediction results of multiple basic prediction models to improve the overall prediction performance of the first meta-model, make up for the deficiencies of using a single basic prediction model for atmospheric density prediction, reduce the overall prediction error when predicting atmospheric density under different environmental conditions, and thus obtain a higher-accuracy atmospheric density. Based on this, when determining the atmospheric drag perturbation term using the atmospheric density, the accuracy of the atmospheric drag perturbation term can be improved, and further, when determining the satellite orbit elements using the atmospheric drag perturbation term, the calculation accuracy of the satellite orbit elements can be improved. Moreover, by combining the small thrust perturbation term, the atmospheric drag perturbation term and the conventional perturbation term, the actual force situation of the satellite orbit can be more comprehensively reflected, so that the orbital mechanics model can analyze the orbit change of the satellite under the action of multiple forces according to the actual force situation, and further improve the accuracy of the satellite orbit elements. After that, considering the accuracy of the predicted value output by the orbital mechanics model, which is affected by the satellite state and the satellite space environment, the present application designs multiple and reasonable learning features for the orbit correction model, including time features, satellite state features and space environment features. By fusing the multiple features, the input information of the model is enhanced, the generalization ability of the orbit correction model for predicting orbit errors is improved, a target error prediction value for the satellite velocity and position parameters is obtained, and the second velocity and position parameter (i.e., the predicted value) output by the orbital mechanics model is corrected using the target error prediction value to obtain a higher-accuracy target velocity and position parameter, thereby improving the prediction accuracy of the satellite position at the prediction moment. Therefore, by combining the accurate prediction of atmospheric density by the first meta-model, the influence of the small thrust perturbation term on the change of the satellite orbit position, and the accurate evaluation of the error value of the velocity and position parameter output by the orbital mechanics model by the orbit correction model, the present application can effectively improve the accuracy of the satellite velocity and position parameters at the prediction moment, and then use the accurately predicted satellite velocity and position parameters to perform orbit decay risk analysis and early warning in advance, avoid the satellite orbit from decaying significantly and then falling to a lower position and crashing, reduce the occurrence of crashing situations, and provide guarantee for the stable on-orbit operation of the satellite. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] In order to more clearly illustrate the technical solution of the embodiments of the present application, the following briefly introduces the drawings in the embodiments of the present application.

[0014] Figure 1 It is a schematic flowchart of a method for ensuring the ultra-low orbit of a near-earth satellite provided by an embodiment of the present application;

[0015] Figure 2 It is a schematic flow chart of a method for ensuring the ultra - low orbit of a near - earth satellite provided by another embodiment of the present application;

[0016] Figure 3 It is a schematic flow chart of a method for ensuring the ultra - low orbit of a near - earth satellite provided by still another embodiment of the present application;

[0017] Figure 4 It is a schematic flow chart of a method for ensuring the ultra - low orbit of a near - earth satellite provided by still another embodiment of the present application;

[0018] Figure 5 It is an exemplary structural schematic diagram of a system for ensuring the ultra - low orbit of a near - earth satellite provided by an embodiment of the present application;

[0019] Figure 6 It is an exemplary schematic diagram of an atmospheric density prediction process provided by an embodiment of the present application;

[0020] Figure 7 It is an exemplary schematic diagram of an orbit calculation process provided by an embodiment of the present application;

[0021] Figure 8 It is a structural schematic diagram of a system for ensuring the ultra - low orbit of a near - earth satellite provided by an embodiment of the present application;

[0022] Figure 9 It is a hardware structural schematic diagram of an electronic device provided by an embodiment of the present application. Detailed implementation manners

[0023] The principles and spirit of the present application will be described below with reference to several exemplary embodiments. It should be understood that the purpose of providing these embodiments is to make the principles and spirit of the present application clearer and more thorough, so that those skilled in the art can better understand and then implement the principles and spirit of the present application. The exemplary embodiments provided herein are only a part of the embodiments of the present application, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments herein without creative efforts fall within the scope of protection of the present application.

[0024] In this document, terms such as first, second, and third are only used to distinguish one entity (or operation) from another entity (or operation), and do not require or imply any order or association between these entities (or operations).

[0025] The following briefly describes the related concepts, technical terms, etc. that may be involved in the embodiments of the present application.

[0026] Orbit Mechanics Model: A high-precision orbit propagation model that takes into account various realistic physical forces acting on a satellite (perturbation terms, such as the atmospheric drag perturbation term), and is used to predict the position and velocity of low-orbit satellites. Common orbit mechanics models include, for example, the Simplified General Perturbation Model 4 (SGP4), the High Precision Orbit Propagator (HPOP), etc.

[0027] Atmospheric Drag Perturbation: Refers to the drag perturbation generated when a satellite orbits due to its interaction with atmospheric molecules. This perturbation affects the satellite's orbit parameters, such as orbital altitude, orbital inclination, and period. For low-orbit satellites, atmospheric drag is the largest in magnitude among non-conservative forces, so its impact on the satellite's orbit must be considered.

[0028] Feature Fusion: The process of combining data features from different sources to enhance the input information of a model. This can be achieved through simple concatenation, weighted combination, or more complex methods (such as dimensionality reduction), aiming to improve the model's understanding and prediction ability for tasks.

[0029] Model Fusion Technology: A type of ensemble learning method that combines the prediction results of multiple models to improve the overall prediction performance. This method can make up for the deficiencies of a single model and enhance the accuracy and robustness of the model.

[0030] Base Prediction Model: Abbreviated as the base model, each base prediction model is trained and predicted independently of other models, generating its own prediction results based on different learning algorithms or different subsets of data. Different base prediction models have different learning biases and capabilities, and they capture different patterns or features in the data.

[0031] ECI: Earth-Centered Inertial (ECI) coordinate system, which does not rotate with the Earth.

[0032] ICRS Coordinate System: That is, the International Celestial Reference System (ICRS), which is the standard celestial reference system adopted by the International Astronomical Union (IAU). It is a geocentric inertial rectangular coordinate system, and the coordinate axes are defined based on the distant starry sky background. This coordinate system is not affected by the Earth's rotation or revolution and is mainly used in astronomy and satellite orbit calculations.

[0033] ECEF Coordinate System: That is, the Earth-Centered, Earth-Fixed (ECEF) coordinate system, which is a geocentric rectangular coordinate system with the origin at the center of the Earth and the coordinate axes pointing to fixed positions on the Earth. The X-axis of the coordinate system usually points to the Greenwich astronomical meridian, the Y-axis points to 90 degrees east longitude, and the Z-axis coincides with the Earth's axis of rotation. This coordinate system changes with the Earth's rotation and is fixed relative to the Earth's position.

[0034] Geodetic Coordinate System (WGS84): It is a geocentric coordinate system widely adopted internationally, with the full name of World Geodetic System 1984 Coordinate System. It is a geocentric coordinate system with the origin at the Earth's mass center, and uses longitude, latitude, and altitude to represent positions, and is defined based on the reference ellipsoid (WGS84 ellipsoid). It also changes with the Earth's rotation and is fixed relative to the Earth's position.

[0035] Solar Radiation Parameters: That is, Solar Indices, including F10.7, S10.7, M10.7, and Y10.7. Among them, the data of F10.7 is daily frequency, and its historical values can be obtained from NOAA (or the Atmospheric Density Prediction Center), and the predicted values for the next 40 days can be obtained from NOAA, or can be predicted using the F10.7 prediction model; S10.7, Y10.7, and M10.7 are daily frequency, and the predicted values for the next 40 days can be obtained using the corresponding prediction models; M10.7 is data with an update frequency of 3 hours, its historical values can be obtained using an algorithm, and the predicted values can be obtained using the corresponding prediction model.

[0036] F10.7 (i.e., F107): It is an index used to describe the intensity of solar radiation. It represents the solar radio flux at a wavelength of 10.7 cm, and thus can reflect the intensity of solar activity. F10.7 is used as a good representative of solar ultraviolet radiation. The measurement of F10.7 is the average value of the solar radio radiation intensity measured on the ground in a 100 MHz frequency band centered at 2800 MHz within one hour, and a value is updated daily. F10.7 is closely related to the activities on the solar surface, and there are also long and short periodic variation laws of 11 years and 27 days. It is an important index characterizing solar activity, and there is a strong correlation between geomagnetic disturbances and the active level of the sun.

[0037] S10.7 (EUV): This is a solar radiation parameter in the extreme ultraviolet band, representing the solar radiation in the 26 - 34 nm wavelength band, and this part of the radiation mainly comes from solar active regions.

[0038] M10.7: This is a solar radiation parameter in the far ultraviolet band, representing the solar radiation around approximately 280 nm. This part of the radiation includes the photospheric continuum and chromospheric line radiation. M10.7 is an index used to describe solar activity. It is calculated through the ratio of the core to the wing of the spectral line in the Mg II 280 nm band. This ratio reflects the emission in the solar active regions in the chromosphere and is an index that is theoretically independent of the change in instrument sensitivity. The M10.7 index is used as a good representative of the solar far ultraviolet (FUV) and extreme ultraviolet (EUV) radiation. The radiation in these two band regions has a significant impact on the Earth's atmosphere, especially the thermosphere (located between approximately 85 km and 600 km above the Earth's surface).

[0039] Y10.7: This is a composite solar index, mainly representing solar X-ray and Lyman-α radiation. X-rays mainly come from the solar corona, while Lyman-α radiation is generated in the upper chromosphere and transition region of the sun.

[0040] Geomagnetic parameters: That is, geomagnetic parameters, including the ap, Ap, Ap*, and Dst indices. Among them, the ap index is predicted by a deep learning model for 24 steps (each step is 3 days); Ap is the daily geomagnetic parameter, obtained by averaging 8 ap values on the same day; Ap* is the sliding geomagnetic parameter, obtained by averaging the ap values of the previous 8 prediction cycles of the target prediction period; Dst is a geomagnetic parameter with an hourly frequency.

[0041] ap index: Also known as the equivalent planetary amplitude of geomagnetic disturbance, with the unit of 2 nT. A new ap value is generated every 3 hours, and 8 ap indices are generated in chronological order every day. According to relevant standards, an ap value greater than 39 corresponds to a magnetic storm period. The maximum value of the ap index is 400.

[0042] Ap index: It is an index of the global daily geomagnetic disturbance intensity, also known as the equivalent daily amplitude of the planet, and is an indicator to measure the level of geomagnetic activity. A new Ap value is generated every day, which is the average of 8 ap values on the same day. Therefore, the Ap index can only show the overall situation of the daily geomagnetic disturbance.

[0043] Ap* index: It is an index of the geomagnetic disturbance intensity in the prediction period, and is an indicator to measure the level of geomagnetic activity in real time. Compared with the Ap index, it has higher real-time performance, and thus can more accurately evaluate the overall situation of the geomagnetic disturbance in a certain prediction period.

[0044] Dst Index: This is the Disturbance Storm Time (Dst) index, which is used to indicate the intensity of the ring current during a magnetospheric storm. The Dst index is determined from hourly magnetic field measurements at four points around the Earth's equator and is used to simulate global density changes during a geomagnetic storm.

[0045] The spatio-temporal parameters include altitude, longitude and latitude, universal time, DOY, month, local time (or mean solar time, and the mean solar time cancels the whole-hour limit).

[0046] The following combines the accompanying drawings and details the method for ensuring the ultra-low orbit of a near-earth satellite provided by the embodiments of the present application through specific embodiments and their application scenarios.

[0047] Figure 1 It is a schematic flowchart of the method for ensuring the ultra-low orbit of a near-earth satellite provided by an embodiment of the present application. The execution subject of the method for ensuring the ultra-low orbit of the near-earth satellite can be an ultra-low orbit assurance system.

[0048] The following takes the execution subject of the method for ensuring the ultra-low orbit of a near-earth satellite as an ultra-low orbit assurance system as an example to illustrate the method for ensuring the ultra-low orbit of a near-earth satellite of the present application. It should be noted that the above execution subject and application scenario do not constitute a limitation to the present application.

[0049] As Figure 1 shown, the method for ensuring the ultra-low orbit of a near-earth satellite provided by the embodiments of the present application may include Step 110 - Step 180.

[0050] Step 110: Input the historical environmental parameters corresponding to the target prediction period and M target prediction values into the first meta-model, and assign M first weight values to the M target prediction values through the first meta-model;

[0051] Step 120: Use the M first weight values to perform weighted summation on the M target prediction values to determine the atmospheric density of the target prediction period;

[0052] Step 130: Based on the real-time windward mass ratio, drag coefficient, atmospheric density of the satellite at the current moment, as well as the speed scalar and speed vector of the satellite relative to the atmosphere, determine the atmospheric drag perturbation term of the satellite at the current moment;

[0053] Step 140: Based on the atmospheric drag perturbation term, the conventional perturbation term, and the small thrust perturbation term, determine the satellite orbit elements at the current moment, and input the satellite orbit elements at the current moment into the orbit mechanics model;

[0054] Step 150: Based on the orbital mechanics model and the satellite orbital elements, determine the velocity and position parameters of the satellite at the prediction moment in the geocentric inertial rectangular ICRS coordinate system, obtain the first velocity and position parameters, and convert the first velocity and position parameters into the Earth-centered, Earth-fixed rectangular ECEF coordinate system to obtain the second velocity and position parameters at the prediction moment;

[0055] Step 160: Input the time characteristics, satellite state characteristics, space environment characteristics, and the second velocity and position parameters at the current moment into the orbit correction model, so that the orbit correction model evaluates the error of the second velocity and position parameters to obtain the target error prediction value;

[0056] Step 170: Adjust the second velocity and position parameters based on the target error prediction value to obtain the target velocity and position parameters of the satellite at the prediction moment;

[0057] Step 180: Conduct an early warning of the orbit decay risk based on the target velocity and position parameters of the satellite at the prediction moment.

[0058] The low-earth orbit guarantee method for near-earth satellites provided by the embodiments of the present application combines the accurate prediction of atmospheric density by the first meta-model, the influence of the small thrust perturbation term on the change of the satellite orbit position, and the accurate evaluation of the error value of the velocity and position parameters output by the orbit mechanics model by the orbit correction model. The present application can effectively improve the accuracy of the satellite velocity and position parameters at the prediction moment, and then use the accurately predicted satellite velocity and position parameters to conduct an early analysis and warning of the orbit decay risk in advance, avoid the satellite orbit from decaying significantly and then descending to a lower position and crashing, reduce the occurrence of crashing situations, and provide guarantee for the stable in-orbit operation of the satellite.

[0059] The following combines specific embodiments to introduce the specific implementation manners of the above steps in detail.

[0060] Regarding Step 110, input the historical environmental parameters corresponding to the target prediction period and M target prediction values into the first meta-model, and the first meta-model assigns M first weight values to the M target prediction values.

[0061] In Step 110, by inputting the historical environmental parameters corresponding to the target prediction period into M basic prediction models respectively, M target prediction values are obtained. The M target prediction values are output by the M basic prediction models based on the historical environmental parameters corresponding to the target prediction period. The basic prediction model is the basic atmospheric prediction model, including but not limited to the JB series models (Jacchia-Bowman), MSIS series models (Mass Spectrometer Incoherent Scatter), DTM series models (Drag Temperature Model), and M is a positive integer.

[0062] Historical environmental parameters include geomagnetic parameters, solar radiation parameters, and spatio-temporal parameters. Among them, both geomagnetic parameters and solar radiation parameters are time-series characteristics, including ap, Ap*, F10.7, M10.7, S10.7, Y10.7, etc., and spatio-temporal parameters are non-time-series characteristics , including longitude, latitude, altitude, DOY, month, and mean solar time.

[0063] The first meta-model is a fusion model for fusing the prediction results of M basic prediction models. The first weight value is used to characterize the prediction accuracy of each basic prediction model under the historical environmental parameters corresponding to the target prediction period. The higher the prediction accuracy, the higher the first weight value. Therefore, through the weight model of the first meta-model, the historical environmental parameters of the target prediction period can be analyzed, and M first weight values are assigned to M target prediction values, so that the weight assignment can conform to the spatial environmental conditions corresponding to the target prediction period.

[0064] It involves step 120 of using M first weight values to perform weighted summation on M target prediction values to determine the atmospheric density of the target prediction period.

[0065] It involves step 130 of determining the atmospheric drag perturbation term of the satellite at the current moment based on the real-time windward mass ratio, drag coefficient, atmospheric density of the satellite at the current moment, as well as the speed scalar and speed vector of the satellite relative to the atmosphere.

[0066] In step 130, the current moment is within the target prediction period, and the atmospheric drag perturbation term can be expressed by the second acceleration , where , is the second acceleration, C represents the drag coefficient, A represents the real-time windward area of the satellite, m represents the mass of the satellite, represents the atmospheric density, v and represent the speed scalar and speed vector of the satellite relative to the atmosphere.

[0067] In the embodiment of the present application, based on the mass and real-time windward area of the satellite in the atmospheric drag perturbation term, the real-time windward mass ratio of the satellite can be determined, which can more accurately reflect the change in the windward area of the satellite in different attitudes. Then, based on this real-time windward mass ratio, the second acceleration corresponding to the atmospheric drag perturbation term is determined, which can improve the calculation accuracy of the second acceleration in the satellite orbit mechanics model, and further improve the accuracy of the satellite orbit elements based on this second acceleration.

[0068] It involves step 140 of determining the satellite orbit elements at the current moment based on the atmospheric drag perturbation term, the conventional perturbation term, and the small thrust perturbation term, and inputting the satellite orbit elements at the current moment into the orbit mechanics model.

[0069] In step 140, the low-earth orbit satellite orbit transfer has the characteristics of high maneuvering frequency, long continuous maneuvering time, and very small engine thrust. In order to maintain the continuity of orbit prediction, this application adds a small thrust perturbation term that did not exist in the past to the orbit mechanics model to calculate the acceleration generated by the ion engine thrust, so as to consider the effect of this acceleration on the satellite when determining the satellite orbit elements. Conventional perturbation terms may include third-body gravitational perturbation terms, earth non-spherical perturbation terms, solid tide force perturbation terms, solar radiation pressure perturbation terms, and other perturbation terms, etc.

[0070] Regarding step 150, based on the orbit mechanics model and the satellite orbit elements, determine the velocity and position parameters of the satellite at the prediction time in the geocentric inertial rectangular ICRS coordinate system, obtain the first velocity and position parameters, and convert the first velocity and position parameters into the earth-fixed rectangular ECEF coordinate system to obtain the second velocity and position parameters at the prediction time.

[0071] In step 150, the orbit mechanics model can be SGP4, HPOP, SGT4 or other models, and this application does not make specific limitations on this. The prediction time is a future time, after the current time, and there can be a prediction period between the current time and the prediction time. The period duration of this prediction period can be preset according to specific requirements, and this application does not make specific limitations on this. The prediction frequencies of the orbit mechanics model and the orbit correction model are the same, and both predict once every prediction period.

[0072] The velocity and position parameters in this application include velocity parameters and position parameters. By inputting the satellite orbit elements at the current time into the orbit mechanics model, the orbit mechanics model can predict the velocity and position parameters at the future time based on the satellite orbit elements, and represent the predicted velocity and position parameters in the ICRS coordinate system to obtain the first velocity and position parameters.

[0073] This application can use a professional math library or Geographic Information System (GIS) software (such as the geographic toolboxes of PROJ, GDAL, or MATLAB) to implement the coordinate system conversion of the velocity and position parameters. The conversion process may also need to consider parameters such as the oblateness of the earth and the reference ellipsoid model. Among them, PROJ is applied in the field of geographic information systems (GIS) and map projection, and generally refers to an open-source library for coordinate conversion and map projection, which supports the conversion between multiple map projection methods and coordinate reference systems (CRS). GDAL (Geospatial Data Abstraction Library) is an open-source library for reading and writing raster and vector geospatial data formats. MATLAB (Matrix Laboratory) is a high-level programming language and environment for numerical calculation and scientific calculation.

[0074] It involves step 160 of inputting the time feature, satellite state feature, space environment feature at the current moment, and the second velocity position parameter into the orbit correction model, so that the orbit correction model evaluates the error of the second velocity position parameter to obtain the target error prediction value.

[0075] In step 160, the satellite state feature is characterized in the ECEF coordinate system, and the space environment feature is characterized in the geodetic coordinate system wgs84. Both the wgs84 and ECEF coordinate systems change with the rotation of the earth and are fixed relative to the earth's position. Therefore, they are relatively stationary and both belong to the earth-fixed coordinate system. After the time feature, satellite state feature, space environment feature at the current moment, and the second velocity position parameter are subjected to feature splicing through the feature fusion technology to obtain the input feature, it can be input into the orbit correction model. The orbit correction model is pre-trained and is used to evaluate the error of the predicted value of the velocity position parameter output by the orbit mechanics model. This error is relative to the true value. During the training process of the orbit correction model based on historical training data, this true value can be actually detected by the satellite detection system for the satellite. Both the above-mentioned first velocity position parameter and the second velocity position parameter are the predicted values of the velocity position parameter at the prediction moment output by the orbit mechanics model, and the difference between them is that they are characterized in different coordinate systems.

[0076] This application designs diverse and reasonable learning features for the orbit correction model, including time features, satellite state features, and space environment features. By fusing the diverse features, the input information of the model is enhanced, and the generalization ability of the orbit correction model for orbit error prediction is improved. Moreover, both the input feature and the output label of the orbit correction model are uniformly characterized in the earth-fixed coordinate system, which can fit the error between the predicted value and the true value output by the orbit mechanics model under different satellite states and different space environment states, analyze the performance ability of the predicted value output by the orbit mechanics model in different prediction scenarios, realize the global analysis and modeling of the prediction error of the orbit mechanics model, obtain the target error prediction value for the predicted value, and use this target error prediction value to correct the second velocity position parameter (i.e., the predicted value) output by the orbit mechanics model to obtain a more accurate target velocity position parameter, thereby improving the prediction accuracy of the satellite position at the prediction moment.

[0077] It should be noted that this application can input the input feature after feature fusion into the orbit correction model, or input the time feature, satellite state feature, space environment feature at the current moment, and the second velocity position parameter into the orbit correction model respectively, and the orbit correction model performs feature fusion on various features.

[0078] The current moment is and the prediction moment is , , is the cycle duration of the prediction cycle. The time feature corresponding to the current moment may include the month of the current moment, the DOY (day of year), and the mean solar time of the longitude where the satellite is located at the current moment. DOY refers to the day of the year when the current moment is located.

[0079] In some embodiments of the present application, the satellite state feature may include the mass of the satellite, the windward area, and the velocity position parameters of the satellite in the ECEF coordinate system at the current moment. The space environment feature may include the position of the satellite in the geodetic coordinate system at the current moment, the relative position between the main celestial body and the satellite, and the solar radiation parameters and geomagnetic parameters of the satellite.

[0080] Specifically, the satellite state feature may include the mass m of the satellite, the windward area , and the velocity position parameters of the satellite at the current moment in the ECEF coordinate system. For easy distinction, this velocity position may be the third velocity position parameter, which may specifically include three position component parameters ( , , ) and three velocity component parameters ( , , ).

[0081] In the embodiments of the present application, by converting the velocity position parameters of the satellite at the current moment in the input features into the ECEF coordinate system, the input features and output features of the orbit correction model can be unified into the geodetic coordinate system that is relatively fixed with respect to the Earth, reducing the training difficulty of the orbit correction model and enhancing the interpretability of the orbit correction model. In addition to time features and satellite state features, the input features may also include the relative positions of the main celestial bodies (the Earth, the Sun, and the Moon), solar radiation parameters, and geomagnetic parameters, and various position features, time features, and model prediction results are all uniformly represented in the geodetic coordinate system, enabling the orbit correction model to directly learn the relationship between the positions of the Earth, the Sun, the Moon, and the prediction error of the orbit mechanics model. In this way, by inputting multivariate features into the orbit correction model, the orbit correction model can analyze and evaluate the prediction error of the orbit mechanics model under the influence of the current satellite state and the satellite space environment, and obtain the target error prediction value that can accurately correct the output prediction value of the orbit mechanics model.

[0082] In some embodiments of the present application, in addition to being able to characterize in the ECEF (Earth-Centered, Earth-Fixed coordinate system), there is an alternative solution for the second speed position parameter, the third speed position parameter, the prediction result of the orbital mechanics model, and the prediction error of the orbit correction model in the present application, that is, to perform coordinate transformation on the above parameters and convert them into the Earth-fixed polar coordinate system, so that they all characterize in the Earth-fixed polar coordinate system, and the position and speed expressions in this Earth-fixed polar coordinate system have a more direct correlation with other characteristics characterized in the geodetic coordinate system.

[0083] In some embodiments of the present application, the satellite position may include the longitude, latitude, and altitude of the satellite, and the relative position between the main celestial body and the satellite includes the longitude and latitude of the direct solar point on the Earth, the relative longitude and latitude difference between the satellite and the sun, the longitude and latitude of the direct lunar point on the Earth, and the relative longitude and latitude difference between the satellite and the moon.

[0084] Specifically, the space environment characteristics may include the following items: the longitude of the satellite , latitude and altitude; the longitude and latitude of the direct solar point on the Earth, and the relative longitude and latitude difference and between the satellite and the sun; the longitude and latitude of the direct lunar point on the Earth, and the relative longitude and latitude difference and between the satellite and the moon; solar radiation parameters F10.7, M10.7, S10.7, Y10.7; geomagnetic parameters ap, Ap*.

[0085] The longitude and latitude of the satellite can be calculated from the satellite orbital elements, and the longitude and latitude of the sun and the moon at a specific moment can also be directly calculated by professional astronomical software. The relative longitude and latitude difference calculation formulas (1), (2), (3), and (4) can be as follows:

[0086] (1)

[0087] (2)

[0088] (3)

[0089] (4)

[0090] In some embodiments, the above astronomical software may be, for example, Astropy. Astropy is a Python software package for astronomical data analysis and processing, an open-source project promoted by the community, aiming to provide a powerful core toolkit for astronomy and astrophysics in the Python programming environment.

[0091] In some embodiments of the present application, in order to train an orbit correction model that can accurately evaluate the prediction error of the orbit mechanics model based on the current satellite state and the satellite space environment state, before the above step 160, the following steps may further be included: obtaining time features, satellite state features, space environment features, and satellite orbit elements at a plurality of first moments; inputting the satellite orbit elements at the first moment into the orbit mechanics model to obtain predicted values of velocity and position parameters for a second moment output by the orbit mechanics model, where a preset period of time elapses between the first moment and the second moment; constructing historical training data by combining the time features, satellite state features, and space environment features at each first moment, as well as the predicted values and true values of the velocity and position parameters for the second moment; training a preset model based on the historical training data to obtain an orbit correction model associated with the orbit mechanics model.

[0092] Specifically, the first moment and the second moment are two adjacent historical prediction moments, and the second moment is after the first moment. The true value of the velocity and position parameters for the second moment can be obtained by actual detection of the satellite by a satellite detection system. The mean square error can be used as the loss function during model training, and the backpropagation algorithm can be used for training. The orbit correction model is associated with the orbit mechanics model. The preset model can be set according to specific requirements, and the present application does not make specific limitations thereto.

[0093] The above training of the preset model based on the historical training data may specifically include: using the time features, satellite state features, and space environment features at the first moment, as well as the predicted values of the velocity and position parameters for the second moment as input training data, and using the difference between the predicted values and true values of the velocity and position parameters for the second moment as output training data (i.e., output labels) for training.

[0094] Among them, the difference between the predicted values and true values of the velocity and position parameters for the second moment may be the true error value, and the mean square error between the predicted training error value output by the preset model and the true error value is calculated to obtain the loss function value.

[0095] It should be noted that in the historical training data, the satellite state features are in the ECEF coordinate system, the space environment features are in the geodetic coordinate system, and the predicted values and true values of the velocity and position parameters for the second moment are both in the ECEF coordinate system, ensuring that the historical training data are all uniformly represented in the earth-fixed coordinate system.

[0096] In the embodiments of the present application, during the training process of the orbit correction model, the training data used is derived from the predicted values actually output by the orbit mechanics model. Based on this, during the training process of the orbit correction model, it can learn the deviations between the predicted values and the true values output by the orbit mechanics model under different time characteristics, different satellite state characteristics, and different space environment characteristics. Furthermore, the trained orbit correction model can have the ability to reasonably and accurately evaluate the errors of the predicted values output by the orbit mechanics model under different satellite orbit prediction scenarios by combining time characteristics, satellite state characteristics, and space environment characteristics, so as to obtain an orbit correction model adapted to the orbit mechanics model, enabling the orbit correction model to have a better prediction error evaluation ability for the associated orbit mechanics model. Moreover, for different orbit mechanics models, a related orbit correction model can be trained for each orbit mechanics model. The correction ability of the orbit correction model trained for a specific orbit mechanics model is only effective for that orbit mechanics model. For example, the correction model of HPOP is basically ineffective in correcting the errors of SGP4. This can make the adaptability and correlation between the orbit mechanics model and the orbit correction model stronger, and further improve the prediction error evaluation ability of the orbit correction model for the associated orbit mechanics model.

[0097] It involves step 170 of adjusting the second velocity position parameter based on the target error predicted value to obtain the target velocity position parameter of the satellite at the predicted moment.

[0098] In step 170, the target error predicted value also represents that in the ECEF coordinate system, the target error predicted value can be added to the second velocity position parameter to obtain the target velocity position parameter at the predicted moment. The target error predicted value can include the velocity error predicted value and the position error predicted value.

[0099] Exemplarily, the second velocity position parameter includes the velocity parameter and the position parameter , and the orbit correction model outputs the velocity error predicted value and the position error predicted value in the ECEF coordinate system of the satellite. Then, combining the second velocity position parameter and the prediction result of the correction model, the corrected target velocity position parameter is calculated according to formulas (5) and (6). The target velocity position parameter includes the velocity parameter and the position parameter :

[0100] (5)

[0101] (6)

[0102] Finally, and Convert to the final prediction value in the ICRS coordinate system and report it.

[0103] In some embodiments of the present application, both the second velocity position parameter and the target velocity position parameter include a position parameter and a velocity parameter. The position parameter includes three position component parameters on the x, y, and z axes, and the velocity parameter includes three velocity component parameters on the x, y, and z axes. Correspondingly, the velocity error prediction value includes three velocity component error prediction values on the x, y, and z axes, and the position error prediction value includes three position component error prediction values on the x, y, and z axes.

[0104] Based on this, both the second velocity position parameter and the target velocity position parameter include three position component parameters and three velocity component parameters. The orbit correction model is constructed by six sub-fusion models. Each sub-fusion model consists of multiple base models and a target fusion model. Each base model is used to evaluate the deviation between the predicted value and the true value of the corresponding component parameter in the velocity position parameter at the prediction moment, and obtain the predicted component error associated with the corresponding component parameter. The target fusion model is used to fuse the multiple predicted component errors output by the multiple base models to obtain the component error prediction value output by the sub-fusion model;

[0105] Specifically, the structures of the six sub-fusion models are the same. The multiple base models include at least two of lightGBM, random forest algorithm, support vector machine SVM algorithm, long short-term memory network LSTM, and encoder transformer encoder.

[0106] The multiple base models are trained based on historical training data. Specifically, the time feature, satellite state feature, space environment feature at the first moment, and the predicted value of the velocity position parameter for the second moment are used as input training data, and the difference between the predicted value and the true value of the corresponding component parameter of the velocity position parameter for the second moment is used as output training data (i.e., output label) for training. The target fusion model is used to fuse the output results of the multiple base models. For example, a linear regression algorithm can be used to perform weighted summation on the output results of the multiple base models to obtain the target velocity position parameter.

[0107] The above step 160 may specifically include: based on the six sub-fusion models, respectively evaluate the errors of the six component parameters in the second velocity position parameter to obtain a target error prediction value including six component error prediction values. Among them, the six component error prediction values include three velocity component error prediction values and three position component error prediction values. The six component error prediction values are used to adjust the six component parameters in the second velocity position parameter respectively to obtain the target velocity position parameter.

[0108] Exemplarily, the prediction moment is , and the current moment is . Among the third position parameters of the satellite at the moment of , there are three position component parameters ( , , ) and three velocity component parameters ( , , ); among the second velocity position parameters of the satellite at the moment of , there are three position component parameters , and three velocity component parameters .

[0109] In this way, the present application trains a sub-fusion model for each component parameter, enabling the sub-fusion model to have the ability to evaluate the error of a single component parameter, with stronger pertinence and adaptability. Therefore, the accuracy of the error evaluation of the predicted value output by the orbit correction model for the orbit mechanics model can be improved.

[0110] It involves step 180 of warning of the orbit decay risk based on the target velocity position parameter of the satellite at the prediction moment.

[0111] In step 180, based on the target velocity position parameter of the satellite at the prediction moment, the predicted decay height of the satellite can be determined, and when the predicted decay height reaches the preset level threshold, an active orbit lift strategy is initiated. That is, when the orbit decay risk of the satellite is pre-monitored, the satellite engine is controlled to continuously push, and an orbit lift operation is actively performed on the satellite before the satellite orbit decays to a lower height, reducing the falling risk.

[0112] It should be noted that the present application can set an orbit decay period, which is a period for predicting the decay height of the satellite and adjusting the real-time operating height of the satellite based on the decay height. The orbit decay period can include at least one prediction period. Exemplarily, if the period durations of both the orbit decay period and the prediction period are 3h, then based on the velocity position parameter of the satellite at the current moment and the target velocity position parameter at the prediction moment, the predicted decay height of the satellite within the orbit decay period can be determined; if the orbit decay period is 24h and the prediction period is 3h, then the target velocity position parameter is the velocity position parameter 3h after the current moment. On this basis, it is necessary to continue to use the target velocity position parameter to predict the velocity position parameter 6h after the current moment. After multiple predictions, based on the velocity position parameter (predicted value) 21h after the current moment, the velocity position parameter 24h after the current moment is predicted, and in combination with the two velocity position parameters separated by 24h, the predicted decay height of the satellite within the orbit decay period is determined.

[0113] In some embodiments of the present application, in order to improve the accuracy and robustness of the first meta-model, and reduce the overall prediction error when predicting the atmospheric density in the face of different environmental conditions, before the above step 110, the method may further include steps 210 to 230.

[0114] Step 210, obtain target training data, where the target training data includes historical environmental parameters, measured values, and M predicted values corresponding to each prediction period in multiple prediction periods. The measured value is the atmospheric density actually detected in each prediction period, and the M predicted values are obtained by M basic prediction models predicting the atmospheric density in each prediction period based on the historical environmental parameters of each prediction period.

[0115] The measured value is the atmospheric density actually detected in each prediction period. For example, a large amount of historical satellite telemetry data can be combined, and the detection value of the atmospheric density detector carried on the low-Earth orbit satellite can be used as the measured value. The M predicted values are obtained by M basic prediction models predicting the atmospheric density in each prediction period based on the historical environmental parameters of each prediction period.

[0116] Exemplarily, the historical prediction periods include period 1, period 2, and period 3. Taking period 1 as an example, the target training data contains the historical environmental parameters, measured value, and M predicted values corresponding to period 1. The measured value is the atmospheric density actually detected in period 1, and the M predicted values are obtained by M basic prediction models predicting the atmospheric density in period 1 based on the historical environmental parameters of period 1.

[0117] Step 220, train the first meta-model using the target training data, so that the first meta-model assigns corresponding M first weight values to the M predicted values in each prediction period based on the historical environmental parameters corresponding to each prediction period.

[0118] In step 220, feature extraction is performed on the geomagnetic parameters and solar radiation parameters for N consecutive time steps before each prediction period to obtain time series features, and feature extraction is performed on the spatio-temporal parameters of the prediction period to obtain non-time series features. The time series features, non-time series features, and M predicted values corresponding to each prediction period are used as input features, and the M measured values are used as labels to perform model training on the first meta-model. N is a positive integer and can be specifically set according to specific requirements, for example, set to 40.

[0119] The first meta-model obtains the M predicted values in each prediction period, and based on all the predicted values in multiple prediction periods, learns how to adjust the weight values of each basic prediction model under different environmental conditions, and assigns the first weight values to the M predicted values, that is, assigns the first weight values to the M basic prediction models.

[0120] Exemplarily, the M basic prediction models include the MSIS20 model, the MSIS00 model, the DTM2020 model, and the JB2008 model. All 4 basic prediction models can predict the atmospheric density for each prediction period based on the historical environmental parameters of the prediction period, and obtain 4 predicted values corresponding to each prediction period. Among them, the average value of the predicted values of JB2008 for three consecutive hours is used as the predicted value with a 3-hour time step. The first meta-model can assign 4 first weight values to the 4th predicted value based on the historical environmental parameters corresponding to the Nth prediction period, so as to perform weighted summation on the 4 predicted values based on the 4 first weight values, and output the predicted value of the atmospheric density for the Nth period after fusion.

[0121] Step 230: When the first loss function value of the first meta-model satisfies the first preset condition, it is determined that the training of the first meta-model is completed. The first loss function value is the mean square error between the first predicted value and the measured value corresponding to each prediction period among multiple prediction periods. The first predicted value is obtained by performing weighted summation on the M predicted values corresponding to each prediction period using the M first weight values assigned by the first meta-model in each prediction period.

[0122] In step 230, the first loss function value is the mean square error between the first predicted value and the measured value corresponding to each prediction period among multiple prediction periods. The first predicted value is obtained by performing weighted summation on the M predicted values corresponding to each prediction period using the M first weight values assigned by the first meta-model in each prediction period. This application uses the stacking method to fuse the models.

[0123] The first meta-model is a fusion model for fusing the prediction results of M basic prediction models. The first predicted value is the fusion result obtained by performing weighted summation on the prediction results of the M basic prediction models by the first meta-model. Therefore, calculate the mean square error between the first predicted value and the measured value for multiple prediction periods. When the mean square error is less than the first preset error threshold, it is determined that the first loss function value of the first meta-model satisfies the first preset condition, and the training of the first meta-model is completed.

[0124] In some embodiments of this application, the first meta-model includes a weight model and a fusion layer. The network structure of the weight model is a transformer encoder or an Lstm. The weight model is used to assign corresponding M first weight values to the M predicted values of different prediction periods. The fusion layer is used to perform weighted summation on the M predicted values using the M first weight values. Therefore, the fusion layer can also be a weighted summation layer.

[0125] Specifically, when the network structure of the weight model is a Transformer encoder, the Transformer encoder is used as a learner to train the weight model. When the network structure of the weight model is an LSTM, the LSTM is used as a learner to train the weight model. The weights of each basic prediction model are dynamically calculated according to the environmental features corresponding to the historical environmental parameters. Finally, the prediction values of each basic prediction model are weighted and summed to obtain the prediction result of the first meta-model.

[0126] The input features of the weight model include the features corresponding to geomagnetic parameters, solar radiation parameters, and spatio-temporal parameters. The output result of the weight model is the first weight value of each basic prediction model (the number of output weights is the same as the number of models). In addition to the input of the weight model, the input features of the first meta-model also include the prediction results of the basic prediction models, that is, M prediction values. The output of the first meta-model is the final result of this fusion.

[0127] In one example, for instance, there are 4 basic prediction models, and the prediction values of the basic prediction models for the prediction period are [pred1, pred2, pred3, pred4] respectively. At the same moment, the weight vector W = [w1, w2, w3, w4, b] output by the weight model, where w1, w2, w3, w4 are the 4 first weight values respectively, and b is the bias term. Then the prediction result of the fusion layer of the first meta-model at this moment As shown in formula (7):

[0128] (7)

[0129] In the embodiments of the present application, various atmospheric density prediction models are used as basic prediction models to make a preliminary prediction of the atmospheric density. The first meta-model that fuses the weight model and the final weighted summation layer together. The input of the first meta-model includes the input features of the weight model and the prediction results of each basic prediction model. The first meta-model is trained using the mature machine learning algorithms Transformer encoder or LSTM, enabling the first meta-model to analyze the performance capabilities of each basic prediction model under different spatial environmental conditions, combining the prediction accuracies of each basic prediction model for the atmospheric density under different spatial environmental conditions, and assigning different first weight values to each basic prediction model under different spatial environmental conditions, so that the accuracy of the prediction results of each basic prediction model after fusion is higher, and a more accurate atmospheric density prediction result is output.

[0130] In some embodiments of the present application, when the network structure of the weight model is Lstm, step 220 of training the first meta-model using the target training data may specifically include the following steps: extracting features from the geomagnetic parameters and solar radiation parameters for N consecutive time steps before each prediction period to obtain time series features , and extracting features from the spatio-temporal parameters of each prediction period to obtain non-time series features ; after normalizing the time series features and inputting them into the weight model, obtaining the high-order features H of the last hidden layer of the Nth round of the weight model ; merging the features of H and the non-time series features, inputting them into the fully connected layer (FC), followed by a softmax layer, and outputting the weight vector W, where the weight vector contains M first weight values; performing a weighted sum of the prediction values of the M basic prediction models based on the weight vector to obtain the first prediction value; using the measured value of the atmospheric density as the label, using the mean square error between the measured value and the first prediction value as the loss function, and using the backpropagation algorithm to train the first meta-model

[0131] Specifically, both the geomagnetic parameters and the solar radiation parameters are time series features , including ap, Ap*, F10.7, M10.7, S10.7, Y10.7. The update frequencies of ap, Ap*, and M10.7 are 3 hours, and the update frequencies of F10.7, S10.7, and Y10.7 are 1 day. The time series features use 3 hours as the time step. The time series features need to be normalized before being input into the Lstm. The spatio-temporal parameters are non-time series features , including longitude, latitude, altitude, DOY, month, and mean solar time. The Lstm sequence length can be 40 or other values, the hidden layer depth can be 1, 2 or other values, and the hidden layer dimension can be 32, 64 or other values. The present application does not make specific limitations on this

[0132] In some embodiments of the present application, before step 130 of determining the atmospheric drag perturbation term of the satellite at the current moment, steps 310 - 330 may further be included: step 310, when it is determined that the velocity direction of the satellite is the windward direction, obtaining the wind direction unit vector in the satellite body coordinate system, the satellite solar panel area, and the unit normal vector of the plane where the satellite solar panel is located in the satellite body coordinate system; step 320, converting the wind direction unit vector in the satellite body coordinate system to the satellite body coordinate system; step 330, based on the satellite solar panel area, and the wind direction unit vector and unit normal vector in the satellite body coordinate system, determining the real-time windward area of the satellite

[0133] Specifically, if the windward direction is the velocity direction of the satellite, then the wind direction unit vector in the satellite body coordinate system , on this basis, the wind direction unit vector can be transformed from the on-board coordinate system to the satellite body coordinate system by using the rotation matrix.

[0134] In the embodiment of the present application, based on the satellite solar panel area, as well as the wind direction unit vector and the unit normal vector on the satellite body coordinate system, the real-time windward area of the satellite is determined, and the fixed windward area in the atmospheric drag perturbation term is updated to the real-time windward area. Compared with the fixed constant, the real-time windward area can be dynamically calculated when the windward area changes due to the movement of components such as the attitude, orbital altitude, and solar panels. Therefore, the accuracy of the real-time windward area is higher. In this way, the present application optimizes and upgrades the atmospheric drag perturbation term. When determining the satellite orbit elements, the acceleration generated by the atmospheric drag perturbation term can be determined based on the more accurate real-time windward area, thereby improving the accuracy of satellite orbit element calculation.

[0135] In some embodiments of the present application, the above step 320 of transforming the wind direction unit vector of the on-board coordinate system to the satellite body coordinate system may specifically include: obtaining the second rotation matrix of the satellite attitude angle relative to the satellite body coordinate system; based on the second rotation matrix, transforming the wind direction unit vector from the on-board coordinate system to the satellite body coordinate system.

[0136] Specifically, the satellite attitude angle includes the roll angle , the pitch angle , and the yaw angle . The second rotation matrix from the on-board coordinate system to the satellite body coordinate system can be calculated by method (8):

[0137] (8)

[0138] where respectively represent the rotation matrices around the z-axis, y-axis, and x-axis of the on-board coordinate system, and are respectively:

[0139]

[0140] Transform the wind direction unit vector from the on-board coordinate system to the satellite body coordinate system, , and through calculation, it can be obtained that .

[0141] In this way, after obtaining the second rotation matrix, the second rotation matrix can be used to successfully transform the wind direction unit vector from the on-board coordinate system to the satellite body coordinate system, and an effective expression of the wind direction unit vector in the satellite body coordinate system can be obtained.

[0142] In some embodiments of the present application, in order to accurately calculate the real-time windward area of the satellite, step 330 above determines the real-time windward area of the satellite based on the satellite solar panel area, as well as the wind direction unit vector and unit normal vector in the satellite body coordinate system, and may specifically include: calculating the absolute value of the product of the wind direction unit vector and the unit normal vector in the satellite body coordinate system; multiplying the satellite solar panel area by the absolute value to obtain the real-time windward area of the satellite.

[0143] Specifically, the windward area of the satellite can be calculated using formula (9) :

[0144] (9)

[0145] where S is the solar panel area, and the unit normal vector of the plane where the satellite solar panel is located in the satellite body coordinate system (the general satellite configuration is [0,0,1]), the satellite body can be approximately regarded as a sphere, s is the area of the satellite sphere, and finally the real-time windward area of the satellite is obtained through calculation .

[0146] In the embodiments of the present application, by calculating the absolute value of the product of the wind direction unit vector and the unit normal vector in the satellite body coordinate system and multiplying the satellite solar panel area by the absolute value, the real-time windward area of the satellite can be effectively calculated. In this way, in the scenario where the satellite's attitude changes due to factors such as orbital adjustment and solar panel rotation, by calculating the windward area in real time, these complex attitude changes can be effectively adapted, and the calculation error of the windward area caused by attitude changes can be avoided.

[0147] In some embodiments of the present application, in order to obtain the small thrust perturbation term, Figure 4 is a schematic flowchart of the ultra-low orbit guarantee method for a low-earth orbit satellite provided in another embodiment of the present application. Before step 140 above determines the satellite orbit elements at the current moment based on the atmospheric drag perturbation term, the conventional perturbation term, and the small thrust perturbation term, it may further include Figure 4 the steps 410-step 430 shown.

[0148] Step 410, when the thrust magnitude of the satellite's ion thruster is constant, obtain the thrust vector of the satellite engine in the satellite body coordinate system.

[0149] Specifically, the thrust generated by the ion thruster is a small thrust, and this low thrust is regarded as a perturbation term in orbital dynamics because it produces continuous small perturbations to the orbit of the spacecraft. Although the small thrust has a small magnitude, it can accumulate a significant velocity increment over a long time, and the continuous action of the small thrust will cause slow changes in the orbit. If not considered, it may lead to the accumulation of orbit prediction errors.

[0150] Step 420: Obtain the first rotation matrix of the satellite attitude angle with respect to the geocentric inertial coordinate system, and based on the first rotation matrix, convert the thrust vector of the satellite engine in the satellite body coordinate system to the geocentric inertial coordinate system to obtain the thrust components of the thrust vector on the three coordinate axes of the geocentric inertial coordinate system.

[0151] Specifically, the conversion of the same vector between different coordinate systems can be completed through the rotation matrix. The satellite attitude is usually described with respect to the ECI coordinate system. Therefore, in this application, the satellite attitude angle can be obtained through the satellite's attitude sensor, and then the first rotation matrix can be determined using the satellite attitude angle. .

[0152] The rotation matrix from the three-dimensional coordinate system A to the coordinate system B is a 3×3 matrix that defines how to transform the representation of a vector from the coordinate system A to the representation in the coordinate system B . The vector can be transformed between the A and B coordinate systems through formula (10):

[0153] (10)

[0154] The rotation matrix , in the matrix represents the projection of the unit vector in the direction of the first axis of the A coordinate system on each coordinate axis direction in the B coordinate system, and the other elements are analogized accordingly.

[0155] Based on this, the thrust vector of the engine is , and the thrust direction of the engine points to the rear of the satellite. Then the representation in the satellite body coordinate system (BFC) is , where T is the magnitude of the thrust generated by the engine.

[0156] Using the first rotation matrix to perform coordinate transformation, and obtain the representation in the ECI coordinate system . Through calculation, the components of T on the three coordinate axes of the ECI can be obtained:

[0157]

[0158] Step 430: Combine the satellite mass, the real-time on / off state of the engine, and the thrust components on the three coordinate axes to obtain the small thrust perturbation term.

[0159] Specifically, after inputting the small thrust perturbation term into the orbital mechanics model, the orbital mechanics model can calculate the acceleration generated by the small thrust perturbation term based on the satellite mass, the real-time on / off state of the engine, and the thrust components on the three coordinate axes.

[0160] In some embodiments of the present application, step 140 of determining the satellite orbit elements based on the atmospheric drag perturbation term, the conventional perturbation term, and the small thrust perturbation term may specifically include: determining the first acceleration corresponding to the small thrust perturbation term based on the satellite mass, the real-time on / off state of the engine, and the thrust components on the three coordinate axes; determining the acceleration generated by non-conservative forces based on the first acceleration corresponding to the small thrust perturbation term and the second acceleration corresponding to the atmospheric drag perturbation term; determining the satellite orbit elements based on the acceleration generated by non-conservative forces, the Earth's gravitational acceleration, and the third-body gravitational acceleration, where the Earth's gravitational acceleration and the third-body gravitational acceleration are calculated from the conventional perturbation term.

[0161] Specifically, the transmitter can generate a small thrust when it is in the on state, and the engine cannot generate a small thrust when it is in the off state. Therefore, the first acceleration determined under different on / off states is different. The Earth's gravitational acceleration and the third-body gravitational acceleration are calculated from the conventional perturbation term.

[0162] In addition to the small thrust perturbation term and the atmospheric drag perturbation term, the non-conservative force may also include other non-conservative force perturbation terms. Therefore, the acceleration generated by the non-conservative force can be obtained by adding the first acceleration, the second acceleration, and the acceleration generated by other non-conservative force perturbation terms.

[0163] Exemplarily, if the orbital mechanics model is HPOP, the dynamic equation of HPOP can be described by Newton's second law of formula (11):

[0164] (11)

[0165] where represents the second derivative of the position vector with respect to time, represents the Earth's gravitational acceleration, , represents the third-body gravitational acceleration. The atmospheric drag and the small thrust generated by the engine both belong to non-conservative forces, and the accelerations generated by these two forces are represented by and respectively.

[0166] In the embodiments of the present application, a small thrust perturbation term is added on the basis of the atmospheric drag perturbation term and the conventional perturbation term. Based on this, when calculating the satellite orbital elements in the subsequent orbital mechanics model, the first acceleration corresponding to the small thrust perturbation term and the second acceleration corresponding to the atmospheric drag perturbation term can be determined. Furthermore, the acceleration generated by the non-conservative force is determined by combining the first acceleration and the second acceleration. Considering the small thrust perturbation term that was previously ignored, the calculated acceleration generated by the non-conservative force will be more accurate. By comprehensively considering these perturbation factors in the model, the dynamic changes of the satellite orbit can be analyzed more comprehensively, thereby improving the accuracy and reliability of orbit prediction.

[0167] In some embodiments of the present application, in order to improve the accuracy of the first acceleration, the above-mentioned determination of the first acceleration corresponding to the small thrust perturbation term based on the satellite mass, the real-time on / off state of the engine, and the thrust components on the three coordinate axes may include the following steps: Substitute the real-time on / off state of the engine into the working state function to obtain a function value, where different function values correspond to the on state and the off state; Based on the function value, the satellite mass, and the thrust components on the three coordinate axes, determine the first acceleration corresponding to the small thrust perturbation term.

[0168] Specifically, the acceleration expression of the small thrust perturbation term can be shown as formula (12):

[0169] (12)

[0170] Where, is the first acceleration, I represents the working state function of the engine, I = 1 when the engine is on, and I = 0 when the engine is off. The small thrust perturbation term is externally connected to the satellite simulation module to obtain the satellite mass m, the real-time on / off state of the satellite engine, and the thrust magnitudes in each direction, so as to calculate .

[0171] In the embodiments of the present application, the thrust magnitude of the ion thruster is constant, so the thrust vector of the engine remains constant. Based on this, the present application can effectively and accurately calculate the first acceleration generated by the small thrust perturbation term by using the thrust vector of the engine, the satellite mass, and the real-time on / off state of the satellite engine.

[0172] As a specific example, Figure 5 is a schematic structural diagram of an exemplary ultra-low orbit guarantee system for a low-earth orbit satellite provided by an embodiment of the present application. As shown in Figure 5 , the ultra-low orbit guarantee system includes a data collection and management module, a space environment prediction module, an atmospheric density prediction module, an atmospheric density correction module, a satellite simulation module, an orbit calculation module, an orbit correction module, and modules such as satellite situation and risk warning.

[0173] Among them, the functions of the data collection and management module include: collecting and storing various historical data in real time, including measured and predicted values of the space environment, measured values of satellite atmospheric density and model predicted values, measured values of satellite orbital elements and algorithm predicted values, platform parameters of different types of satellites, and satellite telemetry data; providing data support for the current reports and predictions of the situation of the space environment, atmospheric density, satellite orbit, satellite risk, etc.; providing data support for the training and upgrading of AI correction models such as data correlation analysis, atmospheric density and orbit calculation.

[0174] The space environment forecasting module mainly includes two models, namely the F107 forecasting model and the Ap forecasting model. F107 and Ap respectively correspond to the solar radiation intensity on the earth and the degree of geomagnetic disturbance caused by the solar wind, and are input parameters for common atmospheric density models. In Figure 6 the atmospheric density forecasting process shown, the space environment forecasting module extracts the historical values of relevant space environment characteristics required for predicting atmospheric density (such as Ap, F107, solar wind, X-rays, solar active regions, etc.) from the data collection and management module, predicts F107 and Ap, and transmits the prediction results to the subsequent atmospheric density forecasting module, and at the same time transmits them back to the data collection and management module for storage.

[0175] As Figure 6 shown, the atmospheric density forecasting module can be composed of multiple atmospheric models such as MSIS20, MSIS00, DTM, etc. Models can also be added. In actual applications, one output result can be selected from the output results of multiple atmospheric models as the final forecasting result of atmospheric density, or the output results of multiple atmospheric models can be selected as part of the input of the atmospheric density correction model. Finally, the density forecasting values of multiple models at the specified time and space are transmitted back to the data collection and management module for storage. In Figure 6 the atmospheric density forecasting process shown, the atmospheric density forecasting module can receive the predicted values of AP and F107 at a certain moment within the next 3 days output by the space environment forecasting module and the spatial position of the satellite at the same moment output by the orbit calculation module as inputs, and output the forecasting results of multiple models. Combining the input time and spatial position, it is used as part of the input of the atmospheric density correction model.

[0176] The atmospheric density correction model is an AI model, aiming to integrate the output results of multiple models in the atmospheric density forecasting module to further improve the prediction accuracy of atmospheric density. As Figure 6 shown, this correction model is an integrated model, which can be fused by other AIs such as neural network LSTM, lightGBM, Xgboost, SVM, etc. The training data of this correction model comes from the historical data collected by the data collection and management module. As Figure 6As shown in the figure, the inputs required for the correction model to make predictions are respectively sourced from the data collection and management module and the atmospheric density prediction model. The data collection and management module provides the historical and predicted values of space environment indices such as F107 and Ap. The atmospheric density prediction module provides the spatio-temporal location and the prediction results of multiple atmospheric models. The correction model calculates the space environment indices and the prediction results of multiple models, and outputs a more accurate atmospheric density correction value as the final predicted value of the atmospheric density, which is sent back to the data collection and management module for storage and simultaneously output to the orbit calculation module for satellite orbit calculation.

[0177] The satellite simulation module can obtain various parameters of the satellite platform from the data collection and management module, such as satellite mass, volume, solar panel area, attitude, remaining power, engine thrust, etc. Through satellite simulation algorithms, data such as the real-time windward area mass ratio and maneuver acceleration of the satellite can be obtained, thus providing data support for the precise calculation of the satellite orbit and situation display, etc. In Figure 7 In the orbit calculation process shown in the figure, the satellite simulation module obtains the specified satellite model, attitude strategy, and maneuver strategy from the data collection and management module, calculates the drag coefficient and real-time windward area mass ratio (dynamic area mass ratio) of the satellite through simulation algorithms, and passes them into the atmospheric perturbation term of the orbit calculation module, thereby improving the orbit calculation accuracy; obtains the satellite model, engine model, and maneuver strategy from the data collection and management module, and obtains information such as the engine parameters (engine thrust magnitude, direction), (switching on and off) state of the satellite through simulation algorithms, and passes them into the small thrust perturbation term of the orbit calculation module. In this way, by combining the atmospheric drag perturbation term, small thrust perturbation term, and conventional perturbation terms (third-body gravitational perturbation term, Earth non-spherical perturbation term, solid tidal force perturbation term, solar radiation pressure perturbation term, and other perturbation terms), the precise calculation of the satellite orbit in the maneuvering state is realized, and the calculated value of the orbital elements is obtained.

[0178] The orbit calculation module has been optimized and upgraded based on HPOP (High Precision Orbit Algorithm). From Figure 7 As can be seen from the orbit calculation process shown in the figure, the main optimization points are as follows: In view of the significant increase in the influence of atmospheric density on ultra-low orbits, parameters such as the drag coefficient, satellite mass, and windward area in the atmospheric drag perturbation term are opened, and by connecting to the atmospheric density correction module and the satellite simulation module, the calculation accuracy of the orbit algorithm for the influence of atmospheric density is improved; in view of the characteristics of ultra-low orbit satellites with high maneuvering frequencies, long continuous maneuvering times, and very small engine thrusts, a small thrust perturbation term that HPOP does not have is added. The small thrust perturbation term is connected to the satellite simulation module to obtain the real-time state of the satellite engine, thereby calculating the influence of satellite maneuvers on the orbit and realizing the orbit calculation of the satellite in maneuvers by the orbit algorithm.

[0179] The orbit correction module integrates neural network models such as LSTM and Transformers, as well as machine learning models such as SVM and lightGBM. The training data of the models is sourced from the historical data collected by the data collection and management module. As Figure 7 shown, this module obtains historical values or predicted values such as space environment data, satellite dynamics data, spatio-temporal position data, and telemetry data from the data collection and management module, and obtains the predicted values of the orbital elements from the orbit calculation module. Through joint calculations of multiple models, a more accurate corrected value of the orbital elements is obtained, and this corrected value is used as the final orbit calculation result, which is sent back to the data collection and management module for storage, and at the same time output to subsequent modules such as satellite situation and risk warning for display or analysis.

[0180] In this way, the orbit correction module integrates multiple AI models. Based on the calculation results of the orbit calculation module, combined with space environment data, satellite dynamics data, spatio-temporal position data, telemetry data, etc., it corrects the orbital parameters of the satellite, further improving the calculation accuracy of the orbital elements.

[0181] Corresponding to the method embodiment of the present application, the present application also provides an ultra-low orbit guarantee system for low-Earth orbit satellites.

[0182] Figure 8 It is a schematic structural diagram of an ultra-low orbit guarantee system for a low-Earth orbit satellite provided by an embodiment of the present application. As Figure 8 shown, the ultra-low orbit guarantee system 800 for the low-Earth orbit satellite may include: a weight distribution module 810, a determination module 820, a coordinate system conversion module 830, an error evaluation module 840, an orbit correction module 850, and a risk warning module 860.

[0183] Among them, the weight allocation module 810 is used to input the historical environmental parameters corresponding to the target prediction period and M target prediction values into the first meta-model, and allocate M first weight values to the M target prediction values through the first meta-model, where the M target prediction values are output by M basic prediction models based on the historical environmental parameters corresponding to the target prediction period; the determination module 820 is used to perform weighted summation on the M target prediction values by using the M first weight values to determine the atmospheric density of the target prediction period; the determination module 820 is further used to determine the atmospheric drag perturbation term of the satellite at the current moment based on the real-time windward mass ratio, drag coefficient, atmospheric density of the satellite at the current moment, and the speed scalar and speed vector of the satellite relative to the atmosphere, where the current moment is within the target prediction period; the determination module 820 is further used to determine the satellite orbit elements at the current moment based on the atmospheric drag perturbation term, the conventional perturbation term, and the small thrust perturbation term, and input the satellite orbit elements at the current moment into the orbit mechanics model; the coordinate system conversion module 830 is used to determine the velocity position parameters of the satellite at the prediction moment in the geocentric inertial rectangular ICRS coordinate system based on the orbit mechanics model, obtain the first velocity position parameters, and convert the first velocity position parameters into the Earth-centered Earth-fixed rectangular ECEF coordinate system to obtain the second velocity position parameters at the prediction moment; the error evaluation module 840 is used to input the time characteristics, satellite state characteristics, space environment characteristics, and the second velocity position parameters at the current moment into the orbit correction model, so that the orbit correction model evaluates the error of the second velocity position parameters to obtain the target error prediction value; the orbit correction module 850 is used to adjust the second velocity position parameters based on the target error prediction value to obtain the target velocity position parameters of the satellite at the prediction moment; the risk warning module 860 is used to perform orbit decay risk warning based on the target velocity position parameters of the satellite at the prediction moment.

[0184] In some embodiments of the present application, the system further includes: an acquisition module, configured to, before determining the atmospheric drag perturbation term of the satellite at the current moment, in the case where it is determined that the velocity direction of the satellite is the windward direction, acquire the wind direction unit vector of the satellite in the on-board coordinate system, the satellite solar panel area, and the unit normal vector of the plane where the satellite solar panel is located in the satellite body coordinate system; the coordinate system conversion module is further configured to convert the wind direction unit vector of the on-board coordinate system to the satellite body coordinate system; the determination module is further configured to determine the real-time windward area of the satellite based on the satellite solar panel area, and the wind direction unit vector and the unit normal vector in the satellite body coordinate system.

[0185] In some embodiments of the present application, the determination module is specifically configured to: calculate the absolute value of the product of the wind direction unit vector and the unit normal vector in the satellite body coordinate system; multiply the satellite solar panel area by the absolute value to obtain the real-time windward area of the satellite.

[0186] In some embodiments of the present application, it further includes: an acquisition module, further configured to, before determining the satellite orbit elements at the current moment based on the atmospheric drag perturbation term, the conventional perturbation term, and the small thrust perturbation term, and when the thrust magnitude of the ion thruster of the satellite is constant, acquire the thrust vector of the satellite engine in the satellite body coordinate system; a coordinate transformation module, further configured to acquire the first rotation matrix of the satellite attitude angle relative to the geocentric inertial coordinate system, and based on the first rotation matrix, transform the thrust vector of the satellite engine in the satellite body coordinate system to the geocentric inertial coordinate system to obtain the thrust components of the thrust vector on the three coordinate axes of the geocentric inertial coordinate system; a combination module, configured to combine the satellite mass, the real-time on / off state of the engine, and the thrust components on the three coordinate axes to obtain the small thrust perturbation term.

[0187] In some embodiments of the present application, the determination module includes: a first determination unit, configured to determine the first acceleration corresponding to the small thrust perturbation term based on the satellite mass, the real-time on / off state of the engine, and the thrust components on the three coordinate axes; a second determination unit, configured to determine the acceleration generated by non-conservative forces based on the first acceleration corresponding to the small thrust perturbation term and the second acceleration corresponding to the atmospheric drag perturbation term; a third determination unit, configured to determine the satellite orbit elements based on the acceleration generated by non-conservative forces, the earth gravitational acceleration, and the third body gravitational acceleration, where the earth gravitational acceleration and the third body gravitational acceleration are calculated from the conventional perturbation term.

[0188] In some embodiments of the present application, the first determination unit is specifically configured to: substitute the real-time on / off state of the engine into the working state function to obtain a function value, where different function values correspond to the on state and the off state; determine the first acceleration corresponding to the small thrust perturbation term based on the function value, the satellite mass, and the thrust components on the three coordinate axes.

[0189] In some embodiments of the present application, it further includes: an acquisition module, configured to acquire target training data before inputting historical environmental parameters and M target prediction values corresponding to a target prediction period into a first meta-model, where the target training data includes historical environmental parameters, measured values, and M prediction values corresponding to each prediction period among multiple prediction periods, the measured value is the atmospheric density actually detected for each prediction period, and the M prediction values are obtained by M basic prediction models predicting the atmospheric density for each prediction period based on the historical environmental parameters of each prediction period; a model training module, configured to train the first meta-model using the target training data, so that the first meta-model assigns corresponding M first weight values to the M prediction values for each prediction period based on the historical environmental parameters corresponding to each prediction period; a determination module, configured to determine that the training of the first meta-model is completed when the first loss function value of the first meta-model satisfies a first preset condition, where the first loss function value is the mean square error between the first prediction value and the measured value corresponding to each prediction period among multiple prediction periods, and the first prediction value is obtained by weighted summing the M prediction values corresponding to each prediction period using the M first weight values assigned by the first meta-model for each prediction period.

[0190] In some embodiments of the present application, both the second velocity-position parameter and the target velocity-position parameter include three position component parameters and three velocity component parameters. The orbit correction model is constructed by six sub-fusion models. Each sub-fusion model consists of multiple base models and a target fusion model. Each base model is used to evaluate the deviation between the predicted value and the true value of the corresponding component parameter in the velocity-position parameter at the prediction moment, and obtain the predicted component error associated with the corresponding component parameter. The target fusion model is used to fuse the multiple predicted component errors output by the multiple base models to obtain the component error predicted value output by the sub-fusion model; the multiple base models include at least two of lightGBM, random forest algorithm, support vector machine (SVM) algorithm, long short-term memory network (LSTM), and encoder transformer encoder.

[0191] The ultra-low orbit guarantee system for a near-earth satellite provided by the embodiments of the present application can achieve Figure 1-7 each process implemented by the service platform in the method embodiments, and can achieve the same technical effects. To avoid repetition, it will not be elaborated here.

[0192] Figure 9 It is a schematic hardware structure diagram of an electronic device provided by the embodiments of the present application.

[0193] As Figure 9 shown, the electronic device 900 includes a memory 901, a processor 902, and a computer program stored on the memory 901 and executable on the processor 902.

[0194] In one example, the above-mentioned processor 902 may include a central processing unit (CPU), or an application specific integrated circuit (ASIC), or may be an integrated circuit configured to implement one or more embodiments of the present application.

[0195] The memory 901 may include a read-only memory (ROM), a random access memory (RAM), a magnetic disk storage medium device, an optical storage medium device, a flash memory device, or an electrical, optical, or other physical / tangible memory storage device. Thus, generally, the memory includes one or more tangible (non-transitory) computer-readable storage media (e.g., memory devices) encoded with software including computer-executable instructions, and when the software is executed (e.g., by one or more processors), it is operable to perform the operations described with reference to the card opening method in the embodiments of the first aspect of the present application.

[0196] The processor 902 runs a computer program corresponding to the executable program code by reading the executable program code stored in the memory 901, so as to implement the card opening method in the embodiments of the first aspect above.

[0197] In some examples, the electronic device 900 may further include a communication interface 903 and a bus 910. Among them, as Figure 9 shown, the memory 901, the processor 902, and the communication interface 903 are connected through the bus 910 and complete communication with each other.

[0198] The communication interface 903 is mainly used to implement communication between various modules, systems, units, and / or devices in the embodiments of the present application. The input device and / or output device may also be accessed through the communication interface 903.

[0199] Bus 910 includes hardware, software, or both, and couples components of electronic device 900 to each other. By way of example and not limitation, bus 910 may include an Accelerated Graphics Port (AGP) or other graphics bus, an Enhanced Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), a Hyper Transport (HT) interconnect, an Industry Standard Architecture (ISA) bus, an InfiniBand interconnect, a Low Pin Count (LPC) bus, a memory bus, a Micro Channel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-E) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local Bus (VLB) bus, or other suitable bus or a combination of two or more of these. Where appropriate, bus 910 may include one or more buses. Although embodiments of the present application describe and illustrate specific buses, the present application contemplates any suitable bus or interconnect.

[0200] The electronic device provided by the embodiments of the present application can implement Figure 1-7 each process implemented by the electronic device in the method embodiments, and can achieve the same technical effects. To avoid repetition, details are not described herein again.

[0201] In combination with the method for ensuring the ultra-low orbit of a near-earth satellite in the above embodiments, an embodiment of the present application can be implemented by providing a computer storage medium. Computer program instructions are stored on the computer storage medium; when the computer program instructions are executed by a processor, the steps of any one of the methods for ensuring the ultra-low orbit of a near-earth satellite in the above embodiments are implemented.

[0202] In combination with the method for ensuring the ultra-low orbit of a near-earth satellite in the above embodiments, an embodiment of the present application can be implemented by providing a computer program product. The (computer) program product is stored in a non-volatile storage medium, and when the program product is executed by at least one processor, the steps of any one of the methods for ensuring the ultra-low orbit of a near-earth satellite in the above embodiments are implemented.

[0203] Another embodiment of the present application provides a chip, which includes a processor and a communication interface. The communication interface is coupled to the processor. The processor is used to run programs or instructions to implement each process of the above-mentioned embodiment of the method for ensuring the ultra-low orbit of near-earth satellites, and can achieve the same technical effects. To avoid repetition, it will not be elaborated here. It should be understood that the chip mentioned in the embodiment of the present application can also be referred to as a system-on-chip, system chip, chip system, or system-on-a-chip, etc.

[0204] It should be clear that the present application is not limited to the specific configurations and processes described above and shown in the figures. For the sake of brevity, the detailed description of known methods is omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of the present application is not limited to the specific steps described and shown. Those skilled in the art can make various changes, modifications, and additions, or change the order between steps after understanding the spirit of the present application.

[0205] The functional blocks shown in the above structural block diagrams can be implemented as hardware, software, firmware, or a combination thereof. When implemented in hardware, it can be, for example, an electronic circuit, an application specific integrated circuit (ASIC), appropriate firmware, a plug-in, a functional card, and so on. When implemented in software, the elements of the present application are programs or code segments used to perform the required tasks. The program or code segment can be stored in a machine-readable medium, or transmitted through a data signal carried in a carrier wave on a transmission medium or a communication link. "Machine-readable medium" can include any medium capable of storing or transmitting information. Examples of machine-readable media include electronic circuits, semiconductor memory devices, ROM, flash memory, erasable ROM (EROM), floppy disks, CD-ROMs, optical discs, hard disks, fiber optic media, radio frequency (RF) links, and so on. The code segment can be downloaded via a computer network such as the Internet, intranet, etc.

[0206] It should also be noted that the exemplary embodiments mentioned in the present application describe some methods or systems based on a series of steps or systems. However, the present application is not limited to the order of the above steps, that is, the steps can be executed in the order mentioned in the embodiments, or different from the order in the embodiments, or several steps can be executed simultaneously.

[0207] Aspects of the present disclosure have been described above with reference to the flowchart illustrations and / or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the present disclosure. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, enable the implementation of the functions / acts specified in one or more blocks of the flowchart illustrations and / or block diagrams. Such a processor may be, without limitation, a general purpose processor, a special purpose processor, an application specific processor or a field programmable logic circuit. It will also be understood that each block of the block diagrams and / or flowchart illustrations, and combinations of blocks in the block diagrams and / or flowchart illustrations, can also be implemented by dedicated hardware performing the specified functions or acts, or by combinations of dedicated hardware and computer instructions.

[0208] As described above, the foregoing is only a specific implementation manner of the present application. Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, modules and units described above can refer to the corresponding processes in the foregoing method embodiments, and will not be elaborated herein. It should be understood that the protection scope of the present application is not limited thereto. Any person skilled in the art can easily think of various equivalent modifications or substitutions within the technical scope disclosed in the present application, and these modifications or substitutions should all be covered by the protection scope of the present application.

Claims

1. A method for ultra-low orbit support of a near-Earth satellite, characterized in that: include: Inputting historical environment parameters corresponding to the target prediction period and M target prediction values ​​into the first meta-model, and assigning M first weight values ​​to the M target prediction values ​​through the first meta-model, wherein the M target prediction values ​​are output by the M basic prediction models based on the historical environment parameters corresponding to the target prediction period; Using the M first weight values ​​to perform weighted summation on the M target prediction values, to determine the atmospheric density of the target prediction period; Based on the satellite's real-time windward surface mass ratio, drag coefficient, atmospheric density, and the satellite's velocity scalar and velocity vector relative to the atmosphere at the current moment, determine the satellite's atmospheric drag perturbation term at the current moment, where the current moment is in the target prediction period; Based on the atmospheric drag perturbation term, the conventional perturbation term and the small thrust perturbation term, the satellite orbital elements at the current moment are determined, and the satellite orbital elements at the current moment are input into the orbital mechanics model; Based on the orbital mechanics model and the satellite orbit elements, the velocity and position parameters of the satellite at the prediction time are determined in the Earth-centered inertial rectangular ICRS coordinate system to obtain the first velocity and position parameters, and the first velocity and position parameters are converted into the Earth-fixed rectangular ECEF coordinate system to obtain the second velocity and position parameters at the prediction time; Inputting the time characteristics, satellite state characteristics, space environment characteristics, and the second velocity position parameter at the current moment into the orbit correction model, so that the orbit correction model evaluates the error of the second velocity position parameter to obtain a target error prediction value, wherein the time characteristics include the month and DOY at the current moment, and the mean solar time of the longitude of the satellite at the current moment, DOY refers to the day of the year at the current moment, the satellite state characteristics include the mass and windward area of ​​the satellite, and the velocity position parameters of the satellite at the current moment in the ECEF coordinate system, and the space environment characteristics include the satellite position at the current moment in the geodetic coordinate system, the relative position of the main celestial body and the satellite, and the solar radiation parameters and geomagnetic parameters of the satellite; Adjusting the second speed position parameter based on the target error prediction value to obtain the target speed position parameter of the satellite at the prediction time; Orbital decay risk warning is carried out based on the target speed and position parameters of the satellite at the predicted time.

2. The method according to claim 1, characterized in that Before determining the satellite's atmospheric drag perturbation at the current moment, it also includes: When it is determined that the speed direction of the satellite is the windward direction, the wind direction unit vector of the satellite in the on-board coordinate system, the area of ​​the satellite solar panel, and the unit normal vector of the plane where the satellite solar panel is located in the satellite body coordinate system are obtained; Converting the wind direction unit vector of the on-satellite coordinate system to the satellite body coordinate system; Based on the satellite solar panel area, and the wind direction unit vector and unit normal vector in the satellite body coordinate system, the real-time windward area of ​​the satellite is determined.

3. The method according to claim 2, characterized in that Based on the satellite solar panel area, and the wind direction unit vector and unit normal vector on the satellite body coordinate system, the real-time windward area of ​​the satellite is determined, including: Calculate the absolute value of the product of the wind direction unit vector and the unit normal vector in the satellite body coordinate system; The satellite solar panel area is multiplied by the absolute value to obtain the real-time windward area of ​​the satellite.

4. The method according to claim 1, characterized in that Before determining the satellite orbit elements at the current moment based on the atmospheric drag perturbation term, the conventional perturbation term and the small thrust perturbation term, it also includes: When the thrust of the satellite's ion thruster is constant, the thrust vector of the satellite engine in the satellite's body coordinate system is obtained; Acquire a first rotation matrix of the satellite attitude angle relative to the geocentric inertial coordinate system, and based on the first rotation matrix, convert the thrust vector of the satellite engine in the satellite body coordinate system to the geocentric inertial coordinate system to obtain the thrust components of the thrust vector on the three coordinate axes of the geocentric inertial coordinate system; The satellite mass, the real-time on / off status of the engine, and the thrust components on the three coordinate axes are combined to obtain a small thrust perturbation term.

5. The method according to claim 4, characterized in that Based on the atmospheric drag perturbation, conventional perturbation and small thrust perturbation, the satellite orbit elements are determined, including: Determining a first acceleration corresponding to a small thrust perturbation term based on the mass of the satellite, the real-time on / off state of the engine, and the thrust components on the three coordinate axes; Determining the acceleration generated by the non-conservative force based on the first acceleration corresponding to the small thrust perturbation term and the second acceleration corresponding to the atmospheric drag perturbation term; The satellite orbit elements are determined based on the acceleration generated by the non-conservative force, the earth's gravitational acceleration and the third celestial body's gravitational acceleration, wherein the earth's gravitational acceleration and the third celestial body's gravitational acceleration are calculated by conventional perturbation terms.

6. The method according to claim 5, characterized in that Based on the mass of the satellite, the real-time on / off state of the engine and the thrust components on the three coordinate axes, determining a first acceleration corresponding to the small thrust perturbation term includes: Substitute the real-time on / off state of the engine into the working state function to obtain the function value, wherein the on state and the off state correspond to different function values; A first acceleration corresponding to the small thrust perturbation term is determined based on the function value, the satellite mass and the thrust components on the three coordinate axes.

7. The method according to claim 1, characterized in that Before inputting the historical environment parameters corresponding to the target prediction period and the M target prediction values ​​into the first meta-model, it also includes: Obtain target training data, wherein the target training data includes historical environmental parameters, measured values, and M predicted values ​​corresponding to each prediction period in multiple prediction periods, wherein the measured values ​​are actually detected atmospheric density of each prediction period, and the M predicted values ​​are obtained by predicting the atmospheric density of each prediction period by M basic prediction models based on the historical environmental parameters of each prediction period; Using the target training data to train the first meta-model, so that the first meta-model assigns corresponding M first weight values ​​to the M prediction values ​​of each prediction period based on the historical environment parameters corresponding to each prediction period; When the first loss function value of the first meta-model satisfies the first preset condition, it is determined that the training of the first meta-model is completed, wherein the first loss function value is the mean square error between the first predicted value and the measured value corresponding to each prediction period in multiple prediction periods, and the first predicted value is obtained by weighted summing up the M predicted values ​​corresponding to each prediction period using the M first weight values ​​assigned by the first meta-model in each prediction period.

8. The method according to claim 1, characterized in that: The second speed position parameter and the target speed position parameter both include three position component parameters and three speed component parameters. The track correction model is constructed by six sub-fusion models, each of which is composed of multiple base models and a target fusion model. Each base model is used to evaluate the deviation between the predicted value and the true value of the corresponding component parameter in the speed position parameter at the prediction moment to obtain the predicted component error associated with the corresponding component parameter. The target fusion model is used to fuse multiple predicted component errors output by multiple base models to obtain the component error prediction value output by the sub-fusion model. The multiple base models include at least two of lightGBM, random forest algorithm, support vector machine SVM algorithm, long short-term memory network LSTM, and encoder transformer encoder.

9. An ultra-low orbit support system for a near-Earth satellite, characterized in that: include: A weight assignment module, used for inputting historical environment parameters corresponding to the target prediction period and M target prediction values ​​into the first meta-model, and assigning M first weight values ​​to the M target prediction values ​​through the first meta-model, wherein the M target prediction values ​​are output by the M basic prediction models based on the historical environment parameters corresponding to the target prediction period; A determination module, configured to perform weighted summation on the M target prediction values ​​using the M first weight values ​​to determine the atmospheric density of the target prediction period; The determination module is further used to determine the atmospheric drag perturbation term of the satellite at the current moment based on the real-time windward surface mass ratio, drag coefficient, atmospheric density, and the velocity scalar and velocity vector of the satellite relative to the atmosphere at the current moment, wherein the current moment is in the target prediction period; The determination module is further used to determine the satellite orbital elements at the current moment based on the atmospheric drag perturbation term, the conventional perturbation term and the small thrust perturbation term, and input the satellite orbital elements at the current moment into the orbital mechanics model; A coordinate system conversion module is used to determine the velocity position parameters of the satellite at the prediction time in the Earth-centered inertial rectangular ICRS coordinate system based on the orbital mechanics model, obtain the first velocity position parameters, and convert the first velocity position parameters into the Earth-fixed rectangular ECEF coordinate system to obtain the second velocity position parameters at the prediction time; an error evaluation module, for inputting the time characteristics, satellite state characteristics, space environment characteristics, and the second velocity position parameter at the current moment into the orbit correction model, so that the orbit correction model evaluates the error of the second velocity position parameter to obtain a target error prediction value, wherein the time characteristics include the month and DOY at the current moment, and the mean solar time of the longitude of the satellite at the current moment, DOY refers to the day of the year at the current moment, the satellite state characteristics include the mass and windward area of ​​the satellite, and the velocity position parameters of the satellite at the current moment in the ECEF coordinate system, and the space environment characteristics include the satellite position at the current moment in the geodetic coordinate system, the relative position of the main celestial body and the satellite, and the solar radiation parameters and geomagnetic parameters of the satellite; An orbit correction module, used to adjust the second speed position parameter based on the target error prediction value to obtain the target speed position parameter of the satellite at the prediction time; The risk warning module is used to provide orbital decay risk warning based on the target speed position parameters of the satellite at the prediction time.

10. An electronic device, characterized in that: The electronic device comprises: a processor and a memory storing computer program instructions; when the electronic device executes the computer program instructions, the method according to any one of claims 1 to 8 is implemented.

Images