Track geometry deterioration early warning method and device based on vehicle-mounted line inspection instrument

Abstract

The application discloses a track geometry deterioration early warning method and device based on a vehicle-mounted line inspection instrument, and the method comprises the following steps: acquiring total data detected under multiple train services at the same mileage point; extracting invalid data from the total data under each train service by using a density clustering algorithm; comparing the invalid data proportion with a preset ratio range, and evaluating the data quality level of each train service according to the comparison result; selecting two groups of data, determining the significance level between the two groups of data by using the rank sum test method, and judging whether there is a significant difference between the two groups of data according to the significance level; determining that the data quality level classification meets the rationality requirement according to the significant difference judgment result; and performing early warning processing of track geometry deterioration according to the evaluation result of the data quality level of each train service. The application can effectively evaluate the quality of detection data and improve the accuracy of track geometry deterioration early warning.

Description

Technical Field

[0001] This invention relates to the field of railway engineering technology, and in particular to a method and device for early warning of track geometry deterioration based on a vehicle-mounted track inspection instrument. Background Technology

[0002] This section is intended to provide background or context for the embodiments of the invention set forth in the claims. The description herein is not an admission that it is prior art simply because it is included in this section.

[0003] In the field of railway track engineering, specialized inspection trains equipped with dynamic track geometry detection systems are used to conduct periodic track geometry inspections. Simultaneously, onboard track inspection instruments on operating trains are used to monitor the dynamic changes in track geometry quality in real time and provide timely warnings. Therefore, onboard track inspection instruments are a crucial technical means for ensuring stable track geometry quality and railway transportation safety throughout the inspection cycle. With the increasing number of locomotives equipped with onboard track inspection instruments and the increased frequency of inspections, the amount of deviation data generated has also surged. At the same time, this deviation data faces interference from various sources, including monitoring devices (sensor noise, drift, and even malfunctions under harsh vibration environments), operating environments (such as localized strong winds and high-speed trains passing each other on adjacent tracks), and human factors (such as non-compliant installation and inspection practices). This results in a large amount of invalid or abnormal data in the deviation data, seriously affecting data quality and the accuracy of monitoring and early warning. Summary of the Invention

[0004] This invention provides a method for early warning of track geometry deterioration based on a vehicle-mounted track inspection instrument, which effectively evaluates the quality of inspection data and improves the accuracy of track geometry deterioration early warning. The method includes:

[0005] At the same mileage point, acquire the total data detected by the vehicle-mounted line inspection instrument in multiple trips;

[0006] Density clustering algorithm is used to extract invalid data from the total data detected for each train; the proportion of invalid data in the total data of each train is compared with a preset ratio range, and the data quality level of each train is evaluated based on the comparison results.

[0007] Two sets of data are selected from the data quality levels of multiple train trips: the data with the highest data quality level and the data with the lowest data quality level. The rank-sum test is used to determine the significance level between the two sets of data. Based on the significance level, it is determined whether there is a significant difference between the two sets of data. Based on the result of the significance difference judgment, it is determined that the data quality level classification meets the reasonableness requirements.

[0008] Based on the reasonableness requirements of the data quality level classification, early warning processing of track geometry deterioration is carried out according to the evaluation results of the data quality level of each train.

[0009] This invention also provides a track geometry deterioration early warning device based on a vehicle-mounted track inspection instrument, used to effectively evaluate the quality of inspection data and improve the accuracy of track geometry deterioration early warning. The device includes:

[0010] The data acquisition module is used to acquire the total data detected by the vehicle-mounted line inspection instrument in multiple trips at the same mileage point;

[0011] The data quality level assessment module is used to extract invalid data from the total data detected in each train trip using a density clustering algorithm; compares the proportion of invalid data in the total data of each train trip with a preset ratio range; and assesses the data quality level of each train trip based on the comparison results.

[0012] The data quality classification rationality judgment module is used to select two sets of data from multiple train data quality levels: the data with the highest data quality level and the data with the lowest data quality level. The rank-sum test is used to determine the significance level between the two sets of data. Based on the significance level, it is judged whether there is a significant difference between the two sets of data. Based on the result of the significance difference judgment, it is determined that the data quality level classification meets the rationality requirements.

[0013] The track geometry degradation early warning module is used to perform early warning processing of track geometry degradation based on the evaluation results of the data quality level of each train, provided that the data quality level classification meets the reasonable requirements.

[0014] This invention also provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the above-mentioned track geometry deterioration early warning method based on a vehicle-mounted track inspector.

[0015] This invention also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described track geometry deterioration early warning method based on an on-board track inspection instrument.

[0016] This invention also provides a computer program product, which includes a computer program that, when executed by a processor, implements the above-described track geometry deterioration early warning method based on an on-board track inspector.

[0017] In this embodiment of the invention, the total data detected by the vehicle-mounted track inspection instrument under multiple train trips is obtained at the same mileage point; invalid data is extracted from the total data detected under each train trip using a density clustering algorithm; the proportion of invalid data under each train trip in the total data of that train trip is compared with a preset ratio range, and the data quality level of each train trip is evaluated based on the comparison results; two sets of data are selected from the data quality levels of multiple train trips: the data with the highest data quality level and the data with the lowest data quality level, and the rank-sum test is used to determine the significance level between the two sets of data, and the significance level is used to determine whether there is a significant difference between the two sets of data; based on the significance difference judgment result, it is determined that the data quality level classification meets the reasonableness requirements; based on the data quality level classification meeting the reasonableness requirements, the track geometry deterioration early warning processing is performed according to the evaluation results of the data quality level of each train trip. In the above process, the embodiments of the present invention utilize density clustering algorithm to extract invalid data from the total data detected for each train, which significantly improves the data processing efficiency. The rank sum test is used to determine the significant difference between the data with the highest data quality level and the data with the lowest data quality level, verifying the rationality of the data quality level classification, thereby effectively evaluating the quality of the detection data and improving the accuracy of track geometry deterioration early warning. Attached Figure Description

[0018] 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. In the drawings:

[0019] Figure 1 This is a flowchart of a track geometry deterioration early warning method based on a vehicle-mounted track inspection instrument in an embodiment of the present invention;

[0020] Figure 2 This is a schematic diagram illustrating the identification of invalid horizontal acceleration data points based on a density clustering algorithm in an embodiment of the present invention.

[0021] Figure 3 This is a schematic diagram illustrating the identification of invalid vertical acceleration data points based on a density clustering algorithm in an embodiment of the present invention;

[0022] Figure 4 This is a distribution diagram of the detection data of the vehicle-mounted line inspection instrument for grades A and C in an embodiment of the present invention;

[0023] Figure 5 This is a vertical acceleration trend diagram in an embodiment of the present invention;

[0024] Figure 6This is a horizontal acceleration trend diagram in an embodiment of the present invention;

[0025] Figure 7 This is a schematic diagram of a track geometry deterioration early warning device based on a vehicle-mounted track inspector in an embodiment of the present invention. Detailed Implementation

[0026] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the embodiments of the present invention will be further described in detail below with reference to the accompanying drawings. Here, the illustrative embodiments of the present invention and their descriptions are used to explain the present invention, but are not intended to limit the present invention.

[0027] Figure 1 This is a flowchart of a track geometry deterioration early warning method based on a vehicle-mounted track inspection instrument, as described in an embodiment of the present invention. The method includes:

[0028] Step 101: At the same mileage point, acquire the total data detected by the vehicle-mounted line inspection instrument in multiple trips;

[0029] Step 102: Using density clustering algorithm, extract invalid data from the total data detected for each train trip; compare the proportion of invalid data in the total data of each train trip with a preset ratio range, and evaluate the data quality level of each train trip based on the comparison results.

[0030] Step 103: Select two sets of data from the data quality levels of multiple train trips: the data with the highest data quality level and the data with the lowest data quality level. Use the rank-sum test to determine the significance level between the two sets of data. Based on the significance level, determine whether there is a significant difference between the two sets of data. Based on the result of the significance difference judgment, determine whether the data quality level classification meets the reasonableness requirements.

[0031] Step 104: Based on the reasonableness requirements of the data quality level classification, conduct early warning processing for track geometry deterioration according to the evaluation results of the data quality level of each train.

[0032] Each step is explained in detail below.

[0033] In step 101, at the same mileage point, the total data detected by the vehicle-mounted line inspection instrument under multiple train trips is obtained; in step 102, invalid data is extracted from the total data detected under each train trip using a density clustering algorithm; the proportion of invalid data under each train trip in the total data of that train trip is compared with a preset ratio range, and the data quality level of each train trip is evaluated based on the comparison results.

[0034] In a specific embodiment, a vehicle-mounted track inspection device acquires the total data detected by all vehicle-mounted track inspection devices at the same mileage point on Line A within a day. This total data includes horizontal and vertical acceleration deviation data. During the inspection period, the vehicle-mounted track inspection device, as a dedicated device for dynamically monitoring the geometric smoothness of the track, can dynamically monitor the track condition in real time and promptly detect adverse changes. Combining weather data and high-speed traffic passing data, density clustering algorithms are used to filter invalid data from the lateral and vertical acceleration deviation data monitored by the vehicle-mounted track inspection devices at the same mileage point. The device numbers of vehicle-mounted track inspection devices with abnormal data characteristics (invalid data) at the same mileage point are marked. Other mileage points on Line A are marked using the same method. The density clustering algorithm is defined as follows:

[0035] Using data points detected by a vehicle-mounted line inspection instrument at the same mileage point as centers, circles are drawn with a preset radius. The number of monitoring data points within each circle is the density value of that data point. Through extensive data testing, an optimal density threshold is selected. Points containing fewer monitoring data points within a circle than the optimal density threshold are considered low-density points, and vice versa. If a high-density monitoring data point is within the circle of another high-density monitoring data point, these two data points are connected. This process continues, connecting monitoring data points continuously. If a low-density monitoring data point is also within the circle of a high-density monitoring data point, this low-density data point is also connected to the nearest high-density monitoring data point; this is called a boundary point. All connected monitoring data points form a cluster, and low-density monitoring data points not located at any high-density point are invalid data points. Figure 2 This is a schematic diagram illustrating the identification of invalid horizontal acceleration data points based on a density clustering algorithm in an embodiment of the present invention. Figure 3 This is a schematic diagram illustrating the identification of invalid vertical acceleration data points based on a density clustering algorithm in an embodiment of the present invention; wherein... Figure 2 The orange markers in the data represent invalid data points for the horizontal acceleration at 1140km upstream of a certain route. Figure 3 The orange markers indicate invalid data points for vertical acceleration at 1145 km.

[0036] After identifying invalid data points using density clustering algorithms, these invalid data points are then subjected to in-depth correlation analysis with meteorological and traffic meeting data to eliminate other invalid data that may be caused by meteorological conditions (such as strong winds, rain, freezing, etc.) or traffic meeting situations (such as airflow disturbances when encountering other vehicles at close range). This allows for a more accurate identification of vehicle swaying caused by road problems, thereby improving the accuracy and reliability of data analysis.

[0037] In one embodiment, the track geometry deterioration early warning method based on a vehicle-mounted track inspection instrument further includes:

[0038] The weight values ​​corresponding to each preset data quality assessment index are determined using the analytic hierarchy process.

[0039] The data quality assessment result is determined based on the weight value corresponding to each preset data quality assessment indicator.

[0040] In one embodiment, determining the weight value corresponding to each preset data quality assessment indicator includes:

[0041] Based on the preset data quality assessment indicators, a data quality assessment matrix is ​​established using the analytic hierarchy process (AHP). Based on the data quality assessment matrix, the weight values ​​corresponding to the data quality assessment indicators are determined.

[0042] Based on the weight values ​​corresponding to the data quality assessment indicators and the pre-obtained experience weight values ​​by users, a comprehensive assessment function is established to determine the data quality assessment results.

[0043] In a specific embodiment, the Analytic Hierarchy Process (AHP) is used to comprehensively evaluate the quality of data detected by the vehicle-mounted line inspection instrument. This method combines qualitative and quantitative analysis, is highly operable, simple in principle, and possesses strong rigor, making it widely used in the analysis and decision-making of complex systems. Pre-obtained empirical weight values ​​are achieved using an expert scoring method, where the empirical weight values ​​include weight values ​​corresponding to risk loss indicators and weight values ​​corresponding to risk loss occurrence probability indicators. An example of using the AHP to judge the data quality of a specific vehicle-mounted line is as follows:

[0044] First, a first data quality assessment matrix is ​​established. The first category of data quality indicators includes: data integrity assessment indicators, data accuracy assessment indicators, data consistency assessment indicators, and data timeliness assessment indicators. Specifically, the data integrity assessment indicator ensures that the data collected by the vehicle-mounted track inspection instrument covers the entire mileage of the track, is complete, and without omissions or losses. The data accuracy assessment indicator filters out invalid data and checks its accuracy. The data consistency assessment indicator evaluates whether the distribution of monitoring data collected by the vehicle-mounted track inspection instrument at different times and on the same track mileage is consistent, thus determining the reliability and stability of the data. The data timeliness assessment indicator assesses the update speed and real-time performance of the data detected by the vehicle-mounted track inspection instrument, ensuring that the data can promptly reflect changes in track conditions. Second, the first data quality assessment matrix is ​​normalized. Based on the processed first data quality assessment matrix, the weight vectors corresponding to the first category of data quality indicators are calculated. The maximum eigenvalue is calculated based on the weight vectors, and the data quality is judged based on the maximum eigenvalue.

[0045] In a specific embodiment, the proportion of invalid data in the total data of each train trip is calculated to evaluate the data quality level of each train trip. Invalid data proportion = (Number of invalid data entries / Total number of data entries) × 100%.

[0046] Low percentage of invalid data: Data sets with a low percentage of invalid data (e.g., less than 5%) are generally considered to be of higher quality and are rated A.

[0047] Moderate percentage of invalid data: Invalid data percentages between 5% and 10% may require further data cleaning and validation; data grade B.

[0048] High percentage of invalid data: If the percentage of invalid data exceeds 10%, the dataset quality may be low, requiring detailed data cleaning and quality control processes, and the data grade is C.

[0049] An early warning will be issued if the horizontal acceleration data in the detection data of quality grades A and B is greater than or equal to the preset horizontal acceleration threshold, or the vertical acceleration data is greater than or equal to the preset vertical acceleration threshold.

[0050] In step 103, two sets of data are selected from the data quality levels of multiple train trips: the data with the highest data quality level and the data with the lowest data quality level. The rank-sum test is used to determine the significance level between the two sets of data. Based on the significance level, it is determined whether there is a significant difference between the two sets of data. Based on the result of the significance difference judgment, it is determined that the data quality level classification meets the reasonableness requirements.

[0051] In one embodiment, the rank-sum test is used to determine the significance level between two sets of data, and the determination of whether there is a significant difference between the two sets of data based on the significance level includes:

[0052] When two sets of data do not follow a normal distribution, the Wilcoxon rank-sum test is used to calculate the significance level between the two sets of data.

[0053] Based on the significance level between the two sets of data, determine whether there is a significant difference in the overall distribution of the two sets of data.

[0054] like Figure 4 As shown, Figure 4 This is a distribution diagram of the detection data from vehicle-mounted line inspection instruments of grades A and C in an embodiment of the present invention. In a specific embodiment, the Wilcoxon rank-sum test (also known as the Mann-Whitney U test) is a non-parametric hypothesis testing method specifically used to compare whether there is a significant difference between two independent samples (in this example, vehicle-mounted line inspection instrument detection data of grades A and C). It is suitable for vehicle-mounted line inspection instrument monitoring datasets that are difficult to accurately describe using a normal distribution model. By calculating the difference in the rank sums between the monitoring data samples, the Wilcoxon rank-sum test can effectively assess whether there is a statistically significant difference between the two sets of monitoring data samples, thus providing strong support for understanding the distribution characteristics of monitoring data at different grades.

[0055] The formula for calculating the test statistic U is:

[0056]

[0057] Where: n1 and n2 are data samples of grade A and grade C respectively, and R1 is the rank sum of the data samples corresponding to grade A.

[0058] The calculated test statistic U-value needs to be compared with the standard U-distribution table to determine whether to reject the null hypothesis. A significant difference exists between the medians of the two groups at a significance level of alpha = 0.05, thus validating the rationality of the data quality classification.

[0059] In step 104, based on the reasonableness requirements of the data quality level classification, an early warning process for track geometry deterioration is carried out according to the evaluation results of the data quality level of each train.

[0060] In one embodiment, after performing corresponding early warning processing based on the evaluation results of the data quality level of each train trip, the method further includes:

[0061] Based on the evaluation results of the data quality level for each train trip, data with the same data quality level are obtained for different train trips;

[0062] Trend analysis is performed on data of the same quality level at the same mileage point but different train numbers, and potential track geometric anomalies are identified based on the analysis results.

[0063] In a specific embodiment, trend analysis is performed on horizontal acceleration data and vertical acceleration data of the same quality level at the same mileage point but different vehicle trips to extract trend items within the maintenance cycle, such as... Figure 5 This is a vertical acceleration trend diagram in an embodiment of the present invention; Figure 6 This is a horizontal acceleration trend graph in an embodiment of the present invention. The trend graph visually displays the changing trends of values ​​detected by different levels of onboard track inspection instruments over time, helping to quickly identify potential abnormal fluctuations, periodic changes, or stability issues. Through the trend graph, the effectiveness of maintenance measures can be more accurately assessed, and potential maintenance interventions can be identified and predicted in a timely manner, thereby optimizing maintenance strategies and ensuring the smooth operation of the track system.

[0064] This invention also provides a track geometry deterioration early warning device based on a vehicle-mounted track inspector, as described in the following embodiments. Since the principle by which this device solves the problem is similar to the track geometry deterioration early warning method based on a vehicle-mounted track inspector, the implementation of this device can refer to the implementation of the track geometry deterioration early warning method based on a vehicle-mounted track inspector; repeated details will not be elaborated further.

[0065] Figure 7 This is a schematic diagram of a track geometry deterioration early warning device based on a vehicle-mounted track inspector, as described in an embodiment of the present invention. The device includes:

[0066] The data acquisition module 701 is used to acquire the total data detected by the vehicle-mounted line inspection instrument in multiple trips at the same mileage point;

[0067] The data quality level assessment module 702 is used to extract invalid data from the total data detected in each train trip using a density clustering algorithm; compare the proportion of invalid data in the total data of each train trip with a preset ratio range; and assess the data quality level of each train trip based on the comparison results.

[0068] The data quality classification rationality judgment module 703 is used to select two sets of data from the data quality levels of multiple train trips: the data with the highest data quality level and the data with the lowest data quality level. The rank-sum test is used to determine the significance level between the two sets of data. Based on the significance level, it is judged whether there is a significant difference between the two sets of data. Based on the result of the significance difference judgment, it is determined that the data quality level classification meets the rationality requirements.

[0069] The track geometry degradation early warning module 704 is used to perform early warning processing of track geometry degradation based on the evaluation results of the data quality level of each train, on the basis that the data quality level classification meets the reasonable requirements.

[0070] In one embodiment, a data quality assessment module is further included, specifically for:

[0071] The weight values ​​corresponding to each preset data quality assessment index are determined using the analytic hierarchy process.

[0072] The data quality assessment result is determined based on the weight value corresponding to each preset data quality assessment indicator.

[0073] In one embodiment, the data quality assessment module is specifically used for:

[0074] Based on the preset data quality assessment indicators, a data quality assessment matrix is ​​established using the analytic hierarchy process (AHP). Based on the data quality assessment matrix, the weight values ​​corresponding to the data quality assessment indicators are determined.

[0075] Based on the weight values ​​corresponding to the data quality assessment indicators and the pre-obtained experience weight values ​​by users, a comprehensive assessment function is established to determine the data quality assessment results.

[0076] In one embodiment, the data quality grading rationality judgment module 703 is specifically used for:

[0077] When two sets of data do not follow a normal distribution, the Wilcoxon rank-sum test is used to calculate the significance level between the two sets of data.

[0078] Based on the significance level between the two sets of data, determine whether there is a significant difference in the overall distribution of the two sets of data.

[0079] In one embodiment, a trend analysis module is further included, specifically for:

[0080] Based on the evaluation results of the data quality level for each train trip, data with the same data quality level are obtained for different train trips;

[0081] Trend analysis is performed on data of the same quality level at the same mileage point but different train numbers, and potential track geometric anomalies are identified based on the analysis results.

[0082] This invention also provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the above-mentioned track geometry deterioration early warning method based on a vehicle-mounted track inspector.

[0083] This invention also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described track geometry deterioration early warning method based on an on-board track inspection instrument.

[0084] This invention also provides a computer program product, which includes a computer program that, when executed by a processor, implements the above-described track geometry deterioration early warning method based on an on-board track inspector.

[0085] In this embodiment of the invention, the total data detected by the vehicle-mounted track inspection instrument under multiple train trips is obtained at the same mileage point; invalid data is extracted from the total data detected under each train trip using a density clustering algorithm; the proportion of invalid data under each train trip in the total data of that train trip is compared with a preset ratio range, and the data quality level of each train trip is evaluated based on the comparison results; two sets of data are selected from the data quality levels of multiple train trips: the data with the highest data quality level and the data with the lowest data quality level, and the rank-sum test is used to determine the significance level between the two sets of data, and the significance level is used to determine whether there is a significant difference between the two sets of data; based on the significance difference judgment result, it is determined that the data quality level classification meets the reasonableness requirements; based on the data quality level classification meeting the reasonableness requirements, the track geometry deterioration early warning processing is performed according to the evaluation results of the data quality level of each train trip. In the above process, the embodiments of the present invention utilize density clustering algorithm to extract invalid data from the total data detected for each train, which significantly improves the data processing efficiency. The rank sum test is used to determine the significant difference between the data with the highest data quality level and the data with the lowest data quality level, verifying the rationality of the data quality level classification, thereby effectively evaluating the quality of the detection data and improving the accuracy of track geometry deterioration early warning.

[0086] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, 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.

[0087] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), 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.

[0088] 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.

[0089] 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.

[0090] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for early warning of track geometry deterioration based on a vehicle-mounted track inspection instrument, characterized in that, include: At the same mileage point, acquire the total data detected by the vehicle-mounted line inspection instrument in multiple trips; the total data includes horizontal acceleration and vertical acceleration deviation data. Combining weather data and high-speed oncoming traffic data, density clustering algorithms were used to filter out invalid data from the lateral and vertical acceleration deviation data monitored by the vehicle-mounted line inspection instrument at the same mileage point. The density clustering algorithm includes: drawing circles with a preset radius, each centered on a data point detected by a vehicle-mounted route inspection instrument at the same mileage point; counting the number of monitoring data points within each circle; using this number as the density value of the data point corresponding to the center of the circle; selecting an optimal density threshold; points whose number of monitoring data points within the circle is less than the optimal density threshold are considered low-density points, and vice versa; if a high-density monitoring data point is within the circle of another high-density monitoring data point, these two monitoring data points are connected, and so on, continuously connecting monitoring data points; if a low-density monitoring data point is also within the circle of a high-density monitoring data point, this low-density monitoring data point is also connected to the nearest high-density monitoring data point, called a boundary point; all connected monitoring data points form a cluster, and low-density monitoring data points not in any high-density points are invalid data points; using the density clustering algorithm, invalid data is extracted from the total data detected in each trip; the proportion of invalid data in the total data of each trip is compared with a preset ratio range, and the data quality level of each trip is evaluated based on the comparison result; After identifying invalid data points using density clustering algorithms, the invalid data points are then subjected to in-depth correlation analysis with meteorological data and passing data to eliminate other invalid data caused by meteorological conditions or passing situations. Two sets of data are selected from the data quality levels of multiple train trips: the data with the highest data quality level and the data with the lowest data quality level. The rank-sum test is used to determine the significance level between the two sets of data. Based on the significance level, it is determined whether there is a significant difference between the two sets of data. Based on the result of the significance difference judgment, it is determined that the data quality level classification meets the reasonableness requirements. Based on the reasonableness requirements of the data quality level classification, early warning processing of track geometry deterioration is carried out according to the evaluation results of the data quality level of each train.

2. The method as described in claim 1, characterized in that, Also includes: The weight values ​​corresponding to each preset data quality assessment index are determined using the analytic hierarchy process. The data quality assessment result is determined based on the weight value corresponding to each preset data quality assessment indicator.

3. The method as described in claim 2, characterized in that, The weight values ​​corresponding to each preset data quality assessment indicator were determined using the analytic hierarchy process (AHP), including: Based on the preset data quality assessment indicators, a data quality assessment matrix is ​​established using the analytic hierarchy process (AHP). Based on the data quality assessment matrix, the weight values ​​corresponding to the data quality assessment indicators are determined. Based on the weight values ​​corresponding to the data quality assessment indicators and the pre-obtained experience weight values ​​by users, a comprehensive assessment function is established to determine the data quality assessment results.

4. The method as described in claim 1, characterized in that, The rank-sum test is used to determine the significance level between two sets of data. Based on the significance level, it is determined whether there is a significant difference between the two sets of data, including: When two sets of data do not follow a normal distribution, the Wilcoxon rank-sum test is used to calculate the significance level between the two sets of data. Based on the significance level between the two sets of data, determine whether there is a significant difference in the overall distribution of the two sets of data.

5. The method as described in claim 1, characterized in that, After assessing the data quality level of each train trip and implementing corresponding early warning measures, the process also includes: Based on the evaluation results of the data quality level for each train trip, data with the same data quality level are obtained for different train trips; Trend analysis is performed on data of the same quality level at the same mileage point but different train numbers, and potential track geometric anomalies are identified based on the analysis results.

6. A track geometry deterioration early warning device based on a vehicle-mounted track inspector, characterized in that, include: The data acquisition module is used to acquire the total data detected by the vehicle-mounted line inspection instrument in multiple trips at the same mileage point; the total data includes horizontal acceleration and vertical acceleration deviation data. The data quality level assessment module is used to combine weather data and high-speed oncoming traffic data to filter out invalid data by using density clustering algorithms for the lateral and vertical acceleration deviation data monitored by the vehicle-mounted line inspection instrument at the same mileage point, respectively. The density clustering algorithm includes: drawing circles with a preset radius, each centered on a data point detected by a vehicle-mounted line inspection instrument at the same mileage point; counting the number of monitoring data points within each circle; using the number of monitoring data points within the circle as the density value of the data point corresponding to the center of the circle; selecting an optimal density threshold; points whose number of monitoring data points within the circle is less than the optimal density threshold are considered low-density points, and vice versa; if a high-density monitoring data point is within the circle of another high-density monitoring data point, these two monitoring data points are connected, and so on, with monitoring data points continuously connected; if a low-density monitoring data point is also within the circle of a high-density monitoring data point, this low-density monitoring data point is also connected to the nearest high-density monitoring data point, called a boundary point; all connected monitoring data points form a cluster, and low-density monitoring data points not in any high-density points are invalid data points; The data quality level assessment module is used to extract invalid data from the total data detected in each train trip using a density clustering algorithm; compares the proportion of invalid data in the total data of each train trip with a preset ratio range; and assesses the data quality level of each train trip based on the comparison results. After identifying invalid data points using density clustering algorithms, the invalid data points are then subjected to in-depth correlation analysis with meteorological data and passing data to eliminate other invalid data caused by meteorological conditions or passing situations. The data quality classification rationality judgment module is used to select two sets of data from multiple train data quality levels: the data with the highest data quality level and the data with the lowest data quality level. The rank-sum test is used to determine the significance level between the two sets of data. Based on the significance level, it is judged whether there is a significant difference between the two sets of data. Based on the result of the significance difference judgment, it is determined that the data quality level classification meets the rationality requirements. The track geometry degradation early warning module is used to perform early warning processing of track geometry degradation based on the evaluation results of the data quality level of each train, provided that the data quality level classification meets the reasonable requirements.

7. The apparatus as claimed in claim 6, characterized in that, It also includes a data quality assessment module, specifically used for: The weight values ​​corresponding to each preset data quality assessment index are determined using the analytic hierarchy process. The data quality assessment result is determined based on the weight value corresponding to each preset data quality assessment indicator.

8. The apparatus as claimed in claim 7, characterized in that, The data quality assessment module is specifically used for: Based on the preset data quality assessment indicators, a data quality assessment matrix is ​​established using the analytic hierarchy process (AHP). Based on the data quality assessment matrix, the weight values ​​corresponding to the data quality assessment indicators are determined. Based on the weight values ​​corresponding to the data quality assessment indicators and the pre-obtained experience weight values ​​by users, a comprehensive assessment function is established to determine the data quality assessment results.

9. The apparatus as claimed in claim 6, characterized in that, The data quality grading rationality judgment module is specifically used for: When two sets of data do not follow a normal distribution, the Wilcoxon rank-sum test is used to calculate the significance level between the two sets of data. Based on the significance level between the two sets of data, determine whether there is a significant difference in the overall distribution of the two sets of data.

10. The apparatus as claimed in claim 6, characterized in that, It also includes a trend analysis module, specifically used for: Based on the evaluation results of the data quality level for each train trip, data with the same data quality level are obtained for different train trips; Trend analysis is performed on data of the same quality level at the same mileage point but different train numbers, and potential track geometric anomalies are identified based on the analysis results.

11. A computer 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 computer program, it implements the method of any one of claims 1 to 5.

12. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the method of any one of claims 1 to 5.

13. A computer program product, characterized in that, The computer program product includes a computer program that, when executed by a processor, implements the method of any one of claims 1 to 5.

Images