A method for extracting spatial differential perturbations based on in-situ satellite electron density data

Abstract

The present invention discloses a method for extracting spatial differential perturbations based on satellite in-situ electron density data, which includes collecting seismic data and space weather indices, obtaining observation data of in-situ electron density data, and preprocessing the observation data; obtaining spatial differential data based on the observation data and background data, and obtaining magnetically quiet differential data based on the space weather indices and the spatial differential data; clustering the magnetically quiet differential data to obtain abnormal perturbation data, extracting seismic perturbation data in the magnetically quiet differential data based on the seismic data, and extracting ionospheric abnormal data based on the abnormal perturbation data and the seismic perturbation data; constructing a spatial differential perturbation extraction model according to the spatial differential data and the ionospheric abnormal data, inputting the data to be analyzed into the spatial differential perturbation extraction model, and outputting the extraction result. This method can not only improve the accuracy of spatial differential perturbation extraction, but also has good interpretability and can be directly applied to the spatial differential perturbation extraction system.

Description

Technical Field

[0001] The present invention relates to the field of earthquake precursor anomaly research, and particularly to a method for extracting spatial differential perturbations based on in-situ electron density data of satellites. Background Art

[0002] Earthquake disasters can cause huge casualties and property losses, so the research on earthquake precursor anomalies has attracted much attention in the scientific community. Scientists have found that the precursor phenomena of the impending earthquake electromagnetic effect are relatively significant, and this kind of perturbation will be transmitted to the ionosphere through chemical electric fields, DC electric fields, acoustic gravity waves, electromagnetic waves, etc., thus causing changes in the characteristic parameters of the ionosphere.

[0003] Ionospheric measurement techniques can be roughly divided into two categories. One is ground-based observation, including ionospheric vertical sounders, oblique sounders, GNSS ground receiving stations for inverting TEC (Total Electron Content), incoherent scatter radars, VLF radio wave observations, HF radio wave observations, etc.; the other is space observation, including the top ionospheric detection of satellites and the in-situ observations of satellites and rockets. The ground-based ionospheric observation data has the characteristics of fixed location and continuous time, and the data analysis method can choose the common methods of time series analysis. For example, the sliding average method for a single point is a commonly used method for extracting ionospheric perturbations before earthquakes. Satellite detection can achieve all-weather, all-time and global observations, filling the gaps existing in ground observations; the global coverage data greatly increases the sample size of strong earthquake research and is conducive to the understanding of precursor statistical characteristics. However, the in-situ detection data of satellites changes both with time and space, and the commonly used time series analysis methods cannot be directly applied to satellite observation data. In view of this, through scientific research, a method for extracting spatial differential perturbations for in-situ satellite detection data is proposed. Summary of the Invention

[0004] The object of the present invention is to provide a method for extracting spatial differential perturbations based on in-situ electron density data of satellites.

[0005] To achieve the above object, the present invention is implemented according to the following technical solution:

[0006] The present invention includes the following steps:

[0007] Collect earthquake data and space weather indices, obtain the observation data of in-situ electron density data through polar-orbiting satellites, and preprocess the observation data;

[0008] Obtain spatial differential data according to the observation data and background data, and obtain magnetically quiet differential data according to the space weather index and the spatial differential data;

[0009] Cluster the magnetically quiet differential data to obtain abnormal perturbation data, perform seismic impact analysis on the magnetically quiet differential data based on the seismic data to obtain seismic perturbation data, and extract ionospheric abnormal data based on the abnormal perturbation data and the seismic perturbation data;

[0010] Construct a spatial differential perturbation extraction model according to the spatial differential data and the ionospheric abnormal data, input the data to be analyzed into the spatial differential perturbation extraction model, and output the extraction result.

[0011] Further, collect seismic data, obtain observation data of in-situ electron density data through polar-orbiting satellites, and the method for preprocessing the observation data includes:

[0012] The seismic data includes the epicenter location, earthquake magnitude, and earthquake time. Obtain in-situ electron density data within the range of 65° north latitude to 65° south latitude through the Langmuir probe of the polar-orbiting satellite. The time ranges of the observation data and the space weather index are before and during the earthquake;

[0013] Divide the observation data into day-side data and night-side data according to the observation time. The day-side data and night-side data of each day are a set of observation data. Divide the global longitude and latitude coordinate system into grids with a grid size of 5 longitude × 2.5 latitude. For each set of observation data, use the median of the observation data within the grid as the observation data of this grid.

[0014] Further, the method for obtaining spatial differential data according to the observation data and the background data includes:

[0015] Take the median of the observation data of the previous days of the observation date within the grid as the background data of the grid. The calculation formula for the spatial differential data is:

[0016]

[0017] Among them, is the spatial differential data of the observation date in the grid, is the observation data of the observation date in the grid, is the background data of the previous days of the observation date in the grid, is the background data of the previous days of the observation date in the grid, is the date index of the observation data. is the date index of the observation data.

[0018] Further, the method for obtaining magnetically quiet differential data according to the space weather index and the spatial differential data includes:

[0019] Remove the spatial differential data with latitudes greater than 50 degrees north and 50 degrees south, and remove the spatial differential data on non-magnetically quiet days. The criteria for magnetically quiet days are as follows:

[0020] F10.7 < 160 sfu, all 8 Kp values are less than 3, and the absolute values of all 24 Dst indices are less than 30 nT.

[0021] Furthermore, the method for clustering the magnetically quiet differential data to obtain abnormally disturbed data includes:

[0022] Normalize the magnetically quiet differential data and the position coordinates of its corresponding grid to establish data points. Preset two clustering clusters and cluster the data points:

[0023] Update formulas for attractiveness and belongingness:

[0024]

[0025] Among them, S(o,k) is the similarity between data point o and cluster center k, S(o,p) is the similarity between data point o and cluster center p, r t+1 (o,k) is the attractiveness of data point o to cluster center k in the (t + 1)-th iteration, A t (o,p) is the belongingness of data point o to cluster center p in the t-th iteration, A(o,k) is the belongingness of data point o to cluster center k in the (t + 1)-th iteration, r t+1 (p,k) is the attractiveness of cluster center p to cluster center k in the (t + 1)-th iteration,

[0026] Similarity between data points:

[0027]

[0028] Among them, σ is the bandwidth parameter, x o，e and x p，e are the e-th components of data point o and cluster center p respectively, D is the number of dimensions of the data points, w e is the weight of the e-th component of the data point, S(o,o) is the similarity between data point o and cluster center o, representing the tendency of data point o to be selected as the cluster center, h(o) is the preference parameter of data point o, h is the preference parameter, and ∈ is the noise tolerance parameter,

[0029] Update amplitudes of attractiveness and belongingness:

[0030]

[0031] Among them, γ is the attenuation coefficient used to control the update amplitude, 0 < γ < 1, δ is the iterative adjustment parameter of the attenuation coefficient, r t(o, k) is the attraction degree of data point o and clustering center k in the t-th iteration, A t (o, k) is the membership degree of data point o and clustering center k in the t-th iteration,

[0032] Stability index of the clustering cluster:

[0033]

[0034] Among them, H(C1, C2) is the matching degree of C1 and C2, WDX(p, λ) is the stability index of m clustering results obtained from a group of candidate parameters h and γ, H(C b , C b ′) is the matching degree of C b and C b ′, C1 and C2 are two different clustering results, C b and C b ′ are the b-th and b′-th clustering results obtained from candidate parameters h and γ respectively, c1 and c2 are the number of clustering clusters in C1 and C2 respectively, is the number of data points matching the -th clustering cluster in C1 and the -th clustering cluster in C2, is the number of data points matching the I-th clustering cluster in C1 and the -th clustering cluster in C2, is the number of data points matching the -th clustering cluster in C1 and the -th clustering cluster in C2. Select a group of candidate parameters with the largest stability index value,

[0035] Regard the data points in the clustering cluster with fewer data points as abnormal perturbation data.

[0036] Furthermore, based on the seismic data, perform seismic impact analysis on the quiet magnetic differential data to obtain seismic perturbation data. The method for extracting ionospheric anomaly data based on the abnormal perturbation data and the seismic perturbation data includes:

[0037] Extract the skewness, kurtosis and spatial change rate of the quiet magnetic differential data of the revisit orbit closest to the earthquake epicenter, and form spatial differential features,

[0038] Map the spatial differential features as the original data to a high-dimensional space through a non-linear transformation function, and find a hypersphere with the smallest volume in the high-dimensional space. The objective function is:

[0039]

[0040] Among them, is the radius of the hypersphere, n is the number of original data, Φ(·) is a non-linear transformation function, is the i-th original data, a is the center of the hypersphere, ξ i is the i-th relaxation factor, is the penalty parameter, β i is the local density weight, ||·|| represents the Euclidean distance, ρ i is the gradient threshold of the preset non-linear transformation function, is the average value of, and the dual form of the objective function is obtained by the method of Lagrange multipliers:

[0041]

[0042]

[0043] where, G(·) is the kernel function, α i and α j are respectively and the corresponding Lagrange coefficients. Solving the minimum value of the dual form of the objective function, we get as the j-th original data, v i is the gradient constraint the corresponding Lagrange coefficient. Solving the minimum value of the dual form of the objective function, the calculation formulas for the center and radius of the hypersphere are respectively:

[0044]

[0045] where, SV is the set of original data of the support vectors, is the v-th original data in SV,

[0046] the distance from the original data to the center of the hypersphere is

[0047]

[0048] where, is the distance to the center of the hypersphere, is the -th original data. If then is on or inside the hypersphere and belongs to normal data; otherwise the corresponding quiet-time magnetic differential data belongs to seismic disturbance data;

[0049] The overlapping data of the abnormal disturbance data and the seismic disturbance data is the ionospheric anomaly data.

[0050] Further, a method for constructing a spatial differential perturbation extraction model based on the spatial differential data and the ionospheric anomaly data includes:

[0051] Construct a spatial differential perturbation extraction model using a neural network algorithm, which includes an input layer, a function layer, and an output layer;

[0052] The input layer preprocesses the input data and converts it into spatial differential data, and filters out magnetically quiet differential data according to the space weather index; the function layer is a binary classification module that learns the correspondence between the magnetically quiet differential data and the ionospheric anomaly data; the output layer outputs the extraction result of the ionospheric anomaly data.

[0053] In a second aspect, an embodiment of the present application further provides an electronic device, including:

[0054] A processor; and a memory arranged to store computer-executable instructions, and when the executable instructions are executed, the processor executes the method steps described in the first aspect.

[0055] In a third aspect, an embodiment of the present application further provides a computer-readable storage medium, where the computer-readable storage medium stores one or more programs, and when the one or more programs are executed by an electronic device including a plurality of application programs, the electronic device executes the method steps described in the first aspect.

[0056] The beneficial effects of the present invention are:

[0057] The present invention is a method for extracting spatial differential perturbations based on satellite in-situ electron density data. Compared with the prior art, the present invention has the following technical effects:

[0058] The method proposed by the present invention has been integrated into the "Earthquake Demonstration Application" platform developed based on CSES-01 satellite data. After inputting the satellite in-situ electron density data, it can automatically generate daily spatial differential maps on the dayside and nightside for monitoring and tracking in earthquake key monitoring and defense areas.

[0059] The method proposed by the present application can not only be applied to in-situ electron density data, but also prepare data according to the input interface requirements for data such as in-situ detected temperature and single-band electromagnetic fields, perform spatial differential processing, and extract perturbation phenomena in the ionosphere. Therefore, this method also has good application prospects in multi-parameter comparative analysis, which is beneficial to the study of the synchronous changes of different physical parameter perturbations in the ionosphere in space and time. Description of the Drawings

[0060] Figure 1 It is a step flow chart of a method for extracting spatial differential perturbations based on satellite in-situ electron density data of the present invention;

[0061] Figure 2 The distribution diagrams of the observed data before and after preprocessing;

[0062] Figure 3 The spatial distribution diagram of the electron density on the dayside;

[0063] Figure 4 The spatial difference distribution diagram of the electron density;

[0064] Figure 5 The variation diagram of the electron density with latitude for different revisit orbits;

[0065] Figure 6 The statistical distribution diagram of the epicentral distance and the occurrence time of the anomaly;

[0066] Figure 7 The structural schematic diagram of an electronic device in the embodiments of this specification. Detailed implementation manners

[0067] The present invention will be further described below through specific embodiments. The illustrative embodiments and descriptions hereof are used to explain the present invention, but not to limit the present invention.

[0068] A method for extracting spatial differential perturbations based on satellite in-situ electron density data according to the present invention includes the following steps:

[0069] As Figure 1 shown, in this embodiment, it includes the following steps:

[0070] Collect seismic data and space weather indices, obtain the observed data of the in-situ electron density data through a polar-orbiting satellite, and preprocess the observed data;

[0071] In the actual evaluation, the seismic data is a Ms6.9 earthquake in Area A on August 5, 2018, with the epicenter coordinates being (8.258°S, 116.438°E). The polar-orbiting satellite used is the Chinese Seismic Electromagnetic Satellite (CSES-01). The observed data before and after preprocessing is as Figure 2 shown, Figure 2 The observation date of

[0072] is July 28, 2018;

[0073] In the actual evaluation, the background data is obtained based on the observed data of the previous 30 days before the observation date, Figure 4It is a spatial differential distribution map of electron density. The upper figure is the spatial differential data distribution map of the dayside data on July 31, the middle figure is the spatial differential data distribution map of the dayside data on August 4, and the lower figure is the spatial differential data distribution map of the nightside data on July 31. The red and blue pentagrams are the epicenter and its conjugate points, and the blue line in the middle of each subfigure represents the magnetic equator;

[0074] Cluster the magnetically quiet differential data to obtain abnormally disturbed data, perform seismic impact analysis on the magnetically quiet differential data based on the seismic data to obtain seismic disturbance data, and extract ionospheric anomaly data based on the abnormally disturbed data and the seismic disturbance data;

[0075] In actual evaluation, Figure 5 It is a graph of the variation of electron density with latitude for different revisit orbits. The upper figure is the dayside data on July 31, the middle figure is the dayside data on August 4, and the lower figure is the nightside data on July 31. The red line in the figure represents the orbit with ionospheric anomaly data extracted, and the other color curves represent its revisit orbits. The specific dates are shown in the legend. The red and blue pentagrams are the epicenter and its conjugate points, and the two blue vertical lines represent the latitudes 20° away from the epicenter; Figure 6 It is a statistical distribution map of the epicentral distance and the time of anomaly occurrence, showing that the seismic disturbance is statistically significant 11 days, 3 days, and 2 days before the earthquake;

[0076] Construct a spatial differential disturbance extraction model according to the spatial differential data and the ionospheric anomaly data, input the data to be analyzed into the spatial differential disturbance extraction model, and output the extraction result.

[0077] In this embodiment, seismic data is collected, and the method for preprocessing the observational data obtained by polar-orbiting satellites for in-situ electron density data includes:

[0078] The seismic data includes the epicenter location, earthquake magnitude, and earthquake time. The in-situ electron density data within the range of 65° north latitude to 65° south latitude is obtained through the Langmuir probe of the polar-orbiting satellite. The time ranges of the observational data and the space weather index are before and at the epicenter;

[0079] Divide the observational data into dayside data and nightside data according to the observation time. The dayside data and nightside data of each day are a set of observational data. Divide the global longitude-latitude coordinate system into grids with a grid size of 5 longitude × 2.5 latitude. For each set of observational data, take the median of the observational data within the grid as the observational data of this grid.

[0080] In this embodiment, the distributions of the observational data, background data, and spatial differential data are as Figure 3 shown Figure 3It is the spatial distribution map of the electron density on the dayside on May 8, 2021. The upper figure is the observed data, the middle figure is the background data, and the lower figure is the spatial difference data map of the observed data and the background data;

[0081] In this embodiment, the method for obtaining the spatial difference data from the observed data and the background data includes:

[0082] Take the median of the observed data in the grid for the days before the observation date as the background data of the grid. The calculation formula for the spatial difference data is:

[0083]

[0084] where, is the spatial difference data in the grid for the observation date, is the observed data in the grid for the observation date, is the background data in the grid for the days before the observation date, is the date index of the observed data.

[0085] In this embodiment, the method for obtaining the magnetically quiet difference data from the space weather index and the spatial difference data includes:

[0086] Remove the spatial difference data with latitudes greater than 50 degrees north and 50 degrees south, and remove the spatial difference data on non-magnetically quiet days. The criteria for magnetically quiet days are:

[0087] F10.7 < 160 sfu, all 8 Kp values are less than 3, and the absolute values of all 24 Dst indices are less than 30 nT.

[0088] In this embodiment, the method for clustering the magnetically quiet difference data to obtain the abnormal perturbation data includes:

[0089] Normalize the magnetically quiet difference data and the position coordinates of its grid, establish data points, preset two clustering clusters, and cluster the data points:

[0090] Update formulas for the attraction and the belonging degree:

[0091]

[0092]

[0093] where, S(o,k) is the similarity between the data point o and the clustering center k, S(o,p) is the similarity between the data point o and the clustering center p, r t+1(o, k) is the attractiveness of data point o and cluster center k in the (t + 1)-th iteration, A t (o, p) is the membership degree of data point o and cluster center p in the t-th iteration, A(o, k) is the membership degree of data point o and cluster center k in the (t + 1)-th iteration, r t+1 (p, k) is the attractiveness of cluster center p and cluster center k in the (t + 1)-th iteration,

[0094] Similarity between data points:

[0095]

[0096] where σ is the bandwidth parameter, x o,e and x p,e are the e-th dimensional components of data point o and cluster center p respectively, D is the number of dimensions of the data point, w e is the weight of the e-th dimensional component of the data point, S(o, o) is the similarity between data point o and cluster center o, representing the tendency of data point o to be selected as a cluster center, h(o) is the preference parameter of data point o, h is the preference parameter, ∈ is the noise tolerance parameter,

[0097] Update amplitude of attractiveness and membership degree:

[0098]

[0099] where γ is the attenuation coefficient, used to control the update amplitude, 0 < γ < 1, δ is the iterative adjustment parameter of the attenuation coefficient, r t (o, k) is the attractiveness of data point o and cluster center k in the t-th iteration, A t (o, k) is the membership degree of data point o and cluster center k in the t-th iteration,

[0100] Stability index of the cluster:

[0101]

[0102] where H(C1, C2) is the matching degree of C1 and C2, WDX(p, λ) is the stability index of m clustering results obtained by a set of candidate parameters h and γ, H(C b , C b′ ) is the matching degree of C b and C b′ C1 and C2 are two different clustering results, C b and C b′ are the b-th and b ′ -th clustering results obtained by candidate parameters h and γ respectively, c1 and c2 are the number of clusters in C1 and C2 respectively, is the The number of data points that match the th cluster in C2, is the number of data points that match the th cluster in C1 and the th cluster in C2, is the number of data points that match the th cluster in C1 and the

[0103] th cluster in C2. Select a set of candidate parameters with the largest stability index value.

[0104] Use the data points in the cluster with fewer data points as abnormal perturbation data.

[0105] In this embodiment, a method for extracting ionospheric anomaly data based on the seismic data for seismic impact analysis of the magnetic quiet differential data and based on the abnormal perturbation data and the seismic perturbation data includes:

[0106] Extract the skewness, kurtosis, and spatial change rate of the magnetic quiet differential data of the revisit orbit closest to the epicenter to form spatial differential features.

[0107]

[0108] Among them, is the radius of the hypersphere, n is the number of original data, Φ(·) is the nonlinear transformation function, is the i th original data, a is the center of the hypersphere, ξ is the penalty parameter, β i is the local density weight, ∥·∥ represents the Euclidean distance, ρ i is the preset gradient threshold of the nonlinear transformation function, is the average value of. Use the Lagrange multiplier method to obtain the dual form of the objective function:

[0109]

[0110] Among them, G(·) is the kernel function, α i and α j are respectively and the corresponding Lagrange coefficients. Solve the minimum value of the dual form of the objective function to obtain as the i th original data, v i is the gradient constraint For the corresponding Lagrange coefficients, by solving the minimum value of the dual form of the objective function, the calculation formulas for the center and radius of the hypersphere are respectively:

[0111]

[0112] Among them, SV is the original data set of support vectors, is the v-th original data in SV,

[0113] The distance from the original data to the center of the hypersphere is

[0114]

[0115] Among them, is the distance from to the center of the hypersphere, is the -th original data. If then is on or inside the hypersphere and belongs to normal data; otherwise the corresponding quiet-time magnetic differential data belongs to seismic disturbance data;

[0116] The overlapping data of abnormal disturbance data and seismic disturbance data is ionospheric anomaly data.

[0117] In this embodiment, the method for constructing a spatial difference perturbation extraction model based on the spatial difference data and the ionospheric anomaly data includes:

[0118] Use a neural network algorithm to construct a spatial difference perturbation extraction model. The neural network algorithm includes an input layer, a function layer, and an output layer;

[0119] The input layer preprocesses the input data and converts it into spatial difference data, and filters out quiet-time magnetic differential data according to the space weather index; the function layer is a binary classification module that learns the correspondence between quiet-time magnetic differential data and ionospheric anomaly data; the output layer outputs the extraction result of ionospheric anomaly data.

[0120] Figure 2 is a schematic structural diagram of an electronic device according to an embodiment of the present application. Please refer to Figure 2 , at the hardware level, the electronic device includes a processor, and optionally also includes an internal bus, a network interface, and a memory. Among them, the memory may include a memory, such as a high-speed random access memory (Random-Access Memory, RAM), and may also include a non-volatile memory, such as at least one disk memory, etc. Of course, the electronic device may also include other hardware required for other services.

[0121] The processor, network interface, and memory can be interconnected through an internal bus, which can be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component Interconnect) bus, an EISA (Extended Industry Standard Architecture) bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. For the sake of convenience in representation, Figure 2 only a bidirectional arrow is used in the figure, but it does not mean that there is only one bus or one type of bus.

[0122] Memory, used to store programs. Specifically, the program can include program code, and the program code includes computer operation instructions. The memory can include a memory and a non-volatile memory, and provide instructions and data to the processor.

[0123] The processor reads the corresponding computer program from the non-volatile memory into the memory and then runs it, forming a spatial differential perturbation extraction device based on satellite in-situ electron density data at the logical level. The processor executes the program stored in the memory and is specifically used to execute any one of the aforementioned spatial differential perturbation extraction methods based on satellite in-situ electron density data.

[0124] The above is as in the present application Figure 1A method for extracting spatial differential perturbations based on satellite in-situ electron density data disclosed in the illustrated embodiments can be applied to a processor or implemented by a processor. The processor may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by the integrated logic circuit of the hardware in the processor or the instructions in the form of software. The above-mentioned processor may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it may also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components. It can implement or execute the various methods, steps, and logic block diagrams disclosed in the embodiments of the present application. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc. The steps of the method disclosed in combination with the embodiments of the present application can be directly embodied as being executed and completed by the hardware decoding processor, or executed and completed by a combination of the hardware and software modules in the decoding processor. The software module may be located in a mature storage medium in the art such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory, or an electrically erasable programmable memory, a register, etc. This storage medium is located in the memory, and the processor reads the information in the memory and combines its hardware to complete the steps of the above method.

[0125] The electronic device can also execute Figure 1 a method for extracting spatial differential perturbations based on satellite in-situ electron density data, and implement Figure 1 the functions of the illustrated embodiments, which are not elaborated herein in the embodiments of the present application.

[0126] The embodiments of the present application also propose a computer-readable storage medium that stores one or more programs. The one or more programs include instructions that, when executed by an electronic device including a plurality of application programs, execute any of the foregoing methods for extracting spatial differential perturbations based on satellite in-situ electron density data.

[0127] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk memory, CD-ROM, optical memory, etc.) that contain computer-usable program code.

[0128] The present application is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to the embodiments of the present application. It should be understood that each flow and / or block in the flowchart and / or block diagram, and the combination of flows and / or blocks in the flowchart and / or block diagram, can be realized by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate means for realizing the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0129] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory generate a manufactured article including instruction means, and the instruction means realizes the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0130] These computer program instructions can also be loaded onto a computer or other programmable data processing device, so that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process, and thus the instructions executed on the computer or other programmable device provide steps for realizing the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0131] In a typical configuration, a computing device includes one or more processors (CPUs), an input / output interface, a network interface, and a memory.

[0132] The memory may include non-permanent memory in the computer-readable medium, in the form of random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash memory (flashRAM). The memory is an example of a computer-readable medium.

[0133] A computer-readable medium includes permanent and non-permanent, removable and non-removable media that can implement information storage by any method or technology. The information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory, or other memory technologies, compact disc read-only memory (CD-ROM), digital versatile disc (DVD), or other optical storage, magnetic cassette tapes, magnetic tape magnetic disk storage, or other magnetic storage devices, or any other non-transitory medium that can be used to store information accessible by a computing device. As defined herein, a computer-readable medium does not include transitory computer-readable media, such as modulated data signals and carrier waves.

[0134] It should also be noted that the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but also other elements not expressly listed, or elements that are inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the statement "comprising an..." does not exclude the presence of additional identical elements in the process, method, article, or apparatus that comprises the element.

[0135] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0136] The above are only the preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention shall be included within the protection scope of the present invention.

Claims

1. A spatial differential disturbance extraction method based on satellite in-situ electron density data, characterized in that: The following steps are involved: Collecting earthquake data and space weather index, obtaining observation data of in-situ electron density data through polar orbit satellites, and preprocessing the observation data; Obtaining spatial differential data according to the observation data and the background data, and obtaining magnetically quiet differential data according to the space weather index and the spatial differential data; Clustering the magnetically quiet differential data to obtain abnormal disturbance data, performing seismic impact analysis on the magnetically quiet differential data based on the seismic data to obtain seismic disturbance data, and extracting ionospheric abnormality data based on the abnormal disturbance data and the seismic disturbance data; A spatial differential disturbance extraction model is constructed according to the spatial differential data and the ionospheric anomaly data, the data to be analyzed is input into the spatial differential disturbance extraction model, and an extraction result is output.

2. According to claim 1, a spatial differential disturbance extraction method based on satellite in-situ electron density data is characterized in that: The method of collecting seismic data, obtaining observation data of in-situ electron density data through a polar-orbiting satellite, and preprocessing the observation data comprises: Earthquake data include epicenter location, earthquake magnitude and earthquake time. In-situ electron density data within the range of 65°N to 65°S are obtained through the Langmuir probe of the polar-orbiting satellite. The time range of observation data and space weather index is before and at the epicenter of the earthquake. The observation data are divided into dayside data and nightside data according to the observation time. The dayside data and nightside data of each day are a group of observation data. The global longitude and latitude coordinate system is gridded with a grid size of 5 longitude × 2.5 latitude. For each group of observation data, the median of the observation data in the grid is used as the observation data of this grid.

3. The spatial differential disturbance extraction method based on satellite in-situ electron density data according to claim 1 is characterized in that: The method for obtaining spatial difference data according to the observation data and background data comprises: Take the observation date within the grid The median of the observation data of the day is used as the background data of the grid, and the calculation formula of the spatial difference data is: in, is the observation day in the grid The spatial difference data of is the observation day in the grid The observation data, is the observation day in the grid forward Days of background data, The date index of the observation data.

4. The spatial differential disturbance extraction method based on satellite in-situ electron density data according to claim 1 is characterized in that: The method for obtaining magnetically quiet differential data according to the space weather index and the spatial differential data comprises: Remove the spatial difference data of latitudes greater than 50 degrees north latitude and 50 degrees south latitude, and remove the spatial difference data of non-magnetic quiet days. The standard of magnetic quiet days is: F10.7<160sfu, 8 Kp values ​​are all less than 3, and the absolute values ​​of 24 Dst indices are all less than 30nT.

5. The spatial differential disturbance extraction method based on satellite in-situ electron density data according to claim 1 is characterized in that: The method of clustering the magnetic quiet difference data to obtain abnormal disturbance data comprises: Normalize the magnetic quiet differential data and the position coordinates of the grid where they are located, establish data points, preset two clusters, and cluster the data points: Update formula for attraction and belonging: Among them, S(o,k) is the similarity between data point o and cluster center k, S(o,p) is the similarity between data point o and cluster center p, r t+1 (o,k) is the attraction between data point o and cluster center k in the t+1th iteration, A t (o,p) is the degree of belonging between data point o and cluster center p at the tth iteration, A(o,k) is the degree of belonging between data point o and cluster center k at the t+1th iteration, r t+1 (p,k) is the attraction between cluster center p and cluster center k in the t+1th iteration, Similarity between data points: Where σ is the bandwidth parameter, x o,e and x p,e are the e-th dimension components of the data point o and the cluster center p, respectively. D is the number of dimensions of the data point, and w e is the weight of the e-th dimension component of the data point, S(o,o) is the similarity between the data point o and the cluster center o, indicating the tendency of the data point o to be selected as the cluster center, h(o) is the preference parameter of the data point o, h is the preference parameter, ∈ is the noise tolerance parameter, Update range of attraction and belonging: Among them, γ is the attenuation coefficient, which is used to control the update amplitude, 0<γ<1, δ is the iterative adjustment parameter of the attenuation coefficient, r t (o,k) is the attraction between data point o and cluster center k in the tth iteration, A t (o,k) is the degree of belonging between data point o and cluster center k in the tth iteration, Cluster stability index: Among them, H(C1,C2) is the matching degree of C1 and C2, WDX(p,λ) is the stability index of the m clustering results obtained by a set of candidate parameters h and γ, and H(C b ,C b′ ) is C b and C b′ The matching degree, C1 and C2 are two different clustering results, C b and C b′ are the b-th and b'-th clustering results obtained by the candidate parameters h and γ, respectively. c1 and c2 are the number of clusters in C1 and C2, respectively. is the first The clusters and the The number of data points that match the clusters. is the Ith cluster in C1 and the Ith cluster in C2 The number of data points that match the clusters. is the first The clusters and the The number of data points that match the clusters is determined by the number of candidate parameters with the largest stability index value. The data points in the clusters with fewer data points are regarded as abnormal disturbance data.

6. The spatial differential disturbance extraction method based on satellite in-situ electron density data according to claim 1 is characterized in that: The method of performing seismic impact analysis on the magnetic quiet differential data based on the seismic data to obtain seismic disturbance data, and extracting ionospheric abnormality data based on the abnormal disturbance data and the seismic disturbance data comprises: The skewness, kurtosis and spatial variation rate of the magnetic quiet difference data of the revisit orbit closest to the epicenter are extracted to form the spatial difference features. The spatial differential features are used as raw data to map to high-dimensional space through nonlinear transformation functions, and a hypersphere with the smallest volume is found in the high-dimensional space. The objective function is: in, is the radius of the hypersphere, n is the number of original data, Φ(·) is the nonlinear transformation function, is the i-th original data, a is the center of the hypersphere, ξ i is the ith relaxation factor, is the penalty parameter, β i is the local density weight, ||·|| represents the Euclidean distance, ρ i is the gradient threshold of the preset nonlinear transformation function, for The average value of , using the Lagrange multiplier method to obtain the dual form of the objective function: Among them, G(·) is the kernel function, α i and α j They are and The corresponding Lagrange coefficient is used to solve the minimum value of the dual form of the objective function, and we get is the jth original data, v i is the gradient constraint The corresponding Lagrange coefficient is used to solve the minimum value of the dual form of the objective function, and the calculation formulas for the center and radius of the hypersphere are obtained as follows: in, SV is the original data set of support vectors, is the vth original data in SV, The distance from the original data to the center of the hypersphere is in, for The distance to the center of the hypersphere, For the The original data, if but On or inside the hypersphere, it is normal data; otherwise The corresponding magnetic quiet differential data belong to seismic disturbance data; The overlapping data of abnormal disturbance data and earthquake disturbance data are ionospheric anomaly data.

7. The spatial differential disturbance extraction method based on satellite in-situ electron density data according to claim 1 is characterized in that: The method for constructing a spatial differential disturbance extraction model according to the spatial differential data and the ionospheric anomaly data comprises: A spatial differential disturbance extraction model is constructed using a neural network algorithm, which includes an input layer, a function layer, and an output layer; The input layer preprocesses the input data and converts it into spatial differential data, and obtains magnetically quiet differential data based on the space weather index; the function layer is a binary classification module that learns the correspondence between magnetically quiet differential data and ionospheric anomaly data; the output layer outputs the extraction results of ionospheric anomaly data.

8. An electronic device comprising: processor; as well as A memory arranged to store computer executable instructions, which when executed cause the processor to perform the method according to any one of claims 1 to 7.

9. A computer-readable storage medium storing one or more programs, which, when executed by an electronic device including a plurality of application programs, enables the electronic device to execute the method according to any one of claims 1 to 7.

Images