A real-time line galloping monitoring method and system

Abstract

The application belongs to the technical field of line monitoring information, and discloses a real-time line galloping monitoring method and system, which comprises the following steps: constructing an LSTM model through VMD decomposition of the time sequence of the spherical harmonic coefficients of each ionospheric model eigenmode, analyzing the influence size of the principal components according to the data of each eigenmode decomposed by VMD, obtaining the predicted value of the spherical harmonic coefficients of the ionospheric model, and calculating the ionospheric delay through the predicted value of the spherical harmonic coefficients of the ionosphere, so that the ionospheric spherical harmonic coefficients are not needed to wait, the real-time performance of the calculation is improved, and the calculation accuracy is improved without calculating through the past ionospheric spherical harmonic coefficients; the change of each eigenmode before and after reconstruction is measured through the sum of the Euclidean distances of each eigenmode at each time, so that the principal component influence size of each eigenmode is determined.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of line monitoring information, and in particular to a real-time line galloping monitoring method and system. BACKGROUND

[0002] In the application of global navigation satellite system (GNSS), the electron content (TEC) of ionosphere is an important issue affecting signal propagation, especially for positioning services requiring high accuracy, such as line galloping power transmission monitoring. The ionosphere is a charged layer of the Earth's atmosphere, located at an altitude of about 50 to 1000 kilometers. The uneven electron density of the ionosphere can cause changes in the speed and path of radio waves passing through the layer, which is called ionospheric delay, and is one of the main factors affecting the accuracy of GNSS signals.

[0003] In order to effectively predict and compensate for the influence of the ionosphere, a spherical harmonic function model is usually used to describe the global distribution of ionospheric electron content. This model can provide an estimate of the ionospheric TEC on a global scale, which can be used to correct GNSS signals propagating through the ionosphere. However, the traditional spherical harmonic function model contains up to 256 coefficients, and the one-time complete transmission of these coefficients is often not feasible under the existing data transmission format, especially in real-time processing applications. The lack of real-time and complete ionospheric data will directly affect the positioning accuracy, which is extremely critical in applications requiring extremely strict positioning and time synchronization. Therefore, there is an urgent need for a method that can effectively transmit key spherical harmonic function coefficients without increasing the transmission burden, and ensure the balance between real-time and accuracy. SUMMARY

[0004] In view of the above existing problems, the present application is proposed.

[0005] Therefore, the present application provides a real-time line galloping monitoring method and system to solve the problem that the coefficients of the spherical harmonic function model cannot be transmitted completely at one time under the existing data transmission method, lack of real-time and complete ionospheric data, affecting the positioning accuracy, and unable to achieve accurate positioning and time synchronization.

[0006] To solve the above technical problems, the present application provides the following technical solutions:

[0007] In a first aspect, the present application provides a real-time line galloping monitoring method, comprising:

[0008] Real-time acquisition of time series data of ionospheric model spherical harmonic vectors, decomposing each component of the data to obtain the intrinsic modal component of each component;

[0009] Constructing an LSTM model for each component of the intrinsic modal component to obtain the predicted value of each intrinsic modal component of each component;

[0010] The values ​​of each intrinsic mode component are obtained by sliding a window to obtain window data, and the corresponding first matrix is ​​constructed and reconstructed for each intrinsic mode component of the window data.

[0011] The rows of the first matrix of each eigenmode after reconstruction are padded with zeros to form the rows of the second matrix. Each eigenmode of the second matrix is ​​reconstructed. The predicted values ​​of the spherical harmonic coefficients of each ionospheric model at a preset time are combined with the predicted values ​​of the components of each eigenmode to calculate the predicted values ​​of the spherical harmonic coefficients of each ionospheric model at a preset time.

[0012] Obtain the missing state of the spherical harmonic coefficients of the ionospheric model received at a preset time. If missing, combine the predicted value at the preset time to calculate the ionospheric delay of the CORS station.

[0013] Based on the approximate coordinates and grid corrections of the line galloping monitoring points, the tropospheric delay, ionospheric delay, and comprehensive error prediction values ​​at the line galloping monitoring points are obtained to generate virtual observation values ​​for the line galloping monitoring points; wherein, the grid corrections include the ionospheric delay of the CORS station;

[0014] The virtual and actual observation values ​​are used to construct a short-baseline double-difference observation equation. The equation is then subjected to real-time Kalman filtering to obtain the time series of coordinates of the line galloping monitoring points.

[0015] As a preferred embodiment of the real-time line galloping monitoring method of the present invention, wherein:

[0016] When the length N of each intrinsic modal component is greater than the first threshold, a sliding window is set, wherein the length L of the sliding window is less than half of the length N of each intrinsic modal component.

[0017] As a preferred embodiment of the real-time line galloping monitoring method of the present invention, wherein:

[0018] For each intrinsic mode component of the window data, construct and reconstruct the corresponding first matrix, including: constructing and reconstructing the first matrix for all window data corresponding to intrinsic mode i. The rows of the matrix form the first matrix F. i (k) ;

[0019] The first matrix F i (k) SVD decomposition yields:

[0020]

[0021] in, and F i (k) The left singular vector and the right singular vector, representing F i (k) singular value matrix of

[0022] singular value matrix of singular value σ i (k) sorting, screening out the smallest a% of singular values σ i (k) and the corresponding left singular vector and right singular vector

[0023] using the remaining singular values σ i (k)* and the corresponding left singular vector and right singular vector V i (k)T* reconstructing the first matrix of each eigenmodal component of the time series of the kth spherical harmonic coefficient, and the first matrix of each component after reconstruction, denoted as:

[0024]

[0025] As a preferred scheme of the real-time line dance monitoring method, wherein:

[0026] The predicted value of each eigenmodal component is combined to calculate the predicted value of each ionospheric model spherical harmonic coefficient at a preset time, including:

[0027] The L1 norm of the difference between each ionospheric model spherical harmonic vector after reconstruction and before reconstruction is calculated;

[0028] The weight of each eigenmodal component is calculated based on the L1 norm;

[0029] The predicted value of each ionospheric model spherical harmonic coefficient at time t+1 is calculated based on the weight and the predicted value of each eigenmodal component.

[0030] As a preferred scheme of the real-time line dance monitoring method, wherein: further comprising obtaining the missing state of the received ionospheric model spherical harmonic coefficient at a preset time, if not missing, using the received ionospheric model spherical harmonic coefficient for ionospheric delay calculation to obtain the ionospheric delay of the CORS station.

[0031] As a preferred scheme of the real-time line dance monitoring method, wherein: combined with the predicted value at the preset time, the ionospheric delay of the CORS station is calculated, specifically including: arranging the received ionospheric model spherical harmonic coefficient at time t+1 to form a 16*16 square matrix to obtain a missing measurement coefficient matrix

[0032] The ionospheric model spherical harmonic coefficients received at the previous t time points without containing missing values The ionospheric model spherical harmonic coefficients received at the j time point Arranging in order to form a 16*16 complete spherical harmonic coefficient matrix

[0033] The complete spherical harmonic coefficient matrix Each element of the complete spherical harmonic coefficient matrix is added one by one to obtain a merged matrix G, and the merged matrix G is subjected to SVD decomposition;

[0034] Each element of the missing measurement coefficient matrix is added one by one to the element in the merged matrix G to obtain an iteration matrix H, and the missing values in the missing measurement coefficient matrix are taken as variables during the addition;

[0035] The initial value of the missing value is set, and the elements in the matrix H are updated at a fixed step;

[0036] The matrix H is subjected to SVD decomposition each time iteration;

[0037] Each element in the singular value matrix of the matrix H after SVD decomposition is subtracted from the corresponding element in the singular value matrix of the merged matrix G and squared, and the iteration is stopped when the result is less than a set threshold value;

[0038] The size of the missing value when the iteration is stopped is taken as the missing value;

[0039] Each missing value obtained by iteration is averaged with the corresponding predicted value to obtain a final missing value;

[0040] The final missing value is completed in the missing measurement coefficient matrix ;

[0041] Each ionospheric model spherical harmonic coefficient in the missing measurement coefficient matrix is used for solving the ionospheric delay of the CORS station.

[0042] As a preferred scheme of the real-time line dance monitoring method, wherein:

[0043] The generated line dance monitoring point virtual observation value comprises:

[0044] The coordinates of the line dance monitoring point and the CORS station and the satellite ephemeris are used to calculate the carrier observation value of the line dance monitoring point and the pseudo-range observation value of the line dance monitoring point, and the observation value of the CORS station is reduced to the line dance monitoring point;

[0045] The virtual observation value of the line dance monitoring point is calculated by combining the relative difference between the troposphere delay prediction value and the ionosphere delay prediction value of the CORS station and the line dance monitoring point.

[0046] In a second aspect, the present application provides a real-time line dance monitoring system, comprising:

[0047] The acquisition decomposition module is configured to acquire time series data of the ionosphere model spherical harmonic vector in real time, decompose each component of the data, and obtain the intrinsic modal component of each component.

[0048] The first construction module is configured to construct an LSTM model for each intrinsic modal component of each component, and obtain the prediction value of each intrinsic modal component of each component.

[0049] The second construction module is configured to obtain window data by taking values of each intrinsic modal component through a sliding window, and construct and reconstruct a corresponding first matrix for each intrinsic modal component of the window data.

[0050] The first calculation module is configured to take rows of the reconstructed first matrix to form rows of a second matrix, reconstruct each intrinsic mode of the second matrix, and combine the weight of each intrinsic modal component to calculate the prediction value of each ionosphere model spherical harmonic coefficient at a preset time.

[0051] The solving module is configured to obtain the missing state of the ionosphere model spherical harmonic coefficient received at a preset time, and if missing, combine the prediction value at the preset time to obtain the ionosphere delay of the CORS station.

[0052] The second calculation module is configured to obtain the troposphere delay, ionosphere delay and comprehensive error prediction value at the line dance monitoring point according to the approximate coordinates of the line dance monitoring point and the grid correction number, to generate the virtual observation value of the line dance monitoring point; the grid correction number includes the ionosphere delay of the CORS station.

[0053] The third calculation module is configured to combine the virtual observation value and the actual observation value to form a short baseline double difference observation equation, and perform real-time Kalman filtering on the equation to obtain the coordinate time series of the line dance monitoring point.

[0054] In a third aspect, the present application provides a computing device, comprising:

[0055] A memory and a processor;

[0056] The memory is configured to store computer executable instructions, and the processor is configured to execute the computer executable instructions, which implement the steps of the real-time line dance monitoring method when executed by the processor.

[0057] In a fourth aspect, the present application provides a computer readable storage medium storing computer executable instructions, which, when executed by a processor, implement the steps of the real-time line galloping monitoring method.

[0058] Compared with the prior art, the present application has the following beneficial effects: the present application constructs an LSTM model by VMD decomposition of the time series of each ionospheric model spherical harmonic coefficient, analyzes the influence size of the principal component according to the data of each intrinsic mode of VMD decomposition, obtains the predicted value of the ionospheric model spherical harmonic coefficient, and calculates the ionospheric delay by the predicted value of the ionospheric spherical harmonic coefficient, which can improve the real-time performance of the calculation without waiting for the ionospheric spherical harmonic coefficient, and improve the calculation accuracy without calculating by the past ionospheric spherical harmonic coefficient; the present application gives higher weight to the intrinsic mode with greater influence of the principal component, otherwise gives lower weight, which reduces the influence of noise and the like on the ionospheric spherical harmonic coefficient, thereby improving the positioning accuracy of the line galloping monitoring; the present application measures the change of each intrinsic mode before and after reconstruction by the sum of the Euclidean distances of each time and each intrinsic mode, thereby determining the principal component influence size of each intrinsic mode; in the present application, the reception and transmission of a large amount of data are completed by the CORS site, and only a small amount of correction is needed at the line galloping monitoring point, which greatly reduces the data reception and transmission amount of a single line galloping monitoring point, and is of great significance for reducing the cost of line galloping monitoring data reception and transmission. BRIEF DESCRIPTION OF DRAWINGS

[0059] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed in the embodiment description. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0060] Figure 1 The overall flow logic diagram of the real-time line galloping monitoring method described in the first embodiment of the present application. DETAILED DESCRIPTION

[0061] In order to make the above-mentioned purposes, features and advantages of the present application more apparent and easy to understand, the specific embodiments of the present application will be described in detail below with reference to the drawings of the specification. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor should belong to the protection scope of the present application.

[0062] Embodiment 1

[0063] Reference Figure 1For an embodiment of the present application, a real-time line dance monitoring method is provided, comprising:

[0064] S100: Real-time acquisition of time series data of ionospheric model spherical harmonic vectors;

[0065] Specifically, the components of the ionospheric model spherical harmonic vector are spherical harmonic coefficients.

[0066] S200: Decompose each component of the ionospheric model spherical harmonic vector time series data to obtain the intrinsic modal component of each component;

[0067] Specifically, each component is decomposed by VMD to obtain the intrinsic modal component of each component of the ionospheric model spherical harmonic vector Where k represents the serial number of the spherical harmonic coefficient, and i represents the serial number of the intrinsic modal component.

[0068] In the embodiments of the present application, when the length N of each intrinsic modal component of each component is greater than a first threshold, a sliding window is set, and the length L of the sliding window is less than half of the length N of each intrinsic modal component.

[0069] S300: Construct an LSTM model for each intrinsic modal component of each component to obtain the predicted value of each intrinsic modal component of each component

[0070] S400: Slide the intrinsic modal component to obtain window data, and construct and reconstruct a corresponding first matrix F for each intrinsic modal component of the window data i (k) ;

[0071] Illustratively, the sliding window can slide from the beginning of the intrinsic modal component to obtain window data each time the sliding window is slid The sliding step is 1;

[0072] In the embodiments of the present application, a corresponding first matrix F is constructed and reconstructed for each intrinsic modal component of the window data i (k) , comprising:

[0073] All window data corresponding to the intrinsic modal i are taken as rows of the matrix to form the first matrix F i (k) ;

[0074] The first matrix F i (k) is subjected to SVD decomposition to obtain:

[0075]

[0076] wherein, and respectively represent the left singular vector and the right singular vector of F i (k) represent the singular value matrix of F i (k)

[0077] sort the singular values σ in the singular value matrix i (k) , filter out the smallest a% of the singular values σ i (k) corresponding to the time series of the kth spherical harmonic coefficient, and the corresponding left singular vector and the right singular vector

[0078] use the remaining singular values σ i (k)* and the corresponding left singular vector and the right singular vector to reconstruct the first matrix of each eigenmodal component of the time series of the kth spherical harmonic coefficient, and the first matrix of each component after reconstruction is represented as:

[0079]

[0080] S500: take rows of the first matrix of each eigenmodal after reconstruction to form the rows of the second matrix, and take the average of each column of the second matrix to reconstruct each eigenmodal, combine the weight of each eigenmodal component, and calculate the predicted value of each ionospheric model spherical harmonic coefficient at a preset time;

[0081] For example, the zero padding step is: sequentially take each row of the first matrix of each eigenmodal after reconstruction from top to bottom, and pad zeros at the beginning and end of all taken rows. The padding rule is: pad m-1 zeros before the first element of the mth row, and pad N-L zeros after the Lth element of the mth row. All rows after padding are sequentially combined to form the rows of the second matrix .

[0082] Take the average of each column of the second matrix to reconstruct each eigenmodal, which is represented as:

[0083]

[0084] In the embodiments of the present application, combining the weight of each eigenmodal component to calculate the predicted value of each ionospheric model spherical harmonic coefficient at a preset time includes: ​​

[0085] The L1 norm of the difference between each ionospheric model spherical harmonic vector after reconstruction and before reconstruction is calculated;

[0086] Specifically, the L1 norm is represented as:

[0087]

[0088] Where t is a time point;

[0089] The weight of each eigenmode component is calculated based on the L1 norm;

[0090] Specifically, the weight of each eigenmode component is calculated as:

[0091]

[0092] The predicted value of each ionospheric model spherical harmonic coefficient at time t+1 is calculated based on the weight and the predicted value of each eigenmode component;

[0093] Specifically, the predicted value y (k) at time t+1 is calculated as:

[0094]

[0095] S600: Obtain the missing state of the ionospheric model spherical harmonic coefficient received at a preset time, and if missing, combine the predicted value at the preset time to obtain the ionospheric delay of the CORS station;

[0096] It should be noted that the preset time is the above-mentioned time t+1.

[0097] In the embodiments of the present application, the ionospheric delay of the CORS station is calculated by combining the predicted value at the preset time, and specifically includes: arranging the ionospheric model spherical harmonic coefficient received at time t+1 to form a 16*16 square matrix to obtain a missing measurement coefficient matrix

[0098] The ionospheric model spherical harmonic coefficient received at the previous t time points without containing missing values is The ionospheric model spherical harmonic coefficient received at time j is Arranged in order to form a 16*16 complete spherical harmonic coefficient matrix

[0099] The complete spherical harmonic coefficient matrix Each element of the complete spherical harmonic coefficient matrix is one-to-one corresponding and accumulated to obtain a merged matrix G, and the merged matrix G is subjected to SVD decomposition;

[0100] Specifically, the merged matrix G is represented as:

[0101]

[0102] The merged matrix G is decomposed by SVD, and is expressed as:

[0103]

[0104] wherein, and respectively represent left singular vectors and right singular vectors of G, represents a singular value matrix of G;

[0105] Each element of the missing measurement coefficient matrix is added to an element in the merged matrix G to obtain an iteration matrix H, and the missing value in the missing measurement coefficient matrix is taken as a variable during the addition;

[0106] An initial value of the missing value is set, and the elements in the matrix H are updated at a fixed step;

[0107] The matrix H is decomposed by SVD each time;

[0108] Each element in the singular value matrix after the SVD decomposition of the matrix H is subtracted from a corresponding element in the singular value matrix of the merged matrix G, and squared, and the iteration is stopped when the result is less than a set threshold value;

[0109] Specifically, the calculation process and the obtained result are expressed as:

[0110]

[0111] The iteration is stopped when Ω is less than the set threshold value.

[0112] The size of the missing value when the iteration is stopped is taken as the missing value;

[0113] Each missing value obtained by the iteration is averaged with a corresponding predicted value to obtain a final missing value;

[0114] The final missing value is completed in the missing measurement coefficient matrix ;

[0115] Each ionospheric model spherical harmonic coefficient in the missing measurement coefficient matrix is used for solving the ionospheric delay of the CORS station.

[0116] In the embodiments of the present application, the missing state of the ionospheric model spherical harmonic coefficient received at a preset time is also obtained, and if there is no missing, the received ionospheric model spherical harmonic coefficient is used for solving the ionospheric delay to obtain the ionospheric delay of the CORS station.

[0117] It should be noted that the steps of S100-S600 are all performed by the CORS station, which can greatly reduce the data transmission amount of the single line dance monitoring point and reduce the cost.

[0118] S700: According to the approximate coordinates of the line dance monitoring point and the grid correction number, the troposphere delay, the ionosphere delay and the comprehensive error prediction value at the line dance monitoring point are obtained to generate the virtual observation value of the line dance monitoring point; wherein, the grid correction number includes the ionosphere delay of the CORS station;

[0119] Specifically, the method for obtaining the troposphere delay prediction value at the line dance monitoring point is:

[0120] The troposphere related error is divided into dry delay part and wet delay part, the dry delay part is corrected by GPT2w or ITG model, the wet delay part is calculated according to the troposphere wet delay result of each CORS station site using the Kriging interpolation algorithm based on height correction, and the troposphere correction number of the line dance monitoring point is calculated, and the troposphere delay is calculated according to the troposphere correction number at the line dance monitoring point;

[0121] The Kriging interpolation algorithm based on height correction specifically includes:

[0122] The elevation of each CORS station site is averaged to obtain the elevation reference surface of the troposphere wet delay correction of the interpolation area;

[0123] According to the troposphere wet delay height difference correction model, the troposphere wet delay correction number from each CORS station to the troposphere error correction reference surface is calculated;

[0124] The spatial structure analysis and variability analysis of the troposphere wet delay of each CORS station site are carried out on the elevation reference surface;

[0125] The CORS station interpolates the troposphere wet delay correction number of the troposphere wet delay correction reference surface at the place according to the approximate coordinates of the line dance monitoring point, according to the spatial structure analysis and variability analysis results of the troposphere wet delay;

[0126] According to the elevation of the approximate coordinates of the line dance monitoring point, the height difference from the line dance monitoring point to the troposphere error correction reference surface is calculated;

[0127] According to the troposphere wet delay height difference correction model, the troposphere wet delay prediction value at the approximate coordinates of the line dance monitoring point is calculated.

[0128] Specifically, the method for obtaining the ionosphere delay prediction value at the line dance monitoring point is:

[0129] Carrying out spatial structural analysis and spatial variability analysis at the central ionospheric height,

[0130] Taking the preset height as an ionospheric error correction reference surface;

[0131] Carrying out non-difference ionospheric delay error interpolation calculation at the ionospheric error correction reference surface to obtain ionospheric correction numbers at the line dance monitoring points, and calculating ionospheric delay amounts according to the ionospheric correction numbers at the line dance monitoring points;

[0132] The non-difference ionospheric delay error interpolation calculation includes:

[0133] Under the condition of linear unbiased and optimal estimation, Kriging equation groups for ionospheric total electron content interpolation are obtained, and the Kriging equation groups are:

[0134]

[0135] In the formula, μ represents a Lagrange multiplier factor, d ij represents the distance between the CORS site and the line dance monitoring point, d i0 represents the distance between the interpolation point and the observation value, λ j represents a weighting coefficient;

[0136] The Kriging equation groups are solved to obtain the weighting coefficient;

[0137] The ionospheric total electron content estimation value is calculated based on the weighting coefficient;

[0138] The ionospheric delay amount prediction value at the line dance monitoring point is calculated through the ionospheric total electron content estimation value.

[0139] Specifically, the method for obtaining the comprehensive error prediction value at the line dance monitoring point can obtain the comprehensive error prediction value through interpolation by the inverse distance weighting method.

[0140] In the embodiments of the present application, generating virtual observation values at the line dance monitoring points includes:

[0141] Using the coordinates of the line dance monitoring points and the CORS sites and satellite ephemeris, the carrier observation values at the line dance monitoring points and the pseudorange observation values at the line dance monitoring points are calculated, and the observation values of the CORS sites are reduced to the line dance monitoring points;

[0142] Specifically, the method for reducing the observation values of the CORS sites to the line dance monitoring points is:

[0143] Φ′ V = Φ A - ρ A + ρ V

[0144] P' V = P A - p A + p V

[0145] wherein, Φ' V represents the carrier observation value of the line dance monitoring point, P' V represents the pseudo-range observation value of the line dance monitoring point, Φ A represents the carrier observation value of the CORS site, P A represents the pseudo-range observation value of the CORS site, p A represents the geometric distance from the CORS site to the satellite, p V represents the geometric distance from the line dance monitoring point to the satellite.

[0146] The virtual observation value of the line dance monitoring point is calculated by combining the relative differences between the tropospheric delay prediction values and the ionospheric delay prediction values at the CORS site and the line dance monitoring point.

[0147] Specifically, the difference between the double-difference tropospheric delay error of the CORS site is calculated as follows: The calculation method is as follows:

[0148]

[0149] The difference between the double-difference ionospheric delay error at the CORS site is calculated as follows: The calculation method is as follows:

[0150]

[0151] wherein, f1 and f2 are the frequencies of the L1 and L2 observation values, and are the L1 and L2 double-difference carrier observation values, and are the L1 and L2 double-difference integer ambiguities, and p represents the geometric distance between the satellite and the receiver.

[0152] Specifically, the difference between the double-difference tropospheric delay error at the line dance monitoring point is calculated as follows: and the difference between the double-difference ionospheric delay error is calculated as follows:

[0153] The observation value of the virtual reference station, i.e., the virtual observation value of the line dance monitoring point, is calculated.

[0154] Specifically, the observation value of the virtual reference station of the virtual reference station includes the virtual reference station carrier observation value and the virtual reference station pseudo-range observation value.

[0155] carrier observation value of the virtual reference station V The calculation method is as follows:

[0156]

[0157] wherein, represents the double-difference error correction number of the carrier observation value of the virtual reference station, including the double-difference error correction numbers of the carrier observation values corresponding to frequency point 1 and frequency point 2 respectively and The calculation method is as follows:

[0158]

[0159] pseudo-range observation value P of the virtual reference station V The calculation method is as follows:

[0160]

[0161] wherein, represents the double-difference error correction number of the pseudo-range of the virtual reference station, including the double-difference error correction numbers of the pseudo-range of the line dance monitoring point corresponding to frequency point 1 and frequency point 2 respectively and The calculation method is as follows:

[0162]

[0163] S800: Assemble the virtual observation value and the actual observation value into a short baseline double-difference observation equation, and perform real-time Kalman filtering on the equation to obtain a coordinate time sequence of the line dance monitoring point.

[0164] The above is a schematic scheme of the real-time line dance monitoring method of the embodiment. It should be noted that the technical scheme of the real-time line dance monitoring system belongs to the same concept as the technical scheme of the real-time line dance monitoring method described above. The technical scheme of the real-time line dance monitoring system in the embodiment is not described in detail, and the description of the technical scheme of the real-time line dance monitoring method can be referred to.

[0165] The real-time line dance monitoring system in the embodiment comprises:

[0166] The acquisition decomposition module is configured to acquire time sequence data of the ionospheric model spherical harmonic vector in real time, decompose each component of the data, and obtain eigenmode components of each component.

[0167] The first construction module is configured to construct an LSTM model for each eigenmode component of each component, and acquire predicted values of each eigenmode component of each component.

[0168] The second construction module is configured to obtain window data by taking values of each intrinsic modal component through a sliding window, and construct and reconstruct a corresponding first matrix for each intrinsic modal component of the window data respectively;

[0169] The first calculation module is configured to take rows of the reconstructed first matrix as rows of a second matrix, reconstruct each intrinsic modal of the second matrix, and combine the weight of each intrinsic modal component to calculate a predicted value of each ionospheric model spherical harmonic coefficient at a preset time;

[0170] The solving module is configured to obtain a missing state of the ionospheric model spherical harmonic coefficient received at the preset time, and if missing, combine the predicted value at the preset time to obtain an ionospheric delay of the CORS station;

[0171] The second calculation module is configured to obtain the tropospheric delay, the ionospheric delay and the comprehensive error prediction value of the line dance monitoring point according to the approximate coordinates of the line dance monitoring point and the grid correction number, so as to generate a virtual observation value of the line dance monitoring point; the grid correction number includes the ionospheric delay of the CORS station.

[0172] The third calculation module is configured to group the virtual observation value and the actual observation value into a short baseline double difference observation equation, and perform real-time Kalman filtering on the equation to obtain a coordinate time sequence of the line dance monitoring point.

[0173] The embodiment also provides a computing device suitable for real-time line dance monitoring, comprising:

[0174] The memory is configured to store computer executable instructions, and the processor is configured to execute the computer executable instructions to implement the method for real-time line dance monitoring.

[0175] The embodiment also provides a storage medium having a computer program stored thereon, and the computer program is executed by a processor to implement the method for real-time line dance monitoring.

[0176] The storage medium provided by the embodiment belongs to the same inventive concept as the method for real-time line dance monitoring, and the technical details not described in the embodiment can be referred to the above embodiments, and the embodiment has the same beneficial effects as the above embodiments.

[0177] Those skilled in the art can clearly understand the present application by the above description of the embodiments, and the present application can be realized by software and necessary general hardware, and of course, can also be realized by hardware. Based on such understanding, the technical solutions of the present application or the part that contributes to the prior art can be embodied in the form of a software product, which can be stored in a computer readable storage medium, such as a floppy disk, a read-only memory (ROM), a random access memory (RAM), a FLASH, a hard disk, or an optical disc, and includes a number of instructions to make a computer device (which can be a personal computer, a server, or a network device, etc.) execute the methods of various embodiments of the present application.

[0178] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present application but not limit the present application, and although the present application is described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present application can be modified or replaced equivalently without departing from the spirit and scope of the present application, and all should be covered in the scope of the claims of the present application.

Claims

1. A real-time line galloping monitoring method, characterized in that, The method comprises the following steps: real-time acquisition of time series data of ionospheric model spherical harmonic vectors, decomposition of each component of the data to obtain eigenmode components of each component; constructing an LSTM model for each component of the eigenmode component to obtain the predicted value of each eigenmode component of each component; obtaining window data by taking values of each eigenmode component through a sliding window, and constructing and reconstructing a corresponding first matrix for each eigenmode component of the window data; taking a row of zeros for each eigenmode of the reconstructed first matrix to form a row of a second matrix, reconstructing each eigenmode of the second matrix, and combining the predicted value of each eigenmode component to calculate the predicted value of each ionospheric model spherical harmonic coefficient at a preset time; acquiring the missing state of the received ionospheric model spherical harmonic coefficient at the preset time, and if missing, combining the predicted value at the preset time to obtain the ionospheric delay of the CORS station; obtaining the tropospheric delay, ionospheric delay and comprehensive error prediction value of the line dance monitoring point to generate virtual observation values of the line dance monitoring point according to the approximate coordinates and grid correction of the line dance monitoring point; wherein the grid correction includes the ionospheric delay of the CORS station; constructing a short baseline double difference observation equation for the virtual observation values and the actual observation values, and performing real-time Kalman filtering on the equation to obtain the coordinate time series of the line dance monitoring point.

2. The real-time line dance monitoring method of claim 1, wherein, When the length N of each eigenmode component of each component is greater than a first threshold, a sliding window is set, and the length L of the sliding window is less than half of the length N of each eigenmode component.

3. The real-time line dance monitoring method of claim 2, wherein, Respectively constructing and reconstructing a corresponding first matrix for each intrinsic modal component of the window data, comprising: constructing a first matrix F as rows of the matrix i (k) ; The first matrix F i (k) SVD decomposition is performed to obtain: in, and F i (k) The left singular vector and the right singular vector, F represents i (k) The singular value matrix; singular values matrix singular values σ i (k) sorting, discarding a% of the smallest singular values σ of all singular values corresponding to the time series of the kth spherical harmonic coefficient i (k) and the corresponding left singular vector and the right singular vector using the remaining singular values σ i (k)* and the corresponding left singular vectors and right singular vectors reconstructing the first matrix of each eigenmode component of the time series of the kth spherical harmonic coefficient, the reconstructed first matrix of each component is denoted as:

4. The real-time line dance monitoring method of claim 3, wherein combining the predicted value of each eigenmode component to calculate the predicted value of each ionospheric model spherical harmonic coefficient at a preset time comprises: calculating the L1 norm of the difference between each ionospheric model spherical harmonic vector after reconstruction and before reconstruction; calculating the weight of each eigenmode component based on the L1 norm; calculating the predicted value of each ionospheric model spherical harmonic coefficient at time t+1 based on the weight and the predicted value of each eigenmode component.

5. The real-time line dance monitoring method of claim 4, wherein, Further comprising: acquiring the missing state of the received ionospheric model spherical harmonic coefficient at the preset time, and if not missing, using the received ionospheric model spherical harmonic coefficient for the calculation of the ionospheric delay to obtain the ionospheric delay of the CORS station.

6. The real-time line dance monitoring method of claim 4, wherein The ionospheric delay of the CORS station is calculated according to the predicted value of the preset moment, specifically including: arranging the spherical harmonic coefficients of the ionospheric model received at the moment t+1 to form a 16*16 square matrix to obtain a missing measurement coefficient matrix the ionospheric model spherical harmonic coefficients received at time j in the previous t time instants, not containing missing values arranged in order to form a 16*16 complete spherical harmonic coefficient matrix Each element of the complete spherical harmonic coefficient matrix is one-to-one accumulated to obtain a merging matrix G, and the merging matrix G is subjected to SVD decomposition; Each element of the missing measurement coefficient matrix is added one-to-one corresponding to the element in the merging matrix G to obtain the iteration matrix H, and the missing value in the missing measurement coefficient matrix is taken as a variable when added. setting an initial value for the missing value, updating the elements in the matrix H with a fixed step size; SVD decomposition of the matrix H is performed every iteration; and subtract each element in the singular value matrix of the SVD decomposed matrix H from the corresponding element in the singular value matrix of the merging matrix G and square the result to obtain the final missing value when the result is less than a set threshold; the size of the missing value at the time of stopping iteration is taken as the missing value; the final missing value is obtained by averaging each missing value calculated by iteration and the corresponding predicted value; Completing final missing values to a missing measurement coefficient matrix Institute; Each ionospheric model spherical harmonic coefficient in the missing measurement coefficient matrix is used for the determination of the ionospheric delay at the CORS station.

7. The real-time line dance monitoring method of claim 5 or 6, wherein the generation of virtual observation values of the line dance monitoring point comprises: The carrier observation value of the line dance monitoring point and the pseudo-range observation value of the line dance monitoring point are calculated by using the coordinates and satellite ephemeris of the line dance monitoring point and the CORS station, and the observation value of the CORS station is reduced to the line dance monitoring point; The virtual observation value of the line dance monitoring point is calculated by combining the relative difference between the ionospheric delay prediction value and the tropospheric delay prediction value of the CORS station and the line dance monitoring point.

8. A system for applying the real-time line galloping monitoring method according to claim 1, characterized in that, It comprises: An acquisition decomposition module is configured to acquire time series data of ionospheric model spherical harmonic vectors in real time, decompose each component of the data, and obtain intrinsic modal components of each component; A first construction module is configured to construct an LSTM model for each intrinsic modal component of each component, and obtain prediction values of each intrinsic modal component of each component; A second construction module is configured to take values of each intrinsic modal component through a sliding window to obtain window data, and construct and reconstruct a corresponding first matrix for each intrinsic modal component of the window data; A first calculation module is configured to take rows of the reconstructed first matrix to form rows of a second matrix, reconstruct each intrinsic modal of the second matrix, combine weights of the intrinsic modal components, and calculate prediction values of each ionospheric model spherical harmonic coefficient at a preset time; A solving module is configured to acquire missing states of ionospheric model spherical harmonic coefficients received at a preset time, and if the missing states exist, combine the prediction values at the preset time to obtain ionospheric delay of the CORS station; A second calculation module is configured to acquire tropospheric delay, ionospheric delay, and comprehensive error prediction values at the line dance monitoring point based on approximate coordinates of the line dance monitoring point and grid correction numbers, and generate virtual observation values of the line dance monitoring point; The grid correction numbers comprise the ionospheric delay of the CORS station; A third calculation module is configured to group short baseline double difference observation equations for the virtual observation values and actual observation values, perform real-time Kalman filtering on the equations, and obtain time series of coordinates of the line dance monitoring point. 9.An electronic device, comprising: a memory and a processor; The memory is configured to store computer executable instructions, and the processor is configured to execute the computer executable instructions, which realize the steps of the real-time line dance monitoring method in any one of claims 1 to 7. 10.A computer readable storage medium storing computer executable instructions, which realize the steps of the real-time line dance monitoring method in any one of claims 1 to 7 when executed by a processor.

Images