A railway full-line track state determination method and device

Abstract

The application provides a railway full-line track state determination method and device, relates to the field of track vehicle safe operation state evaluation, and comprises the following steps: performing space-time data registration on operation state time series data of a track vehicle and corresponding line space position data to obtain a space-time data set; performing space-time feature extraction on the space-time data set to generate corresponding space-time feature values; inputting the space-time feature values into a pre-constructed railway full-line track state prediction model to obtain a railway full-line track state; and wherein the railway full-line track state prediction model is pre-constructed according to historical operation state time series data and corresponding historical line space position data. The application can predict the railway full-line track state.

Description

Technical Field

[0001] This application relates to the field of safe operation status assessment of rail vehicles, specifically a method and device for determining the track status of the entire railway line. Background Technology

[0002] In the field of rail vehicle safety operation status assessment, existing onboard equipment monitoring systems and ground equipment monitoring systems are mainly used to assess the safety operation status of the rail vehicles themselves. For example, the Train Control System (TCDS) is an onboard equipment monitoring system that uses acceleration sensors installed on the bogies to acquire vehicle vibration signals and uses signal feature extraction and statistical analysis methods to assess the vehicle's operating status. Another example is the Trackside Dynamic Monitoring System (TPDS), a ground equipment monitoring system that uses fixed ground facilities to monitor mobile equipment on the vehicle.

[0003] However, compared to methods for assessing the safe operating status of rail vehicles themselves, current methods for fault detection and health status assessment of various infrastructure along railway lines (such as bridges, tunnels, culverts, roadbeds, rails, turnouts, and sleepers) are relatively outdated. These methods generally include: first, monitoring the condition of tracks, rails, and bridges using dedicated inspection vehicles, such as high-speed integrated inspection trains, rail flaw detectors, and tunnel inspection vehicles; second, installing monitoring equipment on the ground and analyzing the monitoring data collected by specific sensors. However, these methods suffer from at least the following technical problems:

[0004] First, the various monitoring devices or systems are independent and scattered, with a wide variety of types and inconsistent data storage formats, making subsequent data processing difficult.

[0005] Secondly, current fault detection and health status assessment of railway infrastructure are characterized by heterogeneity and uncertainty due to various factors such as geographical environment, design and construction, transportation organization, and maintenance history. Existing technologies for modeling the deterioration patterns of railway infrastructure generally use uniform empirical formulas or statistical models, neglecting the heterogeneous nature of infrastructure deterioration. This leads to significant errors between predicted and actual values ​​when using models to predict the safety status of infrastructure.

[0006] Therefore, how to more accurately describe the degree of deterioration of the infrastructure along the entire railway line, and thus reflect the condition of the entire railway track, is a technical challenge. Summary of the Invention

[0007] To address the problems in the prior art, this application provides a method and apparatus for determining the track condition of the entire railway line, which can predict the track condition of the entire railway line.

[0008] To solve the above-mentioned technical problems, this application provides the following technical solution:

[0009] Firstly, this application provides a method for determining the track condition of an entire railway line, including:

[0010] The time-series data of the operation status of rail vehicles are spatiotemporally registered with the corresponding spatial location data of the line to obtain a spatiotemporal data set;

[0011] Spatiotemporal features are extracted from the spatiotemporal data set to generate corresponding spatiotemporal feature values;

[0012] The spatiotemporal feature values ​​are input into a pre-constructed railway track state prediction model to obtain the track state of the entire railway line; wherein, the railway track state prediction model is pre-constructed based on historical operating status time series data and corresponding historical line spatial location data.

[0013] Furthermore, before performing spatiotemporal data registration between the time-series data of the rail vehicle's operating status and the corresponding spatial location data of the line, the following steps are also included:

[0014] An infrastructure data dictionary is generated based on the up and down kilometer markers and length information of the preset road segments in the spatial location data of the route;

[0015] The spatial location data of the route is corrected based on the acquired video images of vehicle operation and the infrastructure data dictionary.

[0016] Further, the operational status time-series data includes: the acceleration value of the rail vehicle during operation; the extraction of spatiotemporal features from the spatiotemporal data set to generate corresponding spatiotemporal feature values ​​includes:

[0017] Determine the sliding window for each space based on the spatial location data of the line;

[0018] Calculate the spatiotemporal characteristic value corresponding to each spatial sliding window based on the acceleration value corresponding to each spatial sliding window.

[0019] Further, the spatiotemporal characteristic values ​​include: the orbital impact index; the calculation of the spatiotemporal characteristic values ​​corresponding to each spatial sliding window based on the acceleration values ​​corresponding to each spatial sliding window includes:

[0020] The track impact index is calculated based on the preset sampling frequency and the acceleration value.

[0021] Further, the spatiotemporal characteristic values ​​include: a stationarity index; the calculation of the spatiotemporal characteristic values ​​corresponding to each spatial sliding window based on the acceleration values ​​corresponding to each spatial sliding window includes:

[0022] The stability index is calculated based on the preset sampling frequency, frequency correction coefficient, and acceleration value.

[0023] Further, the spatiotemporal feature values ​​include: effective acceleration values; the calculation of the spatiotemporal feature values ​​corresponding to each spatial sliding window based on the acceleration values ​​corresponding to each spatial sliding window includes:

[0024] The effective value of acceleration is calculated based on the acceleration value and the number of spatial sliding windows.

[0025] Further, the spatiotemporal feature values ​​include: the maximum acceleration value; the calculation of the spatiotemporal feature values ​​corresponding to each spatial sliding window based on the acceleration values ​​corresponding to each spatial sliding window includes:

[0026] The acceleration values ​​corresponding to each spatial sliding window are compared to determine the maximum acceleration value.

[0027] Furthermore, the spatiotemporal feature values ​​are input into a pre-constructed railway track state prediction model to obtain the track state of the entire railway line, including:

[0028] Spatiotemporal data registration is performed on the time-series data of the operating status of rail vehicles and the corresponding spatial location data of the line;

[0029] Spatiotemporal features are extracted from the time-series data of the operation status and the spatial location data of the line after the spatiotemporal data registration is completed, and corresponding stability index, effective value of acceleration, maximum value of acceleration and track impact index are generated.

[0030] Based on the stability index, effective value of acceleration, maximum value of acceleration, track impact index, and the track state prediction model for the entire railway line, the track state of the entire railway line is obtained.

[0031] Secondly, this application provides a railway track condition determination device, comprising:

[0032] The spatiotemporal registration unit is used to perform spatiotemporal data registration between the time-series data of the operating status of rail vehicles and the corresponding spatial location data of the line, so as to obtain a spatiotemporal data set;

[0033] The spatiotemporal feature extraction unit is used to extract spatiotemporal features from the spatiotemporal data set and generate corresponding spatiotemporal feature values.

[0034] The track state determination unit is used to input the spatiotemporal feature values ​​into a pre-constructed railway track state prediction model to obtain the track state of the entire railway line; wherein, the railway track state prediction model is pre-constructed based on historical operating status time series data and corresponding historical line spatial location data.

[0035] Furthermore, the railway track condition determination device for the entire line also includes:

[0036] The data dictionary generation unit is used to generate an infrastructure data dictionary based on the up and down kilometer markers and length information of the preset road segments in the spatial location data of the line;

[0037] The spatial data correction unit is used to correct the spatial location data of the route based on the acquired vehicle operation video images and the infrastructure data dictionary.

[0038] Furthermore, the operational status time-series data includes: the acceleration value of the rail vehicle during operation; the spatiotemporal feature extraction unit includes:

[0039] The sliding window determination module is used to determine each spatial sliding window based on the line spatial location data.

[0040] The spatiotemporal feature extraction module is used to calculate the spatiotemporal feature values ​​corresponding to each spatial sliding window based on the acceleration values ​​corresponding to each spatial sliding window.

[0041] Furthermore, the spatiotemporal feature value includes: the orbital impact index; the spatiotemporal feature extraction module is specifically used to calculate the orbital impact index based on a preset sampling frequency and the acceleration value.

[0042] Furthermore, the spatiotemporal feature values ​​include: a stationarity index; the spatiotemporal feature extraction module is specifically used to calculate the stationarity index based on a preset sampling frequency, a frequency correction coefficient, and the acceleration value.

[0043] Furthermore, the spatiotemporal feature values ​​include: effective acceleration values; the spatiotemporal feature extraction module is specifically used to calculate the effective acceleration values ​​based on the acceleration values ​​and the number of spatial sliding windows.

[0044] Furthermore, the spatiotemporal feature value includes: the maximum acceleration value; the spatiotemporal feature extraction module is specifically used to compare the acceleration values ​​corresponding to each spatial sliding window to determine the maximum acceleration value.

[0045] Furthermore, the track condition determination device for the entire railway line, wherein the track condition determination unit includes:

[0046] The spatiotemporal registration module is used to perform spatiotemporal data registration on the time-series data of the operating status of rail vehicles and the corresponding spatial location data of the line.

[0047] The spatiotemporal feature extraction module is used to extract spatiotemporal features from the time-series data of the operation status and the spatial location data of the line after the spatiotemporal data registration is completed, and to generate the corresponding stability index, effective value of acceleration, maximum value of acceleration and track impact index.

[0048] The track condition determination module obtains the track condition of the entire railway line based on the stability index, effective value of acceleration, maximum value of acceleration, track impact index, and the track condition prediction model of the entire railway line.

[0049] Thirdly, this application provides an electronic device including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the steps of the method for determining the track status of the entire railway line.

[0050] Fourthly, this application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method for determining the track state of the entire railway line.

[0051] Fifthly, this application provides a computer program product, including a computer program / instructions, which, when executed by a processor, implement the steps of the method for determining the track status of the entire railway line.

[0052] To address the problems in existing technologies, the method and apparatus for determining the track status of the entire railway line provided in this application can monitor the vehicle status of different track vehicles under the railway line under test, obtain the time-series data of the track vehicle's operating status, and then combine it with the spatial location of the track at various stations, bridges, tunnels, etc. on the railway line under test, thereby constructing a spatiotemporal big data set of "vehicle-ground monitoring". By studying the evolution law of the full spatial monitoring data of the line over time, a trend evolution model of the track health status of the entire line is established, thereby helping technicians to arrange maintenance plans in advance, improve railway operation efficiency, realize intelligent auxiliary decision-making on the operating status of track equipment on the railway line, and have important significance for reducing the maintenance cost of railway lines. Attached Figure Description

[0053] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0054] Figure 1 This is one of the flowcharts for the method of determining the track status of the entire railway line in the embodiments of this application;

[0055] Figure 2 This is the second flowchart of the method for determining the track status of the entire railway line in the embodiments of this application;

[0056] Figure 3This is a flowchart illustrating the generation of corresponding spatiotemporal feature values ​​in the embodiments of this application;

[0057] Figure 4 This is a flowchart illustrating the construction of a railway track condition prediction model in this embodiment of the application;

[0058] Figure 5 This is one of the structural diagrams of the railway track condition determination device in the embodiments of this application;

[0059] Figure 6 This is the second structural diagram of the railway track condition determination device in the embodiments of this application;

[0060] Figure 7 This is a structural diagram of the spatiotemporal feature extraction unit in an embodiment of this application;

[0061] Figure 8 This is the third structural diagram of the railway track condition determination device in the embodiments of this application;

[0062] Figure 9 This is a schematic diagram of the structure of the electronic device in the embodiments of this application;

[0063] Figure 10 This is a schematic diagram of the health diagnosis process for track equipment based on monitoring data in an embodiment of this application;

[0064] Figure 11 This is a schematic diagram of the health status classification assessment process based on graph neural networks in an embodiment of this application;

[0065] Figure 12 This is a schematic diagram illustrating the effect of training the spatiotemporal data prediction model in the embodiments of this application;

[0066] Figure 13 This is a flowchart illustrating the track status of the entire railway line in this embodiment of the application. Detailed Implementation

[0067] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0068] In one embodiment, see Figure 1 In order to predict the track condition of the entire railway line, this application provides a method for determining the track condition of the entire railway line, including:

[0069] S101: Perform spatiotemporal data registration between the time-series data of the rail vehicle's operating status and the corresponding spatial location data of the line to obtain a spatiotemporal data set;

[0070] S102: Extract spatiotemporal features from the spatiotemporal data set to generate corresponding spatiotemporal feature values;

[0071] S103: Input the spatiotemporal feature values ​​into the pre-constructed railway track state prediction model to obtain the railway track state; wherein, the railway track state prediction model is pre-constructed based on historical operating status time series data and corresponding historical track spatial location data.

[0072] It is understandable that, in order to solve the aforementioned technical problems, this application embodiment combines the positive characteristics of railway track equipment, such as linearity and continuity, as well as the deterioration characteristics such as heterogeneity, uncertainty, memory, and linkage. Instead of assessing the track (safety / health) status of railway lines from the perspective of the track equipment's engineering, electrical, or traction power supply, it starts from the perspective of the line's spatial location, integrates the entire life cycle data of the railway track, and proposes a method for assessing and calculating the track (safety / health) status of railway lines.

[0073] In this embodiment, the rail vehicle and the entire railway track can be regarded as a coupled vibration system. By integrating vibration monitoring data collected by the rail vehicle at small locations on the track, kilometer marker data from the train operation monitoring and recording device (LKJ system), and basic track maintenance data, a large spatiotemporal dataset is continuously accumulated to determine the deterioration and evolution trend of the railway track.

[0074] Generally, rail vehicles operate along fixed routes, and the onboard train operation monitoring and recording (LKJ) system records the vehicle's location, kilometer markers, and operating conditions throughout its entire journey. For a given fixed line, as numerous rail vehicles pass through tunnels, bridges, and turnouts at various geographical locations, the onboard monitoring equipment continuously accumulates real-time and spatial dynamic monitoring data with both temporal and spatial attributes. This vast, ever-accumulating data is of immense value in assessing the fault and health status of the track along that line.

[0075] How to mine massive amounts of real-time and spatial dynamic monitoring data of locomotives and rolling stock to detect track faults and assess the health status of railway lines is a novel research direction in the era of industrial big data. It is of great significance for reducing the maintenance costs of railway infrastructure. This application innovatively monitors the status of different operating track vehicles on a fixed railway line, forming time-series data of their operating status. This time-series data is integrated with the spatial location information of railway infrastructure such as stations, bridges, and tunnels (corresponding to the spatial location data of the railway line) to construct a "vehicle-to-ground monitoring" spatiotemporal big data system. By studying the evolution of spatial monitoring data across the entire railway line over time, a trend evolution model of the health status of the entire track across the entire line is established, enabling intelligent auxiliary decision-making for the operational status of track equipment.

[0076] Figure 11 This is a flowchart illustrating the health diagnosis process for track equipment. Based on modeling and / or statistical analysis methods, the state parameters are analyzed through multi-parameter fusion diagnostic logic mechanisms and diagnostic models to clarify the running gear status. Corresponding measures are then provided based on different state levels (calculated using a pre-built railway track condition prediction model).

[0077] As described above, the method for determining the track status of the entire railway line provided in this application can monitor the vehicle status of different track vehicles under the railway line under test, obtain the time-series data of the track vehicle's operating status, and then combine it with the spatial location of the track at various stations, bridges, tunnels, etc. on the railway line under test, thereby constructing a spatiotemporal big data set of "vehicle-ground monitoring". By studying the evolution law of the full spatial monitoring data of the line over time, a trend evolution model of the track health status of the entire line is established, thereby helping technicians to arrange maintenance plans in advance, improve railway operation efficiency, realize intelligent auxiliary decision-making on the operating status of track equipment on the railway line, and is of great significance for reducing the maintenance cost of railway lines.

[0078] In one embodiment, see Figure 2 Before performing spatiotemporal data registration between the time-series data of the rail vehicle's operating status and the corresponding spatial location data of the line, the process also includes:

[0079] S201: Generate an infrastructure data dictionary based on the up and down kilometer markers and length information of the preset road segments in the spatial location data of the line; wherein, the preset road segments include, but are not limited to, stations, bridges and tunnels, etc.

[0080] S202: Correct the spatial location data of the route based on the acquired vehicle operation video images and the infrastructure data dictionary.

[0081] It is understandable that spatiotemporal data registration refers to mapping the time sequence data of the operating status with the spatial location data of the line in order to know the operating status of each line in each time period or the spatial location status of the entire line in each time period.

[0082] Before performing spatiotemporal data registration between the time-series data of the rail vehicle's operating status and the corresponding spatial location data of the line, it is first necessary to obtain the time-series data of the operating status and the corresponding spatial location data of the line.

[0083] For operational status time-series data, an existing railway operating line can be selected as the research target. Status monitoring data of track vehicles under fixed lines during operation can be collected, such as bogie acceleration data, up and down kilometer markers, speed, operating conditions and cumulative operating time, etc., and transmitted to the ground diagnostic system (the execution subject of this application embodiment, the entity can be a server).

[0084] For track spatial location data, an onboard communication protocol can be developed between onboard equipment of rail vehicles (such as passenger car traffic safety monitoring system and train operation monitoring and recording device) so that the onboard equipment host can obtain the up and down kilometer markers, track data and other data recorded by the train operation monitoring and recording device through Ethernet.

[0085] To better achieve spatiotemporal data registration, information from the entire route can be collected, including but not limited to the kilometer markers and lengths of stations, bridges, and tunnels along the fixed route. This forms an infrastructure data dictionary; the spatial location data of the route is then corrected using pre-acquired vehicle operation video images and the infrastructure data dictionary.

[0086] It should be noted that, in order to achieve better registration results, abnormal up and down kilometer marker data can be preprocessed. For example, for illogical abnormal data jumps, the historical average value of the track vehicle on that data item can be used to replace the abnormal value; for missing kilometer marker data, regression fitting can be performed (by using the least squares method to establish a fitting curve of spatial location points on the track and speed and running time, thereby solving the problem of missing kilometer marker data).

[0087] Establish a unified spatiotemporal data reference space. Spatiotemporal data refers to observational indicators sampled at fixed time intervals over a fixed continuous spatial region. For vehicle-mounted time-series vibration monitoring data and the three-dimensional information space of vehicle-mounted data, unified spatiotemporal registration is performed according to train number (the unique identifier of the entire train), time, route, and kilometer marker. Through database stored procedures and views, different data retrieval methods, such as grouping by time and spatial location, are implemented, forming database retrieval tables for vehicle searches based on time and kilometer marker location. This achieves spatiotemporal registration of operational status time-series data and track spatial location data. The spatiotemporally registered data forms a spatiotemporal dataset, from which spatiotemporal features can be extracted subsequently.

[0088] As can be seen from the above description, the railway track condition determination method provided in this application can perform spatiotemporal data registration between the time-series data of the operating status of rail vehicles and the corresponding spatial location data of the line.

[0089] In one embodiment, see Figure 3 The operational time-series data includes: the acceleration values ​​of the rail vehicle during operation; the extraction of spatiotemporal features from the spatiotemporal data set to generate corresponding spatiotemporal feature values ​​includes:

[0090] S301: Determine each spatial sliding window based on the spatial location data of the line;

[0091] S302: Calculate the spatiotemporal characteristic value corresponding to each spatial sliding window based on the acceleration value corresponding to each spatial sliding window.

[0092] Specifically, the spatiotemporal characteristic value includes: the orbital impact index; the calculation of the spatiotemporal characteristic value corresponding to each spatial sliding window based on the acceleration value corresponding to each spatial sliding window includes: calculating the orbital impact index based on a preset sampling frequency and the acceleration value.

[0093] Specifically, the spatiotemporal characteristic values ​​include: a stationarity index; the calculation of the spatiotemporal characteristic values ​​corresponding to each spatial sliding window based on the acceleration values ​​corresponding to each spatial sliding window includes: calculating the stationarity index based on a preset sampling frequency, a frequency correction coefficient, and the acceleration values.

[0094] Specifically, the spatiotemporal feature values ​​include: effective acceleration values; the calculation of the spatiotemporal feature values ​​corresponding to each spatial sliding window based on the acceleration values ​​corresponding to each spatial sliding window includes: calculating the effective acceleration values ​​based on the acceleration values ​​and the number of spatial sliding windows.

[0095] Specifically, the spatiotemporal feature value includes: the maximum acceleration value; the calculation of the spatiotemporal feature value corresponding to each spatial sliding window based on the acceleration value corresponding to each spatial sliding window includes: comparing the acceleration values ​​corresponding to each spatial sliding window to determine the maximum acceleration value.

[0096] It is understandable that long-term data tracking of all rail vehicles passing through fixed lines results in a massive spatiotemporal dataset of raw bogie acceleration data. The data in this spatiotemporal dataset of raw vibration data consists of time-series vibration signal sequences, characterized by large data volume, high spatial dimensionality, and high complexity. This application's embodiments employ a model-driven and data-driven approach to extract spatiotemporal features. On the one hand, it preserves as many vibration characteristics as possible of the coupled vibration system composed of the vehicle and the track; on the other hand, it retains as many statistical characteristics of the massive time-series signals as possible in the time domain, frequency domain, and time-frequency analysis.

[0097] Therefore, data preprocessing is performed first. The spatial location of the entire fixed track is divided into J equally spaced spaces (for example, if the length is 2 kilometers and the equal interval is 200 meters, then J is 10) (equivalent to the spatial sliding window mentioned above). For the raw acceleration data, one intersection is taken as a data sampling sample, and the following four features are calculated per second to form a feature vector:

[0098] Where i represents vertical or horizontal; j represents the location information on the line, and the point located in the j-th spatial window is represented as (j-1)×200≤location≤j×200; K represents the different time slices (i.e. different routes) through which the vehicle passes through the spatial location.

[0099] Specifically, the orbital impact index (V ijk ):

[0100] i = 1, 2, where i = 1 represents the lateral direction and i = 2 represents the vertical direction (vehicle-mounted acceleration data is divided into lateral and vertical directions, and all features in this patent are divided into lateral and vertical directions). x j f represents acceleration. s The sampling frequency is 1 second.

[0101] Stationarity index (W) ijk ):

[0102] According to GB 5599-2019 "Specifications for Evaluation and Testing of Dynamic Performance of Locomotives and Rolling Stock", the stability index of rail vehicles is calculated based on a 5-second time interval with 1-second sliding intervals. The formula is as follows: i = 1, 2, where i represents horizontal or vertical; A i Represents the main frequency amplitude; f sThe sampling frequency is 1 second, in Hz; F(f) -- frequency correction coefficient, see Table 1.

[0103] Table 1 Frequency Correction Factors

[0104]

[0105] RMS acceleration value:

[0106] f s A sampling frequency of 1 second

[0107] Maximum acceleration:

[0108] X max =max(x1,x2,...,x fs ), f s The sampling frequency is 1 second.

[0109] Preprocessed feature vectors The mean and variance of the above characteristics are calculated using a sliding spatial window (each window is taken at equal intervals of 200 meters). The calculation of the mean and variance is illustrated using the orbital impact index as an example:

[0110] Mean of orbital impact index

[0111] Orbital Impact Index Variance

[0112] Where n is the number of data points within the j-th interval of the spatial slice.

[0113] The spatiotemporal characteristics of a single intersection sampling are obtained.

[0114]

[0115] In one embodiment, see Figure 13 The spatiotemporal feature values ​​are input into a pre-constructed railway track state prediction model to obtain the track state of the entire railway line, including:

[0116] S401: Perform spatiotemporal data registration on the time sequence data of the operating status of rail vehicles and the corresponding spatial location data of the line;

[0117] S402: Extract spatiotemporal features from the time-series data of the operation status and the spatial location data of the line after the spatiotemporal data registration is completed, and generate the corresponding stability index, effective value of acceleration, maximum value of acceleration and track impact index.

[0118] S403: Based on the stability index, effective value of acceleration, maximum value of acceleration, track impact index, and the track state prediction model of the entire railway line, the track state of the entire railway line is obtained; wherein, the track state prediction model of the entire railway line is obtained by training a graph neural network model.

[0119] Specifically, the mean and variance of the stability index, effective value of acceleration, maximum value of acceleration, and track impact index are input into the track state prediction model of the entire railway line to obtain the track state of the entire railway line.

[0120] It should be noted that, in one embodiment, see [reference needed]. Figure 4 The steps for constructing the track condition prediction model for the entire railway line include:

[0121] S401': Perform historical spatiotemporal data registration on the historical operating status time-series data of rail vehicles and the corresponding historical line spatial location data;

[0122] S402': Extract historical spatiotemporal features from the time-series data of the operation status and the spatial location data of the line after the historical spatiotemporal data registration is completed, and generate the corresponding historical stability index, historical effective value of acceleration, historical maximum value of acceleration and historical track impact index.

[0123] S403': Input the mean and variance of the historical stability index, historical effective acceleration value, historical maximum acceleration value, and historical track impact index into the graph neural network model for training to obtain the historical track status of the entire railway line; wherein, the mean and variance of the historical stability index, historical effective acceleration value, historical maximum acceleration value, and historical track impact index need to be pre-calculated, and the calculation method is the same as the method for calculating the mean and variance of the stability index, effective acceleration value, maximum acceleration value, and track impact index, as detailed in the aforementioned formula;

[0124] S404': Generate the railway track condition prediction model based on the preset railway track condition rating strategy and the historical track condition of the entire railway line.

[0125] It is understood that this application's embodiments correlate the massive amount of raw vibration monitoring data accumulated by rail vehicles under fixed railway operating lines with the spatial information of railway lines, proposing and constructing a full-time and spatiotemporal large dataset for vibration monitoring under fixed railway operating lines. Using a combination of model and data-driven approaches, spatiotemporal feature values ​​are extracted, and a vibration characteristic characterizing the coupled vibration system of vehicles and tracks—the track impact index—is proposed. Simultaneously, as many statistical characteristics of the massive time-series signals in the time domain, frequency domain, and time-frequency analysis are preserved as much as possible.

[0126] This model is a full-line, all-temporal railway track condition prediction model based on Spatial-Temporal Graph Neural Networks (STGNN). The graph neural network model consists of several stacked spatial-temporal blocks (ST blocks). Through the spatial-temporal feature extraction machine of the spatial-temporal blocks and the spatial-temporal graph convolution module, a graded evaluation model of the entire line's degradation trend over time is constructed, divided into four levels: A (normal), B (weakened, requiring continuous monitoring), C (abnormal, requiring inspection and maintenance), and D (fault, requiring immediate repair). See [link to relevant documentation]. Figure 10 Using long-term track maintenance data, data from the entire track lifecycle, from normal operation to the next maintenance, were selected as test data, and data from a period after maintenance were selected as validation data. The results show that the model's prediction accuracy for track conditions is over 90%, validating the effectiveness of the proposed intelligent prediction model.

[0127] Figure 12 This is a spatiotemporal data prediction model (also known as a railway full-line track condition prediction model), with historical data T as input. h The graph neural network model generates a spatiotemporal signal matrix X for each time slice and outputs multi-step prediction results y. The model consists of stacked l spatiotemporal blocks (ST blocks); the input to the l-th spatiotemporal block is... Indicates; d model It is the feature dimension of the intermediate layer data. Each spatiotemporal module contains two key operations. The first is the spatiotemporal feature extraction operation, which dynamically calculates the correlation between the test data and the correlation between time slices based on the input data. The second is the spatiotemporal graph convolution module, which uses the spatial dimension graph convolution kernel to capture the correlation between the temporal and spatial dimensions in the one-dimensional convolution of the temporal dimension.

[0128] Based on spatiotemporal datasets

[0129] This paper proposes a Spatial-Temporal Graph Neural Network (STGNN) model, and the model training process is as follows:

[0130] (1) Input layer: Part of the spatiotemporal dataset is used as training samples, and the rest is used as test samples. The dataset is divided into K time slices and J spatial slices, and each sample is a single sampling in the spatiotemporal space.

[0131]

[0132] The STGNN network takes the graph structure and the features of each node as input, and outputs either node-level results or graph-level results. It typically uses an adjacency matrix A∈R. N×N Represents the structure of a graph, with features X∈R N×F Represent the feature values ​​of each node, Z∈R N×P This represents the output of the model. Each neural network layer can be represented as:

[0133] H l+1 =f(H l ,A), where H 0 =X,H l =Z.

[0134] Based on the input and output formats, it can be equivalent to:

[0135] f(H l+1 ,A)=δ(AH l W l )

[0136] (2) Spatiotemporal module convolution

[0137] Graph neural network models are composed of several spatiotemporal blocks (ST blocks) stacked together, see... Figure 11 As shown. Each spatiotemporal module first has a spatiotemporal feature extraction mechanism, which dynamically calculates the correlation between nodes and the strength of correlation between time slices based on the input data; secondly, it has a spatiotemporal graph convolution module, which uses graph convolution kernels in the spatial dimension and one-dimensional convolutions in the temporal dimension to capture the correlation in the temporal and spatiotemporal dimensions respectively.

[0138] This scheme selects 7 hidden spatiotemporal modules after comprehensive comparison through experiments.

[0139] (3) Output layer

[0140] As the output node for fault diagnosis, it is hoped that the fault mode can be determined directly from the output result. This patent connects a fully connected layer neural network (MLP) + softmax classifier after 7 spatiotemporal convolutional modules, outputting the probability between 0 and 1, and using one-shot encoding format and softmax output for cross-entropy loss. The four nodes of the final output layer are defined as 0001 0010 0100 1000, which correspond to level A (normal), level B (weakened, need to be continuously tracked), level C (abnormal, need to be checked and maintained), and level D (fault, need to be repaired immediately).

[0141] If the entire network is trained 1000 times, the training target error is 1e-6, and the learning rate is 0.01.

[0142] Specifically, the Levenberg-Marquardt optimization algorithm can be used to train the network.

[0143] The training error variation curve of the LM optimization algorithm is shown below. Figure 12 As shown, the training error is 9.4516e-7, and the target error is 1e-6. At this point, the network performance has met the requirements very well.

[0144] During the model validation phase, data from a period following maintenance can be selected as validation data. The results show that the model's prediction accuracy for orbital state is above 90%, validating the effectiveness of the proposed intelligent prediction model.

[0145] As can be seen from the above description, the method for determining the track condition of the entire railway line provided in this application can construct a prediction model for the track condition of the entire railway line.

[0146] Based on the same inventive concept, this application also provides a railway track condition determination device, which can be used to implement the method described in the above embodiments, as described in the following embodiments. Since the principle of the railway track condition determination device in solving the problem is similar to that of the railway track condition determination method, the implementation of the railway track condition determination device can refer to the implementation of the software performance benchmark-based determination method, and repeated details will not be elaborated further. As used below, the terms "unit" or "module" can refer to a combination of software and / or hardware that implements a predetermined function. Although the system described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0147] In one embodiment, see Figure 5 In order to predict the track condition of the entire railway line, this application provides a railway track condition determination device, comprising:

[0148] The spatiotemporal registration unit 501 is used to perform spatiotemporal data registration between the time-series data of the running status of the rail vehicle and the corresponding spatial location data of the line to obtain a spatiotemporal data set.

[0149] The spatiotemporal feature extraction unit 502 is used to extract spatiotemporal features from the spatiotemporal data set and generate corresponding spatiotemporal feature values.

[0150] The track state determination unit 503 is used to input the spatiotemporal feature values ​​into a pre-constructed railway track state prediction model to obtain the track state of the entire railway line; wherein, the railway track state prediction model is pre-constructed based on historical operating status time series data and corresponding historical line spatial location data.

[0151] In one embodiment, see Figure 6The railway track condition determination device for the entire line also includes:

[0152] The data dictionary generation unit 601 is used to generate an infrastructure data dictionary based on the up and down kilometer markers and length information of a preset road segment in the line spatial location data;

[0153] The spatial data correction unit 602 is used to correct the spatial location data of the route based on the acquired vehicle operation video images and the infrastructure data dictionary.

[0154] In one embodiment, see Figure 7 The operational status time-series data includes: the acceleration value of the rail vehicle during operation; the spatiotemporal feature extraction unit 602 includes:

[0155] The sliding window determination module 701 is used to determine each spatial sliding window based on the line spatial location data;

[0156] The spatiotemporal feature extraction module 702 is used to calculate the spatiotemporal feature value corresponding to each spatial sliding window based on the acceleration value corresponding to each spatial sliding window.

[0157] In one embodiment, the spatiotemporal feature value includes: an orbital impact index; the spatiotemporal feature extraction module is specifically used to calculate the orbital impact index based on a preset sampling frequency and the acceleration value.

[0158] In one embodiment, the spatiotemporal feature value includes: a stationarity index; the spatiotemporal feature extraction module is specifically used to calculate the stationarity index based on a preset sampling frequency, a frequency correction coefficient, and the acceleration value.

[0159] In one embodiment, the spatiotemporal feature value includes: an effective acceleration value; the spatiotemporal feature extraction module is specifically used to calculate the effective acceleration value based on the acceleration value and the number of spatial sliding windows.

[0160] In one embodiment, the spatiotemporal feature value includes: the maximum acceleration value; the spatiotemporal feature extraction module is specifically used to compare the acceleration values ​​corresponding to each spatial sliding window to determine the maximum acceleration value.

[0161] In one embodiment, see Figure 8 The railway track condition determination device for the entire line, track condition determination unit 503, includes:

[0162] The spatiotemporal registration module 801 is used to perform spatiotemporal data registration on the time sequence data of the operation status of rail vehicles and the corresponding spatial location data of the line.

[0163] The spatiotemporal feature extraction module 802 is used to extract spatiotemporal features from the time-series data of the running status and the spatial location data of the line after the spatiotemporal data registration is completed, and to generate corresponding stability index, effective value of acceleration, maximum value of acceleration and track impact index.

[0164] The track condition determination module 803 obtains the track condition of the entire railway line based on the stability index, effective value of acceleration, maximum value of acceleration, track impact index, and the track condition prediction model of the entire railway line.

[0165] From a hardware perspective, in order to predict the track condition of the entire railway line, this application provides an embodiment of an electronic device for implementing all or part of the aforementioned method for determining the track condition of the entire railway line. The electronic device specifically includes the following components:

[0166] The system comprises a processor, a memory, a communications interface, and a bus; wherein the processor, memory, and communications interface communicate with each other via the bus; the communications interface is used to realize information transmission between the railway track condition determination device and core business systems, user terminals, and related databases and other related equipment; the logic controller can be a desktop computer, tablet computer, or mobile terminal, etc., and this embodiment is not limited to these. In this embodiment, the logic controller can be implemented with reference to the embodiments of the railway track condition determination method and the railway track condition determination device in the embodiments, the contents of which are incorporated herein, and repeated details will not be described again.

[0167] It is understood that the user terminal may include smartphones, tablet computers, network set-top boxes, portable computers, desktop computers, personal digital assistants (PDAs), in-vehicle devices, smart wearable devices, etc. Among these, the smart wearable devices may include smart glasses, smartwatches, smart bracelets, etc.

[0168] In practical applications, some parts of the method for determining the track condition of the entire railway line can be executed on the electronic device side as described above, or all operations can be completed in the client device. The specific choice depends on the processing power of the client device and the limitations of the user's usage scenario. This application does not impose any limitations on this. If all operations are completed in the client device, the client device may further include a processor.

[0169] The aforementioned client device may have a communication module (i.e., a communication unit) that can communicate with a remote server to achieve data transmission. The server may include a server on the task scheduling center side; in other implementation scenarios, it may also include a server on an intermediate platform, such as a server on a third-party server platform that has a communication link with the task scheduling center server. The server may include a single computer device, a server cluster consisting of multiple servers, or a distributed server structure.

[0170] Figure 9 This is a schematic block diagram illustrating the system configuration of the electronic device 9600 according to an embodiment of this application. Figure 9 As shown, the electronic device 9600 may include a central processing unit 9100 and a memory 9140; the memory 9140 is coupled to the central processing unit 9100. It is worth noting that... Figure 9 This is an example; other types of structures can also be used to supplement or replace this structure to achieve telecommunications functions or other functions.

[0171] In one embodiment, the function of determining the track status of the entire railway line can be integrated into the central processing unit 9100. The central processing unit 9100 can be configured to perform the following control:

[0172] S101: Perform spatiotemporal data registration between the time-series data of the rail vehicle's operating status and the corresponding spatial location data of the line to obtain a spatiotemporal data set;

[0173] S102: Extract spatiotemporal features from the spatiotemporal data set to generate corresponding spatiotemporal feature values;

[0174] S103: Input the spatiotemporal feature values ​​into the pre-constructed railway track state prediction model to obtain the railway track state; wherein, the railway track state prediction model is pre-constructed based on historical operating status time series data and corresponding historical track spatial location data.

[0175] As described above, the method for determining the track status of the entire railway line provided in this application can monitor the vehicle status of different track vehicles under the railway line under test, obtain the time-series data of the track vehicle's operating status, and then combine it with the spatial location of the track at various stations, bridges, tunnels, etc. on the railway line under test, thereby constructing a spatiotemporal big data set of "vehicle-ground monitoring". By studying the evolution law of the full spatial monitoring data of the line over time, a trend evolution model of the track health status of the entire line is established, thereby helping technicians to arrange maintenance plans in advance, improve railway operation efficiency, realize intelligent auxiliary decision-making on the operating status of track equipment on the railway line, and is of great significance for reducing the maintenance cost of railway lines.

[0176] In another embodiment, the railway track condition determination device can be configured separately from the central processing unit 9100. For example, the data composite transmission device railway track condition determination device can be configured as a chip connected to the central processing unit 9100, and the function of the railway track condition determination method can be realized through the control of the central processing unit.

[0177] like Figure 9 As shown, the electronic device 9600 may further include: a communication module 9110, an input unit 9120, an audio processor 9130, a display 9160, and a power supply 9170. It is worth noting that the electronic device 9600 does not necessarily need to include these components. Figure 9 All components shown; in addition, the electronic device 9600 may also include Figure 9 For components not shown, please refer to existing technologies.

[0178] like Figure 9 As shown, the central processing unit 9100, sometimes also referred to as a controller or operating control, may include a microprocessor or other processor device and / or logic device, which receives inputs and controls the operation of various components of the electronic device 9600.

[0179] The memory 9140 may be, for example, one or more of a cache, flash memory, hard drive, removable media, volatile memory, non-volatile memory, or other suitable devices. It may store the aforementioned failure-related information, and also store a program for executing that information. The central processing unit 9100 may execute the program stored in the memory 9140 to perform information storage or processing, etc.

[0180] Input unit 9120 provides input to central processing unit 9100. Input unit 9120 may be, for example, a keypad or touch input device. Power supply 9170 provides power to electronic device 9600. Display 9160 displays images and text. Display may be, for example, an LCD display, but is not limited thereto.

[0181] The memory 9140 can be a solid-state memory, such as a read-only memory (ROM), random access memory (RAM), a SIM card, etc. It can also be a memory that retains information even when power is off, can be selectively erased, and contains more data; examples of this type of memory are sometimes referred to as EPROMs. The memory 9140 can also be some other type of device. The memory 9140 includes a buffer memory 9141 (sometimes referred to as a buffer). The memory 9140 may include an application / function storage unit 9142 for storing application programs and function programs or processes for executing the operation of the electronic device 9600 via the central processing unit 9100.

[0182] The memory 9140 may also include a data storage unit 9143 for storing data, such as contacts, digital data, pictures, sounds, and / or any other data used by the electronic device. The driver storage unit 9144 of the memory 9140 may include various drivers for the electronic device's communication functions and / or for performing other functions of the electronic device (such as messaging applications, address book applications, etc.).

[0183] The communication module 9110 is a transmitter / receiver 9110 that transmits and receives signals via the antenna 9111. The communication module (transmitter / receiver) 9110 is coupled to the central processing unit 9100 to provide input signals and receive output signals, which can be the same as in a conventional mobile communication terminal.

[0184] Based on different communication technologies, multiple communication modules 9110 can be configured in the same electronic device, such as cellular network modules, Bluetooth modules, and / or wireless LAN modules. The communication module (transmitter / receiver) 9110 is also coupled to a speaker 9131 and a microphone 9132 via an audio processor 9130 to provide audio output via the speaker 9131 and receive audio input from the microphone 9132, thereby realizing typical telecommunications functions. The audio processor 9130 may include any suitable buffer, decoder, amplifier, etc. Additionally, the audio processor 9130 is also coupled to a central processing unit 9100, enabling on-device recording via the microphone 9132 and on-device playback of stored sound via the speaker 9131.

[0185] Embodiments of this application also provide a computer-readable storage medium capable of implementing all steps of the railway track state determination method with a server or client as the execution subject in the above embodiments. The computer-readable storage medium stores a computer program that, when executed by a processor, implements all steps of the railway track state determination method with a server or client as the execution subject in the above embodiments. For example, when the processor executes the computer program, it implements the following steps:

[0186] S101: Perform spatiotemporal data registration between the time-series data of the rail vehicle's operating status and the corresponding spatial location data of the line to obtain a spatiotemporal data set;

[0187] S102: Extract spatiotemporal features from the spatiotemporal data set to generate corresponding spatiotemporal feature values;

[0188] S103: Input the spatiotemporal feature values ​​into the pre-constructed railway track state prediction model to obtain the railway track state; wherein, the railway track state prediction model is pre-constructed based on historical operating status time series data and corresponding historical track spatial location data.

[0189] As described above, the method for determining the track status of the entire railway line provided in this application can monitor the vehicle status of different track vehicles under the railway line under test, obtain the time-series data of the track vehicle's operating status, and then combine it with the spatial location of the track at various stations, bridges, tunnels, etc. on the railway line under test, thereby constructing a spatiotemporal big data set of "vehicle-ground monitoring". By studying the evolution law of the full spatial monitoring data of the line over time, a trend evolution model of the track health status of the entire line is established, thereby helping technicians to arrange maintenance plans in advance, improve railway operation efficiency, realize intelligent auxiliary decision-making on the operating status of track equipment on the railway line, and is of great significance for reducing the maintenance cost of railway lines.

[0190] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, apparatus, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0191] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (devices), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0192] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0193] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0194] Specific embodiments have been used to illustrate the principles and implementation methods of this invention. The descriptions of the embodiments above are only for the purpose of helping to understand the method and core ideas of this invention. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this invention. Therefore, the content of this specification should not be construed as a limitation of this invention.

Claims

1. A method for determining the track condition of an entire railway line, characterized in that, include: The time-series data of the operation status of rail vehicles are spatiotemporally registered with the corresponding spatial location data of the line to obtain a spatiotemporal data set; Spatiotemporal features are extracted from the spatiotemporal data set to generate corresponding spatiotemporal feature values; wherein, the spatiotemporal feature values ​​include the orbital impact index, stability index, effective acceleration value, and maximum acceleration value; The spatiotemporal feature values ​​are input into a pre-constructed railway track status prediction model to obtain the track status of the entire railway line; wherein, the railway track status prediction model is pre-constructed based on historical operating status time series data and corresponding historical line spatial location data; The steps for constructing the track condition prediction model for the entire railway line include: Historical spatiotemporal data registration is performed on the historical operating status time-series data of rail vehicles and the corresponding historical line spatial location data. Historical spatiotemporal features are extracted from the time-series data of the operation status and the spatial location data of the line after the historical spatiotemporal data registration is completed, and corresponding historical stability index, historical effective value of acceleration, historical maximum value of acceleration and historical track impact index are generated. The mean and variance of the historical stability index, historical effective value of acceleration, historical maximum value of acceleration and historical track impact index are input into the graph neural network model to train the historical track status of the entire railway line. Based on the preset railway track condition rating strategy and the historical track condition of the entire railway line, a railway track condition prediction model is generated; wherein, from the perspective of the spatial location of the line, the entire life cycle data of the railway track is integrated.

2. The method for determining the track condition of the entire railway line according to claim 1, characterized in that, Before performing spatiotemporal data registration between the time-series data of the rail vehicle's operating status and the corresponding spatial location data of the line, the following steps are also included: An infrastructure data dictionary is generated based on the up and down kilometer markers and length information of the preset road segments in the spatial location data of the route; The spatial location data of the route is corrected based on pre-acquired video images of vehicle operation and the infrastructure data dictionary.

3. The method for determining the track condition of the entire railway line according to claim 1, characterized in that, The operational status time-series data includes: the acceleration value of the rail vehicle during operation; the extraction of spatiotemporal features from the spatiotemporal data set to generate corresponding spatiotemporal feature values ​​includes: Determine the sliding window for each space based on the spatial location data of the line; Calculate the spatiotemporal characteristic value corresponding to each spatial sliding window based on the acceleration value corresponding to each spatial sliding window.

4. The method for determining the track condition of the entire railway line according to claim 3, characterized in that, The step of calculating the spatiotemporal characteristic value corresponding to each spatial sliding window based on the acceleration value corresponding to each spatial sliding window includes: The track impact index is calculated based on the preset sampling frequency and the acceleration value.

5. The method for determining the track condition of the entire railway line according to claim 3, characterized in that, The step of calculating the spatiotemporal characteristic value corresponding to each spatial sliding window based on the acceleration value corresponding to each spatial sliding window includes: The stability index is calculated based on the preset sampling frequency, frequency correction coefficient, and acceleration value.

6. The method for determining the track condition of the entire railway line according to claim 3, characterized in that, The step of calculating the spatiotemporal characteristic value corresponding to each spatial sliding window based on the acceleration value corresponding to each spatial sliding window includes: The effective value of acceleration is calculated based on the acceleration value and the number of spatial sliding windows.

7. The method for determining the track condition of the entire railway line according to claim 3, characterized in that, The step of calculating the spatiotemporal characteristic value corresponding to each spatial sliding window based on the acceleration value corresponding to each spatial sliding window includes: The acceleration values ​​corresponding to each spatial sliding window are compared to determine the maximum acceleration value.

8. The method for determining the track condition of the entire railway line according to claim 1, characterized in that, The step of inputting the spatiotemporal feature values ​​into a pre-constructed railway track state prediction model to obtain the track state of the entire railway line includes: Spatiotemporal data registration is performed on the time-series data of the operating status of rail vehicles and the corresponding spatial location data of the line; Spatiotemporal features are extracted from the time-series data of the operation status and the spatial location data of the line after the spatiotemporal data registration is completed, and corresponding stability index, effective value of acceleration, maximum value of acceleration and track impact index are generated. Based on the stability index, effective value of acceleration, maximum value of acceleration, track impact index, and the track state prediction model for the entire railway line, the track state of the entire railway line is obtained.

9. A device for determining the track condition of an entire railway line, characterized in that, include: The spatiotemporal registration unit is used to perform spatiotemporal data registration between the time-series data of the operating status of rail vehicles and the corresponding spatial location data of the line, so as to obtain a spatiotemporal data set; The spatiotemporal feature extraction unit is used to extract spatiotemporal features from the spatiotemporal data set and generate corresponding spatiotemporal feature values; wherein, the spatiotemporal feature values ​​include the orbital impact index, the stability index, the effective value of acceleration, and the maximum value of acceleration; The track state determination unit is used to input the spatiotemporal feature values ​​into a pre-constructed railway track state prediction model to obtain the track state of the entire railway line; wherein, the railway track state prediction model is pre-constructed based on historical operating status time series data and corresponding historical line spatial location data; The steps for constructing the track condition prediction model for the entire railway line include: Historical spatiotemporal data registration is performed on the historical operating status time-series data of rail vehicles and the corresponding historical line spatial location data. Historical spatiotemporal features are extracted from the time-series data of the operation status and the spatial location data of the line after the historical spatiotemporal data registration is completed, and corresponding historical stability index, historical effective value of acceleration, historical maximum value of acceleration and historical track impact index are generated. The mean and variance of the historical stability index, historical effective value of acceleration, historical maximum value of acceleration and historical track impact index are input into the graph neural network model to train the historical track status of the entire railway line. Based on the preset railway track condition rating strategy and the historical track condition of the entire railway line, a railway track condition prediction model is generated; wherein, from the perspective of the spatial location of the line, the entire life cycle data of the railway track is integrated.

10. The railway track condition determination device according to claim 9, characterized in that, Also includes: The data dictionary generation unit is used to generate an infrastructure data dictionary based on the up and down kilometer markers and length information of the preset road segments in the spatial location data of the line; The spatial data correction unit is used to correct the spatial location data of the route based on the pre-acquired vehicle operation video images and the infrastructure data dictionary.

11. The railway track condition determination device according to claim 9, characterized in that, The operational status time-series data includes: the acceleration value of the rail vehicle during operation; the spatiotemporal feature extraction unit includes: The sliding window determination module is used to determine each spatial sliding window based on the line spatial location data. The spatiotemporal feature extraction module is used to calculate the spatiotemporal feature values ​​corresponding to each spatial sliding window based on the acceleration values ​​corresponding to each spatial sliding window.

12. The railway track condition determination device according to claim 11, characterized in that, The spatiotemporal feature extraction module is specifically used to calculate the orbital impact index based on the preset sampling frequency and the acceleration value.

13. The railway track condition determination device according to claim 11, characterized in that, The spatiotemporal feature extraction module is specifically used to calculate the stationarity index based on the preset sampling frequency, frequency correction coefficient, and acceleration value.

14. The railway track condition determination device according to claim 11, characterized in that, The spatiotemporal feature extraction module is specifically used to calculate the effective value of acceleration based on the acceleration value and the number of spatial sliding windows.

15. The railway track condition determination device according to claim 11, characterized in that, The spatiotemporal feature extraction module is specifically used to compare the acceleration values ​​corresponding to each spatial sliding window in order to determine the maximum acceleration value.

16. The railway track condition determination device according to claim 9, characterized in that, The orbital state determination unit includes: The spatiotemporal registration module is used to perform spatiotemporal data registration on the time-series data of the operating status of rail vehicles and the corresponding spatial location data of the line. The spatiotemporal feature extraction module is used to extract spatiotemporal features from the time-series data of the operation status and the spatial location data of the line after the spatiotemporal data registration is completed, and to generate the corresponding stability index, effective value of acceleration, maximum value of acceleration and track impact index. The track condition determination module is used to obtain the track condition of the entire railway line based on the stability index, effective value of acceleration, maximum value of acceleration, track impact index, and the track condition prediction model of the entire railway line.

17. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the steps of the railway track condition determination method according to any one of claims 1 to 8.

18. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the computer program implements the steps of the method for determining the track condition of the entire railway line as described in any one of claims 1 to 8.

19. A computer program product comprising a computer program / instructions, characterized in that, When the computer program / instruction is executed by the processor, it implements the steps of the method for determining the track condition of the entire railway line as described in any one of claims 1 to 8.

Images