Geosynchronous orbit high-energy electron flux forecasting method, model and model construction method

Abstract

The invention provides a geosynchronous orbit high-energy electron flux forecasting method, a geosynchronous orbit high-energy electron flux forecasting model and a model construction method, and relates to the field of electron flux forecasting. The geosynchronous orbit high-energy electron flux forecasting method comprises the following steps: constructing a geosynchronous orbit high-energy electron flux forecasting model; inputting a time point T, a space coordinate of a space point in a radiation band and an average solar wind speed in a time range of [T-a hours, T-b hours] into the geosynchronous orbit high-energy electron flux forecasting model; calculating the flux of high-energy electrons passing through a space point at the height of the geosynchronous orbit in a calm period; and judging whether the spatial point at the height of the geosynchronous orbit is in the violent time or not, and outputting the high-energy electron flux passing through the spatial point at the height of the geosynchronous orbit in the calm period or the violent time. According to the high-energy electron flux forecasting method provided by the invention, the calm period or burst time flux forecasting mode can be automatically switched, and the forecasting precision under different geomagnetic activity conditions is remarkably improved.

Description

Technical Field

[0001] The present invention relates to the field of electron flux prediction, and in particular to a geosynchronous orbit high-energy electron flux prediction method, model and model construction method. Background Art

[0002] Earth's radiation belts are filled with high-energy charged particles trapped by the geomagnetic field. The flux of these particles varies with spatial location and is subject to disturbances during geomagnetic storms, substorms, and other eruptions. Because the radiation belts coincide with the orbits of most satellites, high-energy charged particles pose a serious threat to the safety of satellites and astronauts. Studies have shown that high-energy particles cause 40% of all satellite failures. Therefore, establishing a comprehensive forecasting model for the distribution of high-energy particles in the radiation belts has long been a goal.

[0003] Currently, commonly used radiation belt high-energy electron flux prediction models fall into two categories. Simple static models, such as NASA's AE-8 model, were proposed relatively recently and lack accuracy. These models lack the ability to adjust their predictions in real time based on space bursts, making them incapable of providing effective radiation warnings for satellites and astronauts. The other category is physical models based on numerical simulations. While these models offer relatively accurate predictions, they consume significant computing resources, have numerous parameters, and require long runtimes, making them inconvenient for widespread use. Therefore, we aim to develop a wide-area, dynamic empirical model of high-energy electrons in the radiation belts based on actual observational data. This model would not only provide real-time radiation warnings but also be convenient and efficient. Summary of the Invention

[0004] In view of the fact that the above-mentioned existing technologies can only simulate the electron flux in quiet periods or during storms, the present application provides a method and model for predicting high-energy electron flux in geosynchronous orbit, which can autonomously switch between forecasting in quiet periods and forecasting during storms, thereby improving the accuracy of electron flux forecasting.

[0005] In one aspect, the present invention provides a method for predicting high-energy electron flux in geosynchronous orbit, comprising:

[0006] Obtain existing electron flux data and construct a geosynchronous orbit high-energy electron flux prediction model;

[0007] Inputs to the geosynchronous orbit high-energy electron flux prediction model include: time point T, spatial coordinates of the spatial point in the radiation belt, and average solar wind speed within a time range of [Ta hours, Tb hours], where a>b;

[0008] Calculate the high-energy electron flux J passing through the space point at the geosynchronous orbit altitude at the time point T during the quiet period. quiet (E);

[0009] Input to the geosynchronous orbit high-energy electron flux prediction model: the average auroral electrocurrent index within the time range [Tc minutes, T], where c≠0;

[0010] If the average auroral electrocurrent index is less than 300 nT, output the high-energy electron flux J passing through the space point at the geosynchronous orbit altitude at the time point T during the quiet period. quiet (E);

[0011] If the average auroral electrocurrent index is greater than or equal to 300 nT, calculate and output the high-energy electron flux J passing through the spatial point at the geosynchronous orbit altitude at the time point T during the burst. storm (E).

[0012] Optionally, during the quiet period, the high-energy electron flux J passing through the spatial point at the geosynchronous orbit altitude at the time point T is calculated. quiet (E) comprising the following steps:

[0013] Determine the transition energy E b ;

[0014] Determine the baseline energy E a ;

[0015] Calculate the reference energy E of the space point passing through the geosynchronous orbit at the time point T during the quiet period a Electron flux J quiet (E a ):

[0016]

[0017] Calculate the flux J of high-energy electrons passing through the space point at the geosynchronous orbit altitude at the time point T during the quiet period. quiet (E):

[0018]

[0019] Among them, V sw represents the average solar wind speed, k ij are the coefficients of a quadratic polynomial, determined by fitting the data set used, MLAT * is the geomagnetic latitude value after normalization, β1 is the energy spectrum slope of the lower energy part, β2 is the energy spectrum slope of the higher energy part, E b represents the transition energy of the energy spectrum, and α represents the sharpness of the transition energy.

[0020] Optionally, the high-energy electron flux passing through the spatial point at the geosynchronous orbit altitude during the burst is:

[0021] J storm (E) = J quiet (E)*e Q ;

[0022] in,

[0023] Among them, J quiet is the flux during the quiet period at the same solar wind speed conditions and the same spatial coordinates. Q is called the exponential growth rate of the flux during the storm. MLT p is the MLT corresponding to the peak value of Q, H(E,AE) is the bias function, and G(E,AE) is the amplitude function.

[0024] Optionally, the MLAT * for:

[0025] Among them, MLT and MLAT are the input radiation belt space coordinates.

[0026] Optionally, the bias function H(E,AE) is:

[0027] The amplitude function G(E,AE) is:

[0028] Among them, h ij and g ij are the polynomial coefficients determined by fitting the data set used, and E0 is a benchmark energy value used to non-dimensionalize the electron energy.

[0029] Optionally, the reference energy E a =205keV;

[0030] During the quiet period, at the time point T, the flux J of high-energy electrons passing through the space point at the geosynchronous orbit altitude is quiet (E) is:

[0031]

[0032] Among them, V sw represents the average solar wind speed, k ij are the coefficients of a quadratic polynomial, determined by fitting the data set used, MLAT * is the geomagnetic latitude value after normalization, β1 is the energy spectrum slope of the lower energy part, β2 is the energy spectrum slope of the higher energy part, E b represents the transition energy of the energy spectrum, and α represents the sharpness of the transition energy.

[0033] Another aspect of the present invention provides a method for constructing a geosynchronous orbit high-energy electron flux prediction model, comprising:

[0034] Constructing a training database, wherein the training database includes historical electron flux data, spatial coordinate data, solar wind speed, and auroral electrocurrent index;

[0035] Design the model architecture, including designing the calm period prediction module and the storm prediction module; design the calm period prediction module including the down-turned double-fold power law spectrum and the pan-spectrum fitting function Obtain a model for the high-energy electron flux at any energy during a quiet period: The design of the storm prediction module includes superimposing the exponential growth term Q on the basis of the quiet period flux, and the high-energy electron flux module at any energy during the storm: J storm (E) = J quiet (E)*e Q ;

[0036] The calm period prediction module and the violent period prediction module are trained based on the training database.

[0037] Optionally, the method further includes preprocessing the data in the training database, wherein the preprocessing includes:

[0038] Data cleaning, removing outliers from the training database;

[0039] Normalize the spatial coordinates and convert MLAT into normalized parameter MLAT * ,

[0040] Another aspect of the present invention provides a geosynchronous orbit high-energy electron flux prediction model, comprising:

[0041] Data acquisition module, which acquires all-round electron flux data, solar wind speed, and auroral electrocurrent index in geosynchronous orbit;

[0042] A training database, used to store the data acquired by the data acquisition module;

[0043] A comparison module is used to determine whether a space point at the geosynchronous orbit altitude is in a storm time;

[0044] The high-energy electron flux prediction module during quiet periods is used to calculate the high-energy electron flux at any energy during quiet periods. The high-energy electron flux model at any energy during quiet periods is:

[0045]

[0046] in,

[0047] The high-energy electron flux prediction module during the burst is used to calculate the high-energy electron flux at any energy during the burst. The high-energy electron flux model at any energy during the burst is:

[0048] J storm (E) = J quiet (E)*e Q ;

[0049] in,

[0050] a flux output module, which obtains and outputs the flux data predicted by the high-energy electron flux prediction module during the quiet period or the high-energy electron flux prediction module during the storm;

[0051] Among them, V sw represents the average solar wind speed, k ij are the coefficients of a quadratic polynomial, determined by fitting the data set used, MLAT * is the geomagnetic latitude value after normalization, β1 is the energy spectrum slope of the lower energy part, β2 is the energy spectrum slope of the higher energy part, E b represents the transition energy of the energy spectrum, α represents the sharpness of the transition energy, J quiet is the flux during the quiet period under the same solar wind speed conditions and at the same spatial coordinates.

[0052] As described above, the geosynchronous orbit high-energy electron flux prediction method, model, and model construction method provided by the present invention have at least the following beneficial technical effects:

[0053] The geosynchronous orbit high-energy electron flux prediction model of the present invention automatically switches between the calm period flux prediction mode and the storm period flux prediction mode by real-time monitoring of the auroral electrocurrent index (AE index), which can significantly improve the prediction accuracy under different geomagnetic activity conditions; the present application integrates Beidou satellite electron flux data, solar wind parameters, AE index and other data, regularly incorporates new observation data to refit the model parameters, and ensures the accuracy of the forecast stage. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 A flow chart is shown for implementing a provided method for predicting high-energy electron flux in geosynchronous orbit.

[0055] Figure 2 Shown is a flow chart of the method for constructing a geosynchronous orbit high-energy electron flux prediction model provided by Example 2.

[0056] Figure 3 Shown is a schematic diagram of the composition of the geosynchronous orbit high-energy electron flux prediction model provided in Example 3.

[0057] Figure 4The diagram shows a comparison of the prediction results of the geosynchronous orbit high-energy electron flux prediction model provided in Example 3 and the prediction results of the prior art model. DETAILED DESCRIPTION

[0058] The following describes the embodiments of the present invention through specific examples. Those skilled in the art will readily understand the other advantages and benefits of the present invention from the disclosure herein. The present invention may also be implemented or applied through various other specific embodiments, and the details in this specification may be modified or altered based on different viewpoints and applications without departing from the spirit of the present invention.

[0059] It should be noted that the illustrations provided in this embodiment only illustrate the basic concept of the present invention in a schematic manner. Although the illustrations only show components related to the present invention and are not drawn according to the number, shape and size of components in actual implementation, the form, quantity, positional relationship and proportion of each component in actual implementation can be changed at will under the premise of realizing the technical solution of this party, and the component layout form may also be more complicated.

[0060] Example 1

[0061] This embodiment provides a method for predicting high-energy electron flux in geosynchronous orbit, such as Figure 1 As shown, a flow chart of the geosynchronous orbit high-energy electron flux prediction method provided by this embodiment is shown; the geosynchronous orbit high-energy electron flux prediction method provided by this embodiment includes: obtaining existing electron flux data, storing it in a training database, and constructing a geosynchronous orbit high-energy electron flux prediction model; inputting into the geosynchronous orbit high-energy electron flux prediction model: time point T, radiation belt spatial coordinates, and average solar wind speed within the time range of [Ta hours, Tb hours], where a>b; calculating the high-energy electron flux J at the geosynchronous orbit altitude at the time point T during the quiet period. quiet (E); inputting the average auroral electrocurrent index within the time range [Tc min, T] into the geosynchronous orbit high-energy electron flux prediction model, where c≠0; if the average auroral electrocurrent index is less than 300nT, outputting the high-energy electron flux J at the geosynchronous orbit altitude during the quiet period quiet (E); If the average auroral electrocurrent index is greater than or equal to 300 nT, calculate and output the high-energy electron flux J at the geosynchronous orbit altitude at the time point T during the storm period. storm (E).

[0062] Specifically, S1: obtain existing electron flux data, store them in a training database, and construct a geosynchronous orbit high-energy electron flux prediction model;

[0063] Specifically, the existing electron flux data include: high-energy electron flux data, solar wind speed data, and auroral electrocurrent index (AE index). The method for obtaining high-energy electron flux data includes using the BeiDou imaging electron spectrometer (BD-IES) on the BeiDou inclined synchronous orbit satellite. Specifically, the satellite orbit has an inclination of 55°, a period of 24 hours, and an altitude of approximately 37,000 km (5.8 R) above the ground. E , R E Earth radius), the BD-IES carried by BeiDou satellite can measure the electron flux of 50-600keV (kiloelectronvolts), with a field of view of 180°×20° and an energy resolution of In this embodiment, we use the omnidirectional average electron flux data provided by BD-IES, specifically covering 8 energy bands in the range of 50-600 keV; the solar wind velocity data can be obtained from the ACE satellite or related space weather monitoring services; the AE index can be obtained from the space weather parameter report of NOAA / SWPC (National Oceanic and Atmospheric Administration / Space Weather Prediction Center) or other related monitoring services.

[0064] Optionally, new observation data is regularly incorporated into the training database to keep the data in the training database regularly updated to ensure the accuracy of the forecast results.

[0065] Specifically, a geosynchronous orbit high-energy electron flux prediction model is constructed based on the data in the training database. The specific construction method is described in Example 2 of this application and will not be described in detail here.

[0066] Specifically, S2: input to the geosynchronous orbit high-energy electron flux prediction model: time point T, spatial coordinates of the spatial point in the radiation belt, and average solar wind speed within the time range of [Ta hours, Tb hours], where a>b.

[0067] Generally, the time point T is the current moment, which is used to mark the target time of the forecast. The geosynchronous orbit high-energy electron flux forecasting method provided in this embodiment integrates historical and real-time data to dynamically output the high-energy electron flux at the geosynchronous orbit altitude at the current moment T at a specific radiation space coordinate. Specifically, the space of the spatial point in the radiation belt includes: magnetic local time (Magnetic Local Time, MLT) and geomagnetic latitude (Magnetic Latitude, MLAT). Generally, magnetic local time (MLT) and geomagnetic latitude (MLAT) can be obtained by converting geographic longitude and latitude into geomagnetic coordinate system parameters through the International Geomagnetic Reference Field Model (IGRF) or the T89 magnetic field model, or the geomagnetic parameters can be directly obtained through satellites equipped with geomagnetic magnetic field detection instruments, or can be obtained through databases such as ground magnetic measurement data and space weather services. Specifically, the average solar wind speed (V sw) is the average solar wind speed within the time range of [Ta hours, Tb hours], where a is greater than b. Specifically, since it takes about 1-3 days for the solar wind to propagate from the sun to the Earth's magnetosphere, the time window should be selected to match the delay effect of solar wind disturbances on radiation belt electrons. Specifically, this embodiment uses the average solar wind speed (V) within the range of [T-25 hours, T-15 hours] sw ) to capture the cumulative compression of the magnetosphere by the high-speed flow of the solar wind.

[0068] Specifically, S3: Calculate the high-energy electron flux J passing through the space point at the geosynchronous orbit altitude at the time point T during the quiet period. quiet (E); specifically comprising the following steps:

[0069] S31, determine the transition energy E b According to the energy electron observation data of BD-IES, the electron energy spectrum at the synchronous orbital altitude during the quiet period is mainly a downward double power-law spectrum (DDPL) and satisfies the universal spectrum fitting function. According to the data fitting in the training database, the transition energy E can be determined. b , E b It is the critical point where the energy spectrum slope changes significantly. b =192keV is the critical point where the energy spectrum slope changes.

[0070] S32, determine the reference energy E a . Select the one that is closer to the transition energy E b The energy is the reference energy E a To ensure the reference energy E a The flux value is highly sensitive to the energy spectrum shape parameters (β1, β2, α), and these parameters can be determined more effectively through data fitting. Specifically, in this embodiment, 205 keV is selected as the reference energy.

[0071] S33, calculate the reference energy E of the space point passing through the geosynchronous orbit at time point T during the quiet period. a Electron flux J quiet (E a ).

[0072] Specifically,

[0073] Among them, k ij are the coefficients of a quadratic polynomial of two variables, determined by fitting the data set used. * is the normalized geomagnetic latitude value:

[0074]

[0075] Where MLT and MLAT in formula (2) are the input radiation belt spatial coordinates, both obtained from the training database; β1, β2 are the normalized geomagnetic latitude values ​​MLAT * There is a relationship with a quadratic function:

[0076]

[0077] Among them, parameter b 10 ,b 11 ,b 12 ,b 20 ,b 21 ,b 22 All are determined by fitting the data set used.

[0078] Specifically, in this embodiment, during the quiet period, at the time point T, the flux J of high-energy electrons with an energy of 205 keV passing through the spatial point at the geosynchronous orbit altitude is quiet (205keV) is:

[0079]

[0080] in, According to the training data set fitting, we can determine α = 10.1, and the parameter b 10 ,b 11 ,b 12 ,b 20 ,b 21 ,b 22 All are determined by fitting the data set used.

[0081] S34: Calculate the high-energy electron flux J passing through the space point at the geosynchronous orbit altitude at the time point T during the quiet period. quiet (E)

[0082] The electron energy spectrum at the synchronous orbital altitude during the quiet period satisfies the pan-spectrum fitting function, which is:

[0083]

[0084] It can be considered that the electron flux J of the reference energy electron passing through the space point in the geosynchronous orbit is quiet (E a ) and the flux of high-energy electrons passing through a point in space at the altitude of the geosynchronous orbit J quiet (E) In the same energy spectrum function at different energy points (E a and E), which are two independent observations of the same distribution. According to the pan-spectral fitting function, the ratio of the two can be expressed as:

[0085]

[0086] Then, by simply transforming Equation (5), we can obtain the flux J of high-energy electrons passing through the space point at the geosynchronous orbit altitude: quiet (E):

[0087]

[0088] Specifically, in this embodiment, during the quiet period, at the time point T, the flux J of high-energy electrons passing through the space point at the geosynchronous orbit altitude is quiet (E) is:

[0089]

[0090] Among them, V sw represents the average solar wind speed, k ij are the coefficients of a quadratic polynomial, determined by fitting the data set used, MLAT * is the geomagnetic latitude value after normalization, β1 is the energy spectrum slope of the lower energy part, β2 is the energy spectrum slope of the higher energy part, E b represents the transition energy of the energy spectrum, and α represents the sharpness of the transition energy.

[0091] Specifically, S4: inputting into the geosynchronous orbit high-energy electron flux prediction model: the average auroral electrocurrent index within the time range of [Tc minutes, T], where c≠0.

[0092] Generally, the AE index is an indicator for measuring the intensity of auroral electroconcentration activity, and is crucial for evaluating the impact of geomagnetic disturbances on the high-energy electron flux in the radiation belt. The present invention determines whether the geomagnetic activity at the current time T is calm or violent by evaluating the AE index.

[0093] Specifically, the time range [Tc minutes, T] for calculating the average AE index is not a fixed interval. The window length c must match the physical time scale of electron acceleration. If the window is too short (e.g., c = 10 minutes), the cumulative effect of the acceleration process may be missed; if it is too long (e.g., c = 6 hours), outdated signals may be introduced. Specifically, in this embodiment, c = 30 minutes is selected.

[0094] People usually use an AE index greater than or equal to 300nT as the standard for the occurrence of geomagnetic substorms. When the input average AE index is less than 300nT, the geomagnetic activity is considered to be in a quiet period; when the input average AE index is greater than or equal to 300nT, the geomagnetic activity is considered to be in a stormy period.

[0095] Specifically, S5: if the average auroral electrocurrent index is less than 300 nT, output the high-energy electron flux J passing through the space point at the geosynchronous orbit altitude at the time point T during the quiet period. quiet (E).

[0096] Specifically, S6: If the average auroral electrocurrent index is greater than or equal to 300 nT, calculate and output the high-energy electron flux J passing through the spatial point at the geosynchronous orbit altitude at the time point T during the burst. storm (E).

[0097] Specifically, the high-energy electron flux passing through the space point at the geosynchronous orbit altitude during the storm is considered to have an increase on the basis of the quiet period. The relationship between the high-energy electron flux passing through the space point at the geosynchronous orbit altitude during the storm and the quiet period is:

[0098] J storm (E) = J quiet (E)*e Q (7);

[0099] in,

[0100] Among them, J quiet is the flux during the quiet period at the same solar wind speed conditions and the same spatial coordinates, Q is called the exponential growth rate of the flux during the storm, and the parameter MLT is p is the MLT corresponding to the peak value of Q. Both the bias function H and the amplitude function G are related to the electron energy E and the AE index. The function form is expressed by a quadratic polynomial, that is:

[0101]

[0102] Among them, h ij and g ij are the polynomial coefficients determined by fitting the data set used, and E0 is a benchmark energy value used to non-dimensionalize the electron energy.

[0103] The geosynchronous orbit high-energy electron flux prediction method provided in this embodiment uses the real-time updated solar wind parameters and auroral electrocurrent index as input parameters, and can predict the high-energy electron flux at any spatial position at the current moment in real time.

[0104] Example 2

[0105] This embodiment provides a method for constructing a geosynchronous orbit high-energy electron flux prediction model. Figure 2 FIG. 1 is a flow chart showing a method for constructing a geosynchronous orbit high-energy electron flux prediction model provided by this embodiment; the method comprises the following steps:

[0106] S1: constructing a training database, wherein the training database includes historical electron flux data, spatial coordinate data, solar wind speed, and auroral electrocurrent index;

[0107] S2: Design the model architecture, including the design of the calm period prediction module and the storm prediction module; the design of the calm period prediction module includes the downward double-fold power law spectrum and the pan-spectrum fitting function Obtain a model for the high-energy electron flux at any energy during a quiet period: The design of the storm prediction module includes superimposing the exponential growth term Q on the basis of the quiet period flux, and the high-energy electron flux module at any energy during the storm: J storm (E) = J quiet (E)*e Q ;

[0108] S3: Training the calm period prediction module and the violent period prediction module based on the training database.

[0109] Specifically, S1: constructing a training database, which includes historical electron flux data, spatial coordinate data, solar wind speed, and auroral electrocurrent index. This includes steps such as data collection, constructing a training database, and data preprocessing.

[0110] Data sources include: historical electron flux data are derived from the omnidirectional average electron flux data in the range of 50-600 keV obtained by the imaging electron spectrometer (BD-IES) on the BeiDou satellite; spatial coordinate data (magnetic local time (MLT) and geomagnetic latitude (MLAT)) are obtained by converting the International Geomagnetic Reference Field Model (IGRF) or direct satellite detection data; solar wind speed is obtained from solar wind monitoring satellites such as ACE and DSCOVR; auroral electrocurrent index (AE index) is calculated based on real-time magnetic field measurement data from the global geomagnetic station network.

[0111] Building a database includes storing data in a table database, including time, MLT, MLAT, flux of each energy level, V sw , AE index and other fields.

[0112] Data preprocessing includes data cleaning, removing outliers in the training database; spatial coordinate normalization, converting MLAT into normalized parameter MLAT * , It also includes feature extraction: extracting the solar wind speed and AE index within a specific time.

[0113] Specifically, S2: Design the model architecture, including designing a calm period prediction module and a storm period prediction module.

[0114] Specifically, a quiet period prediction module was designed. Based on the BD-IES energy electron observation data, it was shown that the electron energy spectrum at the geostationary orbital altitude during quiet periods is primarily a downward double power-law (DDPL) spectrum that satisfies the pan-spectral fitting function. This includes the following steps:

[0115] First, the transition energy E is determined by fitting the data in the training database. b , E b It is the critical point where the energy spectrum slope changes significantly.

[0116] Second, choose a value that is closer to the turning energy E. b The energy is the reference energy E a To ensure the reference energy E a The flux value of is more sensitive to the energy spectrum shape parameters (β1, β2, α), and these parameters can be determined more effectively by data fitting.

[0117] Third, calculate the reference energy E of the space point passing through the geosynchronous orbit at the time point T during the quiet period. a Electron flux J quiet (E a ).

[0118] Specifically,

[0119] Among them, k ij are the coefficients of a quadratic polynomial of two variables, determined by fitting the data set used. * is the normalized geomagnetic latitude value:

[0120]

[0121] Where MLT and MLAT in formula (2) are the input radiation belt spatial coordinates, both obtained from the training database; β1, β2 are the normalized geomagnetic latitude values ​​MLAT * There is a relationship with a quadratic function:

[0122]

[0123] Among them, parameter b 10 ,b 11 ,b 12 ,b 20 ,b 21 ,b 22 All are determined by fitting the data set used.

[0124] Fourth, calculate the high-energy electron flux J passing through the space point at the geosynchronous orbit altitude at the time point T during the quiet period.quiet (E), the electron energy spectrum at the synchronous orbital altitude during the quiet period satisfies the pan-spectrum fitting function, which is:

[0125]

[0126] It can be considered that the electron flux J of the reference energy electron passing through the space point in the geosynchronous orbit is quiet (E a ) and the flux of high-energy electrons passing through a point in space at the altitude of the geosynchronous orbit J quiet (E) In the same energy spectrum function at different energy points (E a and E), which are two independent observations of the same distribution. According to the pan-spectral fitting function, the ratio of the two can be expressed as:

[0127]

[0128] Then, by simply transforming Equation (5), we can obtain the flux J of high-energy electrons passing through the space point at the geosynchronous orbit altitude: quiet (E):

[0129]

[0130] Specifically, a storm prediction module is designed.

[0131] The high-energy electron flux passing through the space point at the geosynchronous orbit altitude during the storm is considered to have an increase on the basis of the quiet period. The relationship between the high-energy electron flux passing through the space point at the geosynchronous orbit altitude during the storm and the quiet period is:

[0132] J storm (E) = J quiet (E)*e Q (7);

[0133] in,

[0134] Among them, J quiet is the flux during the quiet period at the same solar wind speed conditions and the same spatial coordinates, Q is called the exponential growth rate of the flux during the storm, and the parameter MLT is p is the MLT corresponding to the peak value of Q. Both the bias function H and the amplitude function G are related to the electron energy E and the AE index. The function form is expressed by a quadratic polynomial, that is:

[0135]

[0136] Among them, h ij and g ijare the polynomial coefficients determined by fitting the data set used, and E0 is a benchmark energy value used to non-dimensionalize the electron energy.

[0137] Specifically, S3: training the calm period prediction module and the violent period prediction module based on the training database.

[0138] Optionally, the calm period prediction module or the storm period prediction module is trained in stages. The polarity of the model is trained based on the data in the training database. Optionally, the calm period prediction module is trained first, and its parameters are fixed before training the storm period prediction module. Specifically, the training method includes using error analysis combined with adjusting the coefficient of determination. Evaluate the goodness of fit.

[0139] Specifically, it also includes constructing a model selection module. By comparing the size of the input average auroral electrocurrent index, it is judged whether the space point at the geosynchronous orbit altitude is in a storm period, and then the choice is made to use the quiet period module or the storm period module for calculation. Generally, by evaluating the AE index, it is judged whether the geomagnetic activity state at the current time T is a quiet period or a storm period. Specifically, people usually use an AE index greater than or equal to 300nT as the standard for the occurrence of geomagnetic substorms. When the input average AE index is less than 300nT, the geomagnetic activity is considered to be in a quiet period; when the input average AE index is greater than or equal to 300nT, the geomagnetic activity is considered to be in a storm period. Optionally, the specific judgment size can be adjusted according to actual conditions.

[0140] Optionally, the method for constructing a geosynchronous orbit high-energy electron flux prediction model also includes model integration and implementation deployment. The quiet period prediction module and the storm period prediction module are integrated into the same interface, and MLT, MLAT, V sw , AE index, and output the full energy flux forecast value.

[0141] The method for constructing a geosynchronous orbit high-energy electron flux prediction model provided in this embodiment combines data fusion and dynamic mode conversion to achieve high-precision, low-latency prediction of geosynchronous orbit high-energy electron flux. Compared with traditional static models, the model achieves accurate dynamic prediction of radiation belt electrons while ensuring computational efficiency.

[0142] Example 3

[0143] This embodiment provides a geosynchronous orbit high-energy electron flux prediction model, such as Figure 3As shown, a schematic diagram of the composition of the geosynchronous orbit high-energy electron flux prediction model provided by this embodiment is shown. The geosynchronous orbit high-energy electron flux prediction model provided by this embodiment includes a data acquisition module for acquiring all-round electron flux data, solar wind speed, and auroral electrocurrent index in the geosynchronous orbit; a training database for storing data acquired by the data acquisition module; a comparison module for determining whether a spatial point at the geosynchronous orbit altitude is in a storm period; and a quiet period high-energy electron flux prediction module for calculating the high-energy electron flux at any energy in the quiet period. The high-energy electron flux model at any energy in the quiet period is:

[0144]

[0145] in,

[0146] The high-energy electron flux prediction module during the burst is used to calculate the high-energy electron flux at any energy during the burst. The high-energy electron flux model at any energy during the burst is:

[0147] J storm (E) = J quiet (E)*e Q ;

[0148] in,

[0149] a flux output module, which obtains and outputs the flux data predicted by the high-energy electron flux prediction module during the quiet period or the high-energy electron flux prediction module during the storm;

[0150] Among them, V sw represents the average solar wind speed, k ij are the coefficients of a quadratic polynomial, determined by fitting the data set used, MLAT * is the geomagnetic latitude value after normalization, β1 is the energy spectrum slope of the lower energy part, β2 is the energy spectrum slope of the higher energy part, E b represents the transition energy of the energy spectrum, α represents the sharpness of the transition energy, J quiet is the flux during the quiet period under the same solar wind speed conditions and at the same spatial coordinates.

[0151] like Figure 4 , which is a schematic diagram showing a comparison of the prediction results of the geosynchronous orbit high-energy electron flux prediction model provided by this embodiment and the AE-8max model of the prior art; Figure 4 The red dots represent the prediction results of the prior art AE-8max model, and the black dots represent the prediction results of the geosynchronous orbit high-energy electron flux prediction model provided by this embodiment. Figure 4The horizontal axis represents the actual observed electron flux, and the vertical axis represents the electron flux predicted by the model. Figure 4 It can be seen that the closer the electron flux predicted by the model is to the actually observed electron flux, the more concentrated the predicted flux is on both sides of the diagonal.

[0152] Specifically, in this embodiment, (Adjusted Coefficient of Determination) quantitatively evaluates the goodness of fit of the model:

[0153]

[0154] Where N is the number of samples, p is the number of parameters of the model, and y i is the actual observed value, is the mean of the observations, is the model's predicted value. The closer it is to 1, the better the forecast effect.

[0155] Depend on Figure 4 It can be seen that the geosynchronous orbit high-energy electron flux prediction model provided by this embodiment has Existing model AE-8max It can be seen that the model provided by this embodiment is closer to 1, so the prediction result of the geosynchronous orbit high-energy electron flux prediction model provided by this embodiment is more accurate.

[0156] The above embodiments are merely illustrative of the principles and effects of the present invention and are not intended to limit the present invention. Anyone skilled in the art may modify or alter the above embodiments without departing from the spirit and scope of the present invention. Therefore, all equivalent modifications or alterations made by one of ordinary skill in the art without departing from the spirit and technical principles disclosed herein are intended to be covered by the claims of the present invention.

Claims

1. A method for predicting high-energy electron flux in geosynchronous orbit, characterized in that: include: Obtain existing electron flux data and construct a geosynchronous orbit high-energy electron flux prediction model; Inputs to the geosynchronous orbit high-energy electron flux prediction model include: time point T, spatial coordinates of the spatial point in the radiation belt, and average solar wind speed within a time range of [Ta hours, Tb hours], where a>b; Calculate the high-energy electron flux J passing through the space point at the geosynchronous orbit altitude at the time point T during the quiet period. quiet (E); Input to the geosynchronous orbit high-energy electron flux prediction model: the average auroral electrocurrent index within the time range [Tc minutes, T], where c≠0; If the average auroral electrocurrent index is less than 300 nT, output the high-energy electron flux J passing through the space point at the geosynchronous orbit altitude at the time point T during the quiet period. quiet (E); If the average auroral electrocurrent index is greater than or equal to 300 nT, calculate and output the high-energy electron flux J passing through the spatial point at the geosynchronous orbit altitude at the time point T during the burst. storm (E).

2. The method for predicting geosynchronous orbit high-energy electron flux according to claim 1, characterized in that: Calculate the high-energy electron flux J passing through the space point at the geosynchronous orbit altitude at the time point T during the quiet period. quiet (E) The following steps are involved: Determine the transition energy E b ; Determine the baseline energy E a ; Calculate the reference energy E of the space point passing through the geosynchronous orbit at the time point T during the quiet period a Electron flux J quiet (E a ): Calculate the flux J of high-energy electrons passing through the space point at the geosynchronous orbit altitude at the time point T during the quiet period. quiet (E): Among them, V sw represents the average solar wind speed, k ij are the coefficients of a quadratic polynomial, determined by fitting the data set used, MLAT * is the geomagnetic latitude value after normalization, β1 is the energy spectrum slope of the lower energy part, β2 is the energy spectrum slope of the higher energy part, E b represents the transition energy of the energy spectrum, and α represents the sharpness of the transition energy.

3. The method for predicting geosynchronous orbit high-energy electron flux according to claim 1, wherein: The high-energy electron flux passing through the space point at the geosynchronous orbit altitude during the burst is: J storm (E)=J quiet (Sign in Q ; in, Among them, J quiet is the flux during the quiet period at the same solar wind speed conditions and the same spatial coordinates. Q is called the exponential growth rate of the flux during the storm. MLT p is the MLT corresponding to the peak value of Q, H(E,AE) is the bias function, and G(E,AE) is the amplitude function.

4. The method for predicting geosynchronous orbit high-energy electron flux according to claim 2, characterized in that: The MLAT * for: Among them, MLT and MLAT are the input radiation belt space coordinates.

5. The method for predicting geosynchronous orbit high-energy electron flux according to claim 3, characterized in that: The bias function H(E,AE) is: The amplitude function G(E,AE) is: Among them, h ij and g ij are the polynomial coefficients determined by fitting the data set used, and E0 is a benchmark energy value used to non-dimensionalize the electron energy.

6. The method for predicting geosynchronous orbit high-energy electron flux according to claim 2, characterized in that: The reference energy E a =205keV; During the quiet period, at the time point T, the flux J of high-energy electrons passing through the space point at the geosynchronous orbit altitude is quiet (E) is: Among them, V sw represents the average solar wind speed, k ij are the coefficients of a quadratic polynomial, determined by fitting the data set used, MLAT * is the geomagnetic latitude value after normalization, β1 is the energy spectrum slope of the lower energy part, β2 is the energy spectrum slope of the higher energy part, E b represents the transition energy of the energy spectrum, and α represents the sharpness of the transition energy.

7. A method for constructing a geosynchronous orbit high-energy electron flux prediction model, characterized in that: include: Constructing a training database, wherein the training database includes historical electron flux data, spatial coordinate data, solar wind speed, and auroral electrocurrent index; Design the model architecture, including designing the calm period prediction module and the storm prediction module; design the calm period prediction module including the down-turned double-fold power law spectrum and the pan-spectrum fitting function Obtain a model for the high-energy electron flux at any energy during a quiet period: The design of the storm prediction module includes superimposing the exponential growth term Q on the basis of the quiet period flux, and the high-energy electron flux module at any energy during the storm: J storm (E) = J quiet (E)*e Q ; The calm period prediction module and the violent period prediction module are trained based on the training database.

8. The method for constructing a geosynchronous orbit high-energy electron flux prediction model according to claim 7, characterized in that: The method further includes preprocessing the data in the training database, wherein the preprocessing includes: Data cleaning, removing outliers from the training database; Normalize the spatial coordinates and convert MLAT into normalized parameter MLAT * , 9. A geosynchronous orbit high-energy electron flux prediction model, characterized in that: include: Data acquisition module, which acquires all-round electron flux data, solar wind speed, and auroral electrocurrent index in geosynchronous orbit; A training database, used to store the data acquired by the data acquisition module; A comparison module is used to determine whether a space point at the geosynchronous orbit altitude is in a storm time; The high-energy electron flux prediction module during quiet periods is used to calculate the high-energy electron flux at any energy during quiet periods. The high-energy electron flux model at any energy during quiet periods is: in, The high-energy electron flux prediction module during the burst is used to calculate the high-energy electron flux at any energy during the burst. The high-energy electron flux model at any energy during the burst is: J storm (E)=J quiet (Sign in Q ; in, a flux output module, which obtains and outputs the flux data predicted by the high-energy electron flux prediction module during the quiet period or the high-energy electron flux prediction module during the storm; Among them, V sw represents the average solar wind speed, k ij are the coefficients of a quadratic polynomial, determined by fitting the data set used, MLAT * is the geomagnetic latitude value after normalization, β1 is the energy spectrum slope of the lower energy part, β2 is the energy spectrum slope of the higher energy part, E b represents the transition energy of the energy spectrum, α represents the sharpness of the transition energy, J quiet is the flux during the quiet period under the same solar wind speed conditions and at the same spatial coordinates.

Images