Satellite orbit kilometer post parameter prediction method, device, equipment and medium

Abstract

The invention discloses a satellite orbit kilometer post parameter prediction method and device, equipment and a medium, and relates to the technical field of satellite data processing. The method comprises the following steps: converting longitude and latitude in satellite positioning data into UTM coordinates; vector projection is carried out on the UTM coordinates on the line segments, and kilometer posts of all key position points are determined; performing multi-time-step lagging feature splicing and differential processing on the satellite positioning data and the kilometer post respectively, generating final feature matrixes of the satellite positioning data and the kilometer post at each key position point, and performing robust standardization; training the double-layer long-short-term memory network to obtain a kilometer post prediction model; and inputting the satellite feature sequence of the to-be-measured position point on the railway track section into the kilometer post prediction model, and outputting the kilometer post of the to-be-measured position point on the railway track section. According to the method, the accuracy of kilometer post parameter prediction can be effectively improved.

Description

TECHNICAL FIELD

[0001] The application relates to the technical field of satellite data processing, in particular to a satellite orbit kilometer marker parameter prediction method and device, equipment and medium. BACKGROUND

[0002] With the rapid progress of satellite navigation and space situation awareness technology, high-precision satellite orbit prediction has become a core technology to ensure the safety of spacecraft in orbit and improve the efficiency of space missions.

[0003] Traditional satellite orbit parameter prediction methods (such as autoregressive integrated moving average and linear regression) are difficult to capture nonlinear time-dependent relationships, satellite data contains high-frequency noise and outliers, resulting in poor robustness of the prediction model, and single time step features cannot fully represent the dynamic characteristics of orbit changes. Conventional data standardization methods (such as Z-score) are sensitive to data distribution and are easily affected by extreme values. Therefore, there is an urgent need to break through the accuracy bottleneck of existing satellite orbit parameter prediction. SUMMARY

[0004] The purpose of the present application is to provide a satellite orbit kilometer marker parameter prediction method, device, equipment and medium, which can effectively improve the accuracy of kilometer marker parameter prediction.

[0005] To achieve the above purpose, the present application provides the following solutions.

[0006] In a first aspect, the present application provides a satellite orbit kilometer marker parameter prediction method, comprising: acquiring satellite positioning data of a plurality of key position points on a railway track section measured by satellite positioning, and converting the longitude and latitude in the data into UTM coordinates; according to the line segments between the position points with known kilometer markers on the railway track section, performing vector projection on the UTM coordinates on the line segments to determine the kilometer markers of each key position point; concatenating the satellite positioning data and the kilometer markers of each key position point with the lag features of multiple time steps respectively to obtain the extended feature matrices of the satellite positioning data and the kilometer markers at each key position point; determining the difference features and moving average features of each extended feature matrix, and generating the final feature matrix from each extended feature matrix and its difference features and moving average features; performing robust standardization on each final feature matrix to obtain the standard feature matrix of the satellite positioning data and the kilometer markers at each key position point; taking the satellite feature sequence of each time step in the standard feature matrix of the satellite positioning data at each key position point as input, and taking the kilometer marker of each time step in the standard feature matrix of the kilometer markers at each key position point as label, training a double-layer long short-term memory network to obtain a kilometer marker prediction model; inputting the satellite feature sequence of the position point to be measured on the railway track section into the kilometer marker prediction model to output the kilometer marker of the position point to be measured on the railway track section.

[0007] Optionally, the longitude and latitude in the data are converted into UTM coordinates, specifically including: determining the UTM partition number using the longitude in the satellite positioning data; determining the radius of curvature according to the ellipsoid parameters and the latitude in the satellite positioning data using the formula and ; in the formula, the radius of curvature is R, the long semi-axis of the ellipsoid is a, the eccentricity is e, the latitude is φ, and the ellipsoid parameter is e2; determining the UTM east coordinate according to the radius of curvature using the formula ; in the formula, the UTM east coordinate is X, the east pseudo-offset is Xo, the scale factor is k, the longitude difference term is Δλ, , the longitude is λ, the longitude of the central meridian is λ0, the latitude tangent square term is tan2(φ), , the correction term related to the eccentricity and the latitude is e sin(φ), ; determining the UTM north coordinate according to the radius of curvature using the formula ; in the formula, the UTM north coordinate is Y, the ellipsoid distance from the equator to the latitude is N, ; the UTM partition number, the UTM east coordinate and the UTM north coordinate together constitute the UTM coordinates.

[0008] Optionally, according to the line segments between the position points with known kilometer markers on the railway track section, the UTM coordinates are projected on the line segments by vector projection to determine the kilometer markers of the key position points, specifically including: determining the line segments between the position points with known kilometer markers on the railway track section; determining the projection parameters of the UTM coordinates of the key position points on each line segment according to the UTM coordinates of the key position points and the end points of each line segment using the formula ; in the formula, the projection parameter is P, , the distance from the UTM coordinates of the key position point to one end point of the line segment is d, , the UTM coordinates of the key position point are (X, Y), one end point coordinate of the line segment is (X0, Y0), The distance from the UTM coordinates of the critical location point to the other endpoint of the line segment. , Let the coordinates be the coordinates of the other endpoint of the line segment; Based on the projection parameters, using the formula Determine the projection coordinates of the UTM coordinates of each key location point onto each line segment; where, The coordinates of the projection point; Based on the UTM coordinates and projection point coordinates of each key location point, select the projection point coordinates and line segment corresponding to the minimum projection distance, and use the selected line segment as the projection line segment. If the coordinates of the selected projection point lie on the projection line segment, then use the formula Determine the kilometer markers for each key location; where, For key location kilometer markers, The kilometer marker at one endpoint of the projected line segment. The kilometer marker at the other endpoint of the projected line segment; If the selected projection point coordinates are outside the projection line segment, then the kilometer marker of the endpoint on the projection line segment closest to the projection point coordinates will be determined as the kilometer marker of each key location point.

[0009] Optionally, the satellite positioning data of each key location point are concatenated with lag features from multiple time steps to obtain an extended feature matrix of the satellite positioning data at each key location point, specifically including: The satellite positioning data with multiple time steps of lag are used to construct a lag feature vector; By concatenating the satellite positioning data of each key location point with the lag feature vector, an extended feature matrix of the satellite positioning data at each key location point is obtained.

[0010] Optionally, the difference features and moving average features of each extended feature matrix are determined, and the final feature matrix is ​​generated from each extended feature matrix and its difference features and moving average features, specifically including: Using formula Calculate the difference features of each extended feature matrix; where, For time step Difference features of the extended feature matrix, For time step The extended feature matrix, For time step The extended feature matrix; Using formula Calculate the moving average characteristic of each extended characteristic matrix; where, For time step Moving average feature of extended feature matrix, For time step The extended feature matrix, This is the offset within the window. =0,1,2; The final feature matrix is ​​obtained by merging the extended feature matrices with the corresponding difference features and moving average features.

[0011] Optionally, robust normalization is performed on each final feature matrix to obtain standard feature matrices for both satellite positioning data and kilometer markers at each key location point, specifically including: Determine the median and interquartile range of each final feature matrix; Based on the median and the interquartile range, using the formula The final feature matrices are standardized to obtain the standard feature matrices of satellite positioning data and kilometer markers at each key location point; where, For the final characteristic matrix, the th The standardized feature values ​​of each feature For the th in the final feature matrix One characteristic, For the final characteristic matrix, the th The median of the features, For the final characteristic matrix, the th Interquartile range of each feature.

[0012] Optionally, the two-layer long short-term memory network includes: a first long short-term memory network, a first dropout layer, a second long short-term memory network, a second dropout layer, and a fully connected layer connected in sequence; The dual-layer long short-term memory network has an adaptive learning rate.

[0013] Secondly, this application provides a satellite orbit kilometer marker parameter prediction device, including: a conversion module, a vector projection module, a feature stitching module, a feature merging module, a standard module, a training module, and an application module.

[0014] The conversion module is configured to acquire satellite positioning data of a plurality of key position points measured by satellite positioning on a railway track section, and convert longitude and latitude in the data into UTM coordinates; the vector projection module is configured to perform vector projection on the UTM coordinates on a line segment between position points with known kilometer marks on the railway track section, to determine kilometer marks of the key position points; the feature splicing module is configured to splice lag features of multiple time steps for the satellite positioning data and the kilometer marks of the key position points respectively, to obtain extended feature matrices of the satellite positioning data and the kilometer marks at the key position points; the feature merging module is configured to determine difference features and moving average features of the extended feature matrices, and generate final feature matrices from the extended feature matrices and the difference features and the moving average features thereof; the standard module is configured to perform robust standardization on the final feature matrices, to obtain standard feature matrices of the satellite positioning data and the kilometer marks at the key position points; the training module is configured to train a double-layer long short-term memory network by taking satellite features in the standard feature matrices of the satellite positioning data at the key position points at each time step as input and taking kilometer marks in the standard feature matrices of the kilometer marks at the key position points at each time step as labels, to obtain a kilometer mark prediction model; and the application module is configured to input satellite feature sequences of a to-be-measured position point on the railway track section into the kilometer mark prediction model, and output a kilometer mark of the to-be-measured position point on the railway track section.

[0015] In a third aspect, the present application provides a computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor executes the computer program to implement the satellite orbit kilometer mark parameter prediction method in any one of the above.

[0016] In a fourth aspect, the present application provides a computer readable storage medium having a computer program stored thereon, and the computer program is executed by a processor to implement the satellite orbit kilometer mark parameter prediction method in any one of the above.

[0017] According to the specific embodiments provided in the present application, the present application has the following technical effects.

[0018] The application provides a satellite orbit kilometer marker parameter prediction method, device, equipment and medium, a double-layer long short-term memory network is used to predict the kilometer marker, the double-layer long short-term memory network can extract hierarchical time sequence features and effectively capture nonlinear time dependence; the double-layer long short-term memory network can indirectly learn an internal representation that is not sensitive to high-frequency noise and outliers in the training stage, thereby improving the robustness of prediction; the satellite positioning data and the kilometer marker at each key position point are respectively spliced with multi-time step lag features, and the defect that a single time step feature cannot fully represent the dynamic characteristics of orbit changes is overcome; the final feature matrix of the satellite positioning data and the kilometer marker at each key position point is subjected to robust standardization, and the problem that a conventional data standardization method is sensitive to data distribution and is easily affected by extreme values is solved. Through the above technical means, the accuracy of kilometer marker parameter prediction is effectively improved. BRIEF DESCRIPTION OF DRAWINGS

[0019] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the related art, the drawings needed in the embodiments will be briefly introduced as follows. Obviously, the drawings in the following description only constitute some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor on the basis of these drawings.

[0020] Figure 1 A flowchart of a satellite orbit kilometer marker parameter prediction method provided by an embodiment of the present application.

[0021] Figure 2 A kilometer marker determination implementation module diagram provided by an embodiment of the present application.

[0022] Figure 3 A UTM conversion and kilometer marker projection diagram provided by an embodiment of the present application.

[0023] Figure 4 An implementation module diagram of a kilometer marker prediction model provided by an embodiment of the present application.

[0024] Figure 5 A comparison diagram of predicted values and actual values of a kilometer marker prediction model provided by an embodiment of the present application.

[0025] Figure 6 A diagram provided by an embodiment of the present application.

[0026] Figure 7 A correlation diagram of predicted values and actual values of a kilometer marker prediction model provided by an embodiment of the present application.

[0027] Figure 8 A function module diagram of a satellite orbit kilometer marker parameter prediction device provided by an embodiment of the present application.

[0028] Figure 9 Fig. 1 is a structural schematic diagram of a computer device according to an embodiment of the present application. DETAILED DESCRIPTION

[0029] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the scope of protection of the present application.

[0030] The above purposes, features and advantages of the present application will be more apparent and understandable. The present application will be further described in detail below with reference to the drawings and specific embodiments.

[0031] In an exemplary embodiment, as shown in Figure 1 A satellite orbit kilometer marker parameter prediction method is provided, including the following steps 101 to 107.

[0032] Step 101: Obtain satellite positioning data of a plurality of key position points on a railway track section measured by satellite positioning, and convert the longitude and latitude in the data into UTM coordinates.

[0033] Step 102: According to the line segment between the position points with known kilometer markers on the railway track section, perform vector projection on the line segment based on the UTM coordinates to determine the kilometer markers of the key position points.

[0034] Step 103: Concatenate the satellite positioning data and the kilometer markers of each key position point with the lag features of multiple time steps, respectively, to obtain the extended feature matrix of the satellite positioning data and the kilometer markers at each key position point.

[0035] Step 104: Determine the difference features and moving average features of each extended feature matrix, and generate a final feature matrix from each extended feature matrix and its difference features and moving average features.

[0036] Step 105: Perform robust standardization on each final feature matrix to obtain a standard feature matrix of the satellite positioning data and the kilometer markers at each key position point.

[0037] Step 106: Take the satellite feature sequence of each time step in the standard feature matrix of the satellite positioning data at each key position point as input, and take the kilometer marker of each time step in the standard feature matrix of the kilometer marker at each key position point as label, train a double-layer long short-term memory network to obtain a kilometer marker prediction model.

[0038] Step 107: inputting the satellite feature sequence of the to-be-tested position point on the railway track section into the kilometer marker prediction model, and outputting the kilometer marker of the to-be-tested position point on the railway track section.

[0039] The implementation of steps 101 to 107 described above realizes satellite orbit kilometer marker value prediction, and effectively improves the accuracy of prediction parameters in the process of predicting the orbit kilometer marker.

[0040] In another exemplary embodiment of the present application, the satellite positioning data is first preprocessed: the satellite positioning data is arranged in the order of longitude and latitude coordinates to realize the time sequence. Then, the UTM (Universal Transverse Mercator) algorithm is called to convert the longitude and latitude into UTM coordinates by using the WGS84 ellipsoid parameters, and the format of the UTM coordinates is: zone number + east coordinate + north coordinate. The conversion of the longitude and latitude in the satellite positioning data into the UTM coordinates in the above step 101 can be replaced by the following steps 201 to 205.

[0041] Step 201: determine the UTM zone number by using the longitude in the satellite positioning data.

[0042] Step 202: according to the ellipsoid parameters and the latitude in the satellite positioning data, determine the curvature radius by using the formula and ; in the formula, is the curvature radius, is the ellipsoid long semi-axis, is the eccentricity, is the latitude (radian), is the ellipsoid parameter. In the UTM conversion, it is used to correct the curvature in the latitude direction. and are both ellipsoid parameters.

[0043] Step 203: according to the curvature radius, determine the UTM east coordinate by using the formula ; in the formula, is the UTM east coordinate, is the east pseudo-offset, is the scale factor, is the longitude difference term, , is the longitude (radian), is the longitude of the central meridian, is the latitude tangent square term, , is the correction term related to the eccentricity and the latitude, . is determined by the UTM zone number.

[0044] Step 204: According to the radius of curvature, determine the UTM North coordinate by the formula , where is the UTM North coordinate, is the ellipsoidal distance from the equator to the latitude, . In the UTM North coordinate calculation, is used to determine the north-south position of the point relative to the equator.

[0045] UTM North coordinate The formula is expanded in even terms, and all expansions are calculated using WGS84 ellipsoid parameters. The calculation formula of the ellipsoidal distance from the equator to the latitude is the standard meridian arc length formula, usually written in the form of a Fourier series.

[0046] Step 205: Form the UTM coordinate by combining the UTM zone number, UTM East coordinate, and UTM North coordinate.

[0047] In another exemplary embodiment of the present application, the nearest point is determined by vector projection, the projection parameter is calculated using vector dot product, and the range is limited within the line segment; based on the linear interpolation of the kilometer marker of the start and end points of the line segment, the kilometer marker corresponding to the nearest projection point or end point in all line segments is selected. Then, the above step 102 can be replaced by the following steps 301~306.

[0048] Step 301: Determine the line segments between all kilometer marker known position points on the railway track section.

[0049] Step 302: According to the UTM coordinates of each key position point and the end points of each line segment, determine the projection parameter of the UTM coordinates of each key position point on each line segment by the formula , where is the projection parameter, , is the distance from the UTM coordinates of the key position point to one end point of the line segment, , is the UTM coordinates of the key position point, is the coordinates of one end point of the line segment, is the distance from the UTM coordinates of the key position point to the other end point of the line segment, , is the coordinates of the other end point of the line segment.

[0050] Step 303: According to the projection parameter, determine the projection point coordinates of the UTM coordinates of each key position point on each line segment by the formula , where is the projection point coordinates.

[0051] Step 304: Based on the UTM coordinates and projection point coordinates of each key location point, select the projection point coordinates and line segment corresponding to the minimum projection distance, and use the selected line segment as the projection line segment.

[0052] Step 305: If the coordinates of the selected projection point lie on the projection line segment, then use the formula... Determine the kilometer markers for each key location; where, For key location kilometer markers, The kilometer marker at one endpoint of the projected line segment. The kilometer marker is the other endpoint of the projected line segment.

[0053] Step 306: If the selected projection point coordinates are outside the projection line segment, then the kilometer marker of the endpoint on the projection line segment closest to the projection point coordinates is determined as the kilometer marker of each key location point.

[0054] The kilometer markers at each key location can be represented as: In the formula, These are kilometer markers for key locations.

[0055] For example, such as Figure 2 As shown, steps 101 to 102 are implemented by an input module, a UTM conversion module, a cumulative distance calculation module, a projection calculation module, and an output module. The input module receives key points of the route (longitude, latitude, and kilometer markers) and the set of points to be converted; the UTM conversion module calls the UTM algorithm to convert latitude and longitude into planar coordinates; the cumulative distance calculation module generates the mileage information of the path by accumulating line segments; the projection calculation module projects the points to be measured onto the nearest line segment and calculates the kilometer markers within the effective interval; the output module outputs the kilometer marker results in a formatted manner (e.g., K071+500).

[0056] Figure 3 The UTM conversion and kilometer marker projection provided in this application embodiment visualizes the results of converting satellite latitude and longitude coordinates into orbital kilometer markers. Figure 3 Part (a) shows the spatial projection relationship of geographic coordinates. Figure 3 Part (b) shows the UTM coordinate space projection relationship, in Figure 3 In part (b), the projection points cover the points to be measured.

[0057] In another exemplary embodiment of this application, if all satellite positioning data and kilometer markers are set with lag features, then step 103 above, which splices the satellite positioning data of each key location point with lag features of multiple time steps to obtain the extended feature matrix of satellite positioning data at each key location point, can be replaced by steps 401 to 402.

[0058] Step 401: Construct a lag feature vector with the satellite positioning data of the lagged multiple time steps of the satellite positioning data.

[0059] For example, the lag feature vector corresponds to a matrix which can be expressed as: ; In the formula, is the total number of samples; is the maximum lag step number; is the satellite positioning data at time In this application, there are 3 feature values (longitude, latitude, and altitude), so each contains 3 values. .

[0060] Step 402: Concatenate the satellite positioning data at each key location point with the lag feature vector to obtain an extended feature matrix of the satellite positioning data at each key location point.

[0061] For example, if the lag is 3 time steps, the lag feature vector is: ; wherein, is a matrix composed of the satellite positioning data (original features) of the previous 3 time steps , with a dimension of 3x3; , , are the original feature vector values of the previous 3 time steps.

[0062] The extended feature matrix of the satellite positioning data at each key location point is expressed as: ; wherein, is the merged extended feature vector, with a dimension of (original 3 dimensions + lag 9 dimensions). is the original feature vector value, with a dimension of . is a matrix flattening operation that converts a 3x3 matrix into a 9x1 vector.

[0063] The steps of concatenating the kilometer marker with the lag feature of multiple time steps to obtain the extended feature matrix of the kilometer marker at each key location point are the same as steps 401-402.

[0064] In another exemplary embodiment of the application, the short-term change trend is captured by the difference operation within the time window, and the moving average fusion is performed by using the sliding window mean smoothing to eliminate high-frequency noise. The above step 104 of generating the final feature matrix can be replaced by the following steps 501-503.

[0065] Step 501: Using the formula Calculate the difference features of each extended feature matrix; where, For time steps Difference features of the extended feature matrix, For time steps The extended feature matrix, For time steps The extended feature matrix.

[0066] Step 502: Using the formula Calculate the moving average characteristic of each extended characteristic matrix; where, For time steps Moving average feature of extended feature matrix, For time steps The extended feature matrix, The offset within the window. =0,1,2.

[0067] Step 503: Merge each extended feature matrix with the corresponding difference feature and moving average feature to obtain the final feature matrix.

[0068] Difference features refer to the difference between the current value and the historical value, and are used to capture short-term trends. The window size is 3, used to smooth noise and reflect long-term trends.

[0069] For example, based on the above lag of 3 time steps... The final feature matrix can be expressed as: ; in, The final feature vector after merging has a dimension of . (Extended features 12-dimensional + difference 12-dimensional + moving average 12-dimensional).

[0070] In another exemplary embodiment of this application, the above step 105, which robustly normalizes each final feature matrix to obtain the standard feature matrix of each satellite positioning data and kilometer marker at each key location point, can be replaced by the following steps 601 to 602.

[0071] Step 601: Determine the median and interquartile range of each final feature matrix.

[0072] Step 602: Based on the median and the interquartile range, use the formula... The final feature matrices are standardized to obtain the standard feature matrices of satellite positioning data and kilometer markers at each key location point; where, For the final characteristic matrix, the th The standardized feature values ​​of each feature For the th in the final feature matrix One characteristic, For the final characteristic matrix, the th The median of each feature. For the final characteristic matrix, the th Interquartile range (IQR) of each feature. , , For the final characteristic matrix, the th Data for each feature.

[0073] In another exemplary embodiment of this application, the two-layer long short-term memory network includes: a first long short-term memory network, a first dropout layer, a second long short-term memory network, a second dropout layer, and a fully connected layer connected sequentially. The first long short-term memory network outputs the complete sequence, and then connects to the first dropout layer to prevent overfitting, with a ratio of 0.4; the second long short-term memory network outputs the last time step, and then connects to the second dropout layer, with a ratio of 0.3; finally, the fully connected layer progressively reduces the dimensionality to 1 dimension for the output.

[0074] The improvements of the two-layer long short-term memory network compared to the LSTM (Long Short-Term Memory) network include: the use of a two-layer cascaded LSTM structure, dynamic regularization, adaptive learning rate scheduling, and prediction error compensation.

[0075] The dual-layer long short-term memory network adopts a dual-layer cascaded LSTM structure: the first layer (first long short-term memory network) has 256 units that output the complete sequence, and the second layer (second long short-term memory network) has 128 units that output the final state.

[0076] Dynamic regularization: A first Dropout layer with a probability of 0.4 is inserted between the first and second Long Short-Term Memory networks, and a second Dropout layer with a probability of 0.3 is inserted between the second Long Short-Term Memory network and the fully connected layer.

[0077] Adaptive learning rate scheduling: Initial learning rate 0.0005, decreasing by 30% every 200 rounds.

[0078] Prediction error compensation: Based on the error distribution characteristics during the testing period, an error threshold alarm and dynamic correction mechanism is established.

[0079] The two-layer Long Short-Term Memory network was trained using the Adam optimizer with a learning rate of 0.0005 and a decay of 30% every 200 epochs. Overfitting was monitored on the validation set. ; wherein, is a training target function; is a true label of the i-th sample; is a model prediction value of the i-th sample; is a true label of the i-th sample; is a model prediction value of the i-th sample; is a total number of training samples.

[0080] MAE, RMSE, indicators are calculated, and a prediction comparison curve, an error histogram and an actual-prediction scatter plot are generated: ; ; ; wherein, MAE is the mean absolute error, which measures the average deviation of the prediction value from the true value; RMSE is the root mean square error, which is more sensitive to larger errors; is the determination coefficient, which represents the explanatory ability of the model to data variation; is the mean value of the label; is a total number of test samples.

[0081] In another exemplary embodiment of the present application, with reference to Figure 4 , steps 101 to 106 of the method of the present application can be performed by a data preprocessing module, a feature engineering module, a standardization module, an LSTM network module and an evaluation module. The data preprocessing module reads data and converts the kilometer marker string into a numerical value; the feature engineering module constructs lag features, difference features and moving average features; the standardization module processes data using robust standardization (median and interquartile range); the LSTM network module constructs a double-layer LSTM structure and adds Dropout to prevent overfitting; and the evaluation module calculates MAE, RMSE, indicators, and generates an error distribution graph and a prediction comparison graph.

[0082] Figure 5 a comparison between the prediction value and the actual value of the kilometer marker prediction model, Figure 6 a prediction error time series of the kilometer marker prediction model, Figure 7 a correlation between the prediction value and the actual value of the kilometer marker prediction model. It can be seen from Figures 5-7 that the prediction error of the method of the present application is small, the prediction correlation is highly consistent, the prediction value and the actual value have small errors, and the prediction requirements are met.

[0083] The present application utilizes LSTM technology to achieve satellite orbit kilometer marker value prediction, and effectively improves the accuracy of prediction parameters in the process of predicting the orbit kilometer marker.

[0084] Based on the same inventive concept, the application further provides a satellite track kilometer parameter prediction device for implementing the satellite track kilometer parameter prediction method described above. The implementation scheme of the device for solving the problem is similar to the implementation scheme described in the above method, so the specific limitations in one or more satellite track kilometer parameter prediction device embodiments provided below can refer to the limitations of the satellite track kilometer parameter prediction method described above, and will not be repeated here.

[0085] In an exemplary embodiment, as shown in Figure 8 A satellite track kilometer parameter prediction device is provided, including a conversion module, a vector projection module, a feature splicing module, a feature merging module, a standard module, a training module, and an application module.

[0086] The conversion module is configured to obtain satellite positioning data of a plurality of key position points on a railway track section measured by satellite positioning, and convert the longitude and latitude in the data into UTM coordinates; the vector projection module is configured to perform vector projection on the UTM coordinates on a line segment between position points with known kilometer marks on the railway track section, to determine the kilometer marks of each key position point; the feature splicing module is configured to splice the satellite positioning data and the kilometer marks of each key position point with multi-time step lag features respectively, to obtain extended feature matrices of the satellite positioning data and the kilometer marks at each key position point; the feature merging module is configured to determine difference features and moving average features of each extended feature matrix, and generate a final feature matrix from each extended feature matrix and its difference features and moving average features; the standard module is configured to perform robust standardization on each final feature matrix, to obtain standard feature matrices of the satellite positioning data and the kilometer marks at each key position point; the training module is configured to take satellite feature sequences of each time step in the standard feature matrices of the satellite positioning data at each key position point as input, and take kilometer marks of each time step in the standard feature matrices of the kilometer marks at each key position point as labels, to train a double-layer long short-term memory network, to obtain a kilometer mark prediction model; and the application module is configured to input satellite feature sequences of a to-be-measured position point on the railway track section into the kilometer mark prediction model, to output the kilometer mark of the to-be-measured position point on the railway track section.

[0087] In an exemplary embodiment, a computer device is provided, which can be a server or a terminal, and its internal structure diagram can be as shown in Figure 9As shown in the figure. The computer device includes a processor, a memory, an Input / Output (I / O) interface, and a communication interface. Among them, the processor, the memory and the input / output interface are connected through the system bus, and the communication interface is connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store the kilometer mark of the to-be-measured position point on the railway track section. The input / output interface of the computer device is used to exchange information between the processor and external devices. The communication interface of the computer device is used to communicate with external terminals through network connection. The computer program is executed by the processor to implement a satellite orbit kilometer mark parameter prediction method.

[0088] Those skilled in the art can understand that, Figure 9 The structure shown in the figure is only a block diagram of part of the structure related to the scheme of the present application, and does not constitute a limitation on the computer device to which the scheme of the present application is applied. The specific computer device can include more or fewer components than those shown in the figure, or combine certain components, or have a different component arrangement. In an exemplary embodiment, a computer device is provided, including a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the steps in each of the above method embodiments.

[0089] In an exemplary embodiment, a computer readable storage medium is provided, storing a computer program, which is executed by a processor to implement the steps in each of the above method embodiments.

[0090] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in the present application are all information and data authorized by the user or authorized by all parties.

[0091] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer readable storage medium, and when executed, can include the processes of the above-mentioned embodiment methods. Any reference to memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical storage, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration but not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc.

[0092] The database involved in the embodiments provided in the present application can include at least one of a relational database and a non-relational database. The non-relational database can include a distributed database based on a blockchain, etc., without being limited thereto. The processor involved in the embodiments provided in the present application can be a general-purpose processor, a central processing unit, a graphics processing unit, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., without being limited thereto.

[0093] The technical features of the above embodiments can be combined arbitrarily. In order to make the description simple, all possible combinations of the technical features in the above embodiments are not described, however, as long as the combinations of the technical features do not exist contradictory, they should be considered as the scope of the present application.

[0094] The principles and implementation manners of the present application are described herein by using specific examples, and the above examples are only used to help understand the method of the present application and its core idea; meanwhile, for those skilled in the art, according to the idea of the present application, the specific implementation manners and application ranges will have changes. In conclusion, the content of the specification should not be understood as a limitation of the present application.

Claims

1. A method for predicting satellite orbit position parameters, characterized by, The application relates to a method for predicting a kilometer marker of a railway track section. The method comprises the following steps: acquiring satellite positioning data of a plurality of key position points on the railway track section measured by satellite positioning, and converting longitude and latitude in the data into UTM coordinates; projecting the UTM coordinates on a line segment between position points with known kilometer markers on the railway track section to determine the kilometer markers of the key position points; splicing lag features of a plurality of time steps into the satellite positioning data and the kilometer markers of the key position points respectively to obtain extended feature matrices of the satellite positioning data and the kilometer markers at the key position points; determining difference features and moving average features of the extended feature matrices, and generating final feature matrices of the extended feature matrices and the difference features and the moving average features thereof; performing robust standardization on the final feature matrices to obtain standard feature matrices of the satellite positioning data and the kilometer markers at the key position points; training a double-layer long short-term memory network by taking satellite feature sequences of each time step in the standard feature matrices of the satellite positioning data at the key position points as input and taking kilometer markers of each time step in the standard feature matrices of the kilometer markers at the key position points as labels to obtain a kilometer marker prediction model; 2. The method of claim 1, wherein, inputting satellite feature sequences of a to-be-measured position point on the railway track section into the kilometer marker prediction model to output the kilometer marker of the to-be-measured position point on the railway track section. The method further comprises the following steps of converting longitude and latitude in the data into UTM coordinates: According to the ellipsoid parameters and the latitude in the satellite positioning data, the radius of curvature is determined using the formula and ; in the formula, is the radius of curvature, is the long semi-axis of the ellipsoid, is the eccentricity, is the latitude, and is the ellipsoid parameter. The UTM East coordinate is determined using the formula , where is the UTM East coordinate, is the East pseudo offset, is the scale factor, is the longitude difference term, , is the longitude, is the longitude of the central meridian, is the latitude tangent squared term, , is the correction term related to the eccentricity and latitude, ; The UTM North coordinate is determined using the formula , where is the UTM North coordinate, is the ellipsoidal distance from the equator to the latitude, ; determining a UTM partition number by using longitude in the satellite positioning data; 3. The method of claim 1, wherein, constructing the UTM coordinates by using the UTM partition number, a UTM east coordinate and a UTM north coordinate. The method further comprises the following steps of projecting the UTM coordinates on the line segment to determine the kilometer markers of the key position points: According to the UTM coordinates of each key position point, the end points of each line segment, the projection parameters of the UTM coordinates of each key position point on each line segment are determined by using the formula , wherein, is the projection parameter, , is the distance from the UTM coordinates of the key position point to one end point of the line segment, , is the UTM coordinates of the key position point, is the coordinate of one end point of the line segment, is the distance from the UTM coordinates of the key position point to the other end point of the line segment, , is the coordinate of the other end point of the line segment. According to the projection parameter, the projection point coordinates of the UTM coordinates of each key position point on each line segment are determined by using a formula , wherein is the projection point coordinates. determining line segments between all position points with known kilometer markers on the railway track section; If the selected projection point coordinate is on the projection line segment, then the kilometer marker of each key position point is determined by using the formula , where is the kilometer marker of the key position point, is the kilometer marker of one end point of the projection line segment, is the kilometer marker of the other end point of the projection line segment. selecting line segments and projection point coordinates corresponding to the minimum projection distance according to the UTM coordinates and the projection point coordinates of the key position points, and taking the selected line segments as projection line segments; 4. The method of claim 1, wherein, if the selected projection point coordinates are outside the projection line segments, determining the kilometer marker of the nearest end point of the projection line segments to the projection point coordinates as the kilometer marker of the key position points. The method further comprises the following steps of splicing lag features of a plurality of time steps into the satellite positioning data of the key position points to obtain the extended feature matrices of the satellite positioning data at the key position points: constructing a lag feature vector by using satellite positioning data of a plurality of time steps of the satellite positioning data; 5. The method of claim 1, wherein, splicing the satellite positioning data of the key position points and the lag feature vector to obtain the extended feature matrices of the satellite positioning data at the key position points. The difference features of each extended feature matrix are calculated using the formula , where is the difference feature of the extended feature matrix at time step , is the extended feature matrix at time step , is the extended feature matrix at time step ; The moving average features of each extended feature matrix are calculated using the formula wherein is the time step of the moving average feature of the extended feature matrix, is the extended feature matrix at time step , and is the offset within the window, =0,1,2. The method further comprises the following steps of determining the difference features and the moving average features of the extended feature matrices, and generating the final feature matrices of the extended feature matrices and the difference features and the moving average features thereof:

6. The method of claim 1, wherein, merging the extended feature matrices and the corresponding difference features and moving average features to obtain the final feature matrices. The method further comprises the following steps of performing robust standardization on the final feature matrices to obtain the standard feature matrices of the satellite positioning data and the kilometer markers at the key position points: determining the median and the interquartile range of the final feature matrices; According to the median and the interquartile range, a formula is used The final feature matrix is standardized to obtain a standard feature matrix of the satellite positioning data and the kilometer mark at each key position point; in the formula, is the standardized feature value of the th feature in the final feature matrix, is the th feature in the final feature matrix, is the median of the th feature in the final feature matrix, is the interquartile range of the th feature in the final feature matrix.

7. The method of claim 1, wherein, The double-layer long short-term memory network comprises a first long short-term memory network, a first Dropout layer, a second long short-term memory network, a second Dropout layer and a full connection layer connected in sequence. The double-layer long short-term memory network has an adaptive learning rate.

8. A satellite orbit position parameter prediction device characterized by comprising: The method comprises the following steps: The conversion module is configured to acquire satellite positioning data of a plurality of key position points on a railway track section measured by satellite positioning, and convert longitude and latitude in the data into UTM coordinates; The vector projection module is configured to perform vector projection on the UTM coordinates on a line segment between position points with known kilometer marks on the railway track section, to determine kilometer marks of the key position points; The feature splicing module is configured to splice lag features of multiple time steps for the satellite positioning data and the kilometer marks of the key position points respectively, to obtain extended feature matrices of the satellite positioning data and the kilometer marks at the key position points; The feature merging module is configured to determine difference features and moving average features of the extended feature matrices, and generate a final feature matrix from the extended feature matrices and the difference features and the moving average features thereof; The standard module is configured to perform robust standardization on the final feature matrices, to obtain standard feature matrices of the satellite positioning data and the kilometer marks at the key position points; The training module is configured to train the double-layer long short-term memory network by taking satellite features in the standard feature matrices of the satellite positioning data at the key position points as input and taking kilometer marks in the standard feature matrices of the kilometer marks at the key position points as labels, to obtain a kilometer mark prediction model; The application module is configured to input satellite feature sequences of a to-be-measured position point on the railway track section into the kilometer mark prediction model, and output a kilometer mark of the to-be-measured position point on the railway track section.

9. A computer device comprising: The memory, the processor and the computer program stored in the memory and executable on the processor are characterized in that the processor executes the computer program to implement the satellite track kilometer mark parameter prediction method in any one of claims 1-7.

10. A computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to implement the satellite track kilometer mark parameter prediction method in any one of claims 1-7.

Images