A wheel-rail test model establishment method, track wear diagnosis method and system

Abstract

The present invention provides a method for establishing a wheel-rail test model and a track wear diagnosis method and system. The establishment method includes: obtaining track attribute data, vehicle motion attribute data, and wheel-rail relationship data of a railway system; modeling the track attribute data to obtain a track data model of the target track; modeling the vehicle motion attribute data to obtain a vehicle dynamics model of a high-speed train; and, based on the wheel-rail relationship data and a pre-built track component library, performing digital twin modeling of the track data model and the vehicle dynamics model to establish a wheel-rail test model. The solution provided by the present invention utilizes a simulation model established based on real track and train information. The track test model can then be used to conduct wheel-rail relationship testing, resulting in higher testing efficiency and a more convenient and safer testing process.

Description

Technical Field

[0001] The present invention relates to the field of simulation testing technology, and in particular to a method for establishing a wheel-rail test model, and a track wear diagnosis method and system. Background Art

[0002] With the continuous advancement of high-speed railway construction, the requirements for the safety and stability of high-speed train operation are becoming increasingly higher. Among them, the wheel-rail relationship is one of the important factors affecting the safety and stability of high-speed train operation.

[0003] In related technologies, two common wheel-rail relationship testing methods are direct contact testing and indirect contact testing. However, both methods require real-world track conditions, resulting in lengthy testing cycles and potential safety risks, making testing inconvenient.

[0004] Therefore, the traditional wheel-rail relationship testing method has the problems of low testing efficiency and lack of convenience and safety. Summary of the Invention

[0005] The present invention provides a method for establishing a wheel-rail test model, a track wear diagnosis method and a system, which are used to solve the defects of traditional wheel-rail relationship testing methods, such as low testing efficiency, inconvenience and lack of safety.

[0006] In a first aspect, the present invention provides a method for establishing a wheel-rail test model, comprising:

[0007] Obtain track attribute data, vehicle motion attribute data, and wheel-rail relationship data of the railway system;

[0008] Modeling the track attribute data to obtain a track data model of the target track;

[0009] Modeling the vehicle motion attribute data to obtain a vehicle dynamics model of the high-speed train;

[0010] Based on the wheel-rail relationship data and a pre-built track component library, digital twin modeling is performed on the track data model and the vehicle dynamics model to establish a wheel-rail test model.

[0011] According to the method for establishing a wheel-rail test model provided by the present invention, after establishing the wheel-rail test model, the method further includes:

[0012] Inputting the test data samples collected in real time into the wheel-rail test model to obtain the wheel-rail relationship prediction result output by the wheel-rail test model;

[0013] The wheel-rail test model is adjusted based on the wheel-rail relationship prediction result and the wheel-rail relationship measurement result in the test data sample.

[0014] According to the method for establishing a wheel-rail test model provided by the present invention, adjusting the wheel-rail test model based on the wheel-rail relationship prediction result and the wheel-rail relationship measurement result in the test data sample includes:

[0015] Comparing the wheel-rail relationship prediction result with the wheel-rail relationship measurement result to determine a first model error value of the wheel-rail test model;

[0016] If the error value of the first model is higher than a preset error threshold, obtaining real-time monitoring data of the current round of testing process;

[0017] determining root causes of errors in the wheel-rail test model based on the measured monitoring data;

[0018] Based on the error root cause, model parameters of the wheel-rail test model are adjusted.

[0019] According to the method for establishing a wheel-rail test model provided by the present invention, after adjusting the model parameters of the wheel-rail test model, the method further includes:

[0020] Based on the test data samples collected in real time, the wheel-rail test model after the model parameters are adjusted is tested again to determine a second model error value;

[0021] Comparing the second model error value with the preset error threshold and the first model error value respectively to obtain a comparison result;

[0022] Based on the comparison results, the model optimization direction is determined.

[0023] According to the method for establishing a wheel-rail test model provided by the present invention, determining the model optimization direction based on the comparison result includes:

[0024] If the comparison result shows that the second model error value is smaller than the first model error value, and the second model error value is greater than the preset error threshold, then it is determined that the model optimization direction is the same as the optimization direction of the previous round;

[0025] If the comparison result shows that the second model error value is greater than the first model error value, and the second model error value is greater than the preset error threshold, then determining that the model optimization direction is opposite to the optimization direction of the previous round;

[0026] If the comparison result is that the second model error value is smaller than the first model error value, and the second model error value is smaller than the preset error threshold, the wheel-rail test model after the model parameters are adjusted is used as the test result.

[0027] According to the method for establishing a wheel-rail test model provided by the present invention, obtaining real-time monitoring data of the current round of testing process includes:

[0028] Obtain the wheel-rail monitoring video collected in real time during the current round of testing;

[0029] Extracting key image points corresponding to wheel-rail contact areas in at least some key video frames of the wheel-rail monitoring video;

[0030] determining the running state and wheel-rail interaction state of the high-speed train based on the key image points;

[0031] The running status and wheel-rail interaction status of the high-speed train are used as real-time monitoring data.

[0032] In a second aspect, the present invention further provides a rail wear diagnosis method, comprising:

[0033] Obtain wheel-rail relationship data between the high-speed train and the target track;

[0034] Inputting the wheel-rail relationship data into a wheel-rail test model to obtain a wheel-rail relationship prediction result output by the wheel-rail test model; wherein the wheel-rail test model is obtained based on any of the above wheel-rail test model establishment methods;

[0035] Based on the wheel-rail relationship prediction result, wear diagnosis is performed on the target track to obtain a wear diagnosis result.

[0036] According to the track wear diagnosis method provided by the present invention, the wheel-rail relationship prediction result includes the wheel-rail interaction force of the current test cycle and the track wear data of the current test cycle;

[0037] The step of performing wear diagnosis on the target track based on the wheel-rail relationship prediction result to obtain a wear diagnosis result specifically includes:

[0038] Comparing the wheel-rail interaction force of the current test cycle with the safety range of the interaction force to obtain a first comparison result;

[0039] Comparing the rail wear data of the current test cycle with the wear safety range to obtain a second comparison result;

[0040] A wear diagnosis result is obtained based on the first comparison result and / or the second comparison result.

[0041] In a third aspect, the present invention further provides a rail wear diagnostic system, comprising:

[0042] Data collection equipment; and

[0043] An electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. The processor is connected to the data acquisition device. When the processor executes the program, the track wear diagnosis method described above is implemented.

[0044] According to the rail wear diagnostic system provided by the present invention, the data acquisition device includes: a sensor group and an image acquisition device;

[0045] The sensor groups are respectively installed on the high-speed train and the target track, and the image acquisition device is installed around the target track;

[0046] The sensor group includes at least one of an accelerometer, a velocity sensor and a force sensor.

[0047] The wheel-rail test model establishment method and track wear diagnosis method and system provided by the present invention model track attribute data to obtain a track data model, and model vehicle motion attribute data to obtain a vehicle dynamics model. Based on wheel-rail relationship data and a pre-built track component library, digital twin modeling of the track data model and vehicle dynamics model is performed to establish a wheel-rail test model. Because the wheel-rail test model is a simulation model built based on real track and train information, the track test model can be used to conduct subsequent wheel-rail relationship testing. Compared to traditional testing solutions, wheel-rail relationship testing can be performed without the need for a real track environment, resulting in higher testing efficiency, a more convenient, and safer testing process. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] In order to more clearly illustrate the technical solutions in the present invention or the prior art, a brief introduction is given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0049] Figure 1 1 is a flow chart of a method for establishing a wheel-rail test model provided by an embodiment of the present invention;

[0050] Figure 2 1 is a flow chart of a rail wear diagnosis method provided by an embodiment of the present invention;

[0051] Figure 3 is a schematic structural diagram of a rail wear diagnostic system provided by an embodiment of the present invention;

[0052] Figure 4 It is a structural diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0053] Embodiments of the present invention will now be described in greater detail with reference to the accompanying drawings. While the accompanying drawings illustrate embodiments of the present invention, it should be understood that the present invention may be implemented in various forms and should not be limited by the embodiments described herein. Rather, these embodiments are provided to make the present invention more thorough and complete, and to fully convey the scope of the present invention to those skilled in the art.

[0054] The terms used in this invention are for the purpose of describing specific embodiments only and are not intended to limit the invention. The singular forms "a," "the," and "the" used in this invention and the appended claims are also intended to include plural forms unless the context clearly indicates otherwise. It should also be understood that the term "and / or" as used herein refers to and includes any or all possible combinations of one or more of the associated listed items.

[0055] It should be understood that although the terms "first", "second", "third", etc. may be used to describe various information in the present invention, such information should not be limited to these terms. These terms are only used to distinguish information of the same type from each other. For example, without departing from the scope of the present invention, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of such features. In the description of the present invention, "plurality" means two or more, unless otherwise clearly and specifically defined.

[0056] First, the technical terms involved in the embodiments of the present invention are explained:

[0057] First, digital twin refers to the technology of modeling various aspects of the physical system through digital means, and simulating and analyzing it in a virtual environment.

[0058] Second, the wheel-rail relationship refers to the interaction between the wheel and the track, including the effects of various forces such as friction, traction, and braking force.

[0059] The following combination Figures 1 to 4 The invention describes a method for establishing a wheel-rail test model, a rail wear diagnosis method, and a detailed system solution.

[0060] See also Figure 1 The method for establishing a wheel-rail test model provided by an embodiment of the present invention specifically includes:

[0061] Step 110: Acquire track attribute data, vehicle motion attribute data, and wheel-rail relationship data of the railway system.

[0062] It can be understood that the track attribute data mainly represents information such as the shape, size and position of the target track in the railway system, the vehicle motion attribute data mainly represents the operating status of the high-speed train, and the wheel-rail relationship data can represent the interaction state of the high-speed train when running at high speed on the target track.

[0063] Step 120: Model the track attribute data to obtain a track data model of the target track.

[0064] In this embodiment, the track data model is a data model established based on track attribute data and capable of simulating the actual track morphology.

[0065] Step 130: Modeling the vehicle motion attribute data to obtain a vehicle dynamics model of the high-speed train.

[0066] In this embodiment, the vehicle dynamics model is a data model established based on vehicle motion attribute data and capable of simulating the actual train running state.

[0067] Step 140: Based on the wheel-rail relationship data and the pre-built track component library, digital twin modeling is performed on the track data model and the vehicle dynamics model to establish a wheel-rail test model.

[0068] It can be understood that the track component library is a data model library containing various track components, which may specifically include track components such as switches, signal machines, and contact lines.

[0069] In this embodiment, the wheel-rail test model can represent the three-dimensional shape of key objects such as tracks, vehicles, and signal systems, and can simulate the interaction between high-speed trains and target tracks.

[0070] This embodiment establishes a wheel-rail test model to test the wheel-rail relationship in a simulation environment without having to test the wheel-rail relationship in a real track environment. This makes the test process more convenient and safer, and the test efficiency is higher.

[0071] In one embodiment, the track attribute data includes: a starting position, an ending position, curvature data, slope data, and altitude data;

[0072] Model the track attribute data to obtain the track data model of the target track, including:

[0073] Determine a track profile of the target track based on the starting position and the ending position;

[0074] Based on the curvature data, slope data and height data, the track contour line is adjusted in shape to obtain the track data model of the target track.

[0075] In this embodiment, the track attribute data uses real track data, including information such as the track's starting and ending positions, curvature data, slope data, and altitude data. During the simulation modeling of the target track, the starting and ending points of the target track in the simulation environment can be determined based on the starting and ending points. The track profile of the target track can be obtained by connecting the starting and ending points. In practical applications, since tracks typically have a certain slope due to differences in the surrounding terrain and are usually curved rather than straight lines, the track profile can be morphologically adjusted using the curvature data, slope data, and altitude data to obtain a track data model of the target track.

[0076] It can be understood that, since the track data model in this embodiment is obtained based on real track data, during the test process, the target track in the railway system can be simulated more accurately and realistically.

[0077] In some embodiments, the vehicle motion attribute data specifically includes position data, speed data, and acceleration data.

[0078] Accordingly, the vehicle motion attribute data is modeled to obtain the vehicle dynamics model of the high-speed train, which specifically includes:

[0079] A dynamic sub-model is established based on the position data, speed data and acceleration data of the high-speed train;

[0080] Based on the operating conditions of high-speed trains, a constraint condition sub-model is established;

[0081] The dynamics sub-model and the constraint condition sub-model are used as the vehicle dynamics model.

[0082] It can be understood that the dynamics sub-model can characterize the motion state of the high-speed train, and the constraint condition sub-model can constrain the motion state of the high-speed train to ensure the safe operation of the high-speed train.

[0083] In this embodiment, during the process of constructing the wheel-rail test model, the wheel-rail test model needs to establish the following logical relationship with the high-speed wheel-rail relationship:

[0084] First, an accurate track data model needs to be established to accurately characterize key information such as the geometry and size of the target track, so as to accurately simulate the movement and force conditions of high-speed trains on the track.

[0085] Second, an accurate vehicle dynamics model needs to be established to simulate the dynamic behavior of high-speed trains running at high speeds.

[0086] Third, a track-vehicle interaction model needs to be established to simulate the generation and impact of the interaction force between the target track and the high-speed train.

[0087] In one embodiment, after establishing the wheel-rail test model, the method further includes:

[0088] Input the test data samples collected in real time into the wheel-rail test model to obtain the wheel-rail relationship prediction results output by the wheel-rail test model;

[0089] The wheel-rail test model is adjusted based on the wheel-rail relationship prediction results and the actual wheel-rail relationship measurement results in the test data sample.

[0090] This embodiment can be understood as a process of verifying and evaluating the wheel-rail test model, that is, a model testing process. After adjustment, a more accurate wheel-rail test model can be obtained.

[0091] In one embodiment, the wheel-rail test model is adjusted based on the wheel-rail relationship prediction results and the wheel-rail relationship measurement results in the test data sample, including:

[0092] In the first step, the wheel-rail relationship prediction results are compared with the wheel-rail relationship measurement results to determine the first model error value of the wheel-rail test model.

[0093] It can be understood that the actual measurement results of the wheel-rail relationship are used as real data, and the predicted results of the wheel-rail relationship are used as predicted data. By comparing the real data with the predicted data, the first model error value of the wheel-rail test model can be determined, and the first model error value can represent the prediction accuracy of the wheel-rail test model.

[0094] In the second step, if the error value of the first model is higher than the preset error threshold, real-time monitoring data of the current round of testing process is obtained.

[0095] It is understandable that if the first model error value is higher than the preset error threshold, it indicates that the prediction accuracy of the current wheel-rail test model is relatively low. At this time, real-time monitoring data of the current round of testing can be obtained. In this embodiment, the real-time monitoring data is the real video data captured by the image acquisition device during the test.

[0096] The third step is to determine the root causes of errors in the wheel-rail test model based on the measured monitoring data.

[0097] In this embodiment, since the measured monitoring data records the actual interaction conditions between the high-speed train and the target track during this round of testing, including real image data of the wheel-rail contact area during the operation of the high-speed train, the root cause of the low prediction accuracy of the wheel-rail test model, that is, the root cause of the error, can be identified based on the measured monitoring data.

[0098] The fourth step is to adjust the model parameters of the wheel-rail test model based on the root cause of the error.

[0099] It is understandable that after the root cause of the error is determined, the root cause of the error can be used as a data basis to guide the parameter adjustment of the wheel-rail test model, thereby improving the adjustment efficiency of the wheel-rail test model.

[0100] In some embodiments, determining the root cause of the error in the wheel-rail test model based on the measured monitoring data includes:

[0101] Determine multiple key performance indicators of the wheel-rail test model and the error distribution data corresponding to each key performance indicator;

[0102] Determine the root cause of the error based on the measured monitoring data and the error distribution data corresponding to each key performance indicator.

[0103] It's understandable that the key performance indicators (KPIs) of a wheel-rail test model are primarily used to characterize its credibility. For example, a multi-level indicator system can be constructed, encompassing the process, stage, activity, element, and feature layers, yielding multiple KPIs. Error distribution data for each KPI can be used to initially analyze which level or levels of the indicator system are deficient. Measured monitoring data can then be used to further identify the specific error causes behind these deficiencies, i.e., the root causes of the errors.

[0104] In some embodiments, based on the root cause of the error, the model parameters of the wheel-rail test model are adjusted, specifically including:

[0105] According to the root cause of the error, at least one target indicator system with deficiencies can be determined, and then the model parameters to be adjusted and the parameter adjustment direction corresponding to each target indicator system can be determined;

[0106] Adjust the corresponding model parameters according to the parameter adjustment direction.

[0107] It should be understood that the above embodiments address the root cause of errors when the root cause is due to model accuracy. In practical applications, if the root cause also includes errors in data acquisition or processing, adjustments to the data acquisition and processing steps are necessary to ensure the accuracy and reliability of the data input to the wheel-rail test model. If the wheel-rail test model still exhibits low accuracy after adjustments to the data acquisition and processing steps, adjustments to the model parameters are then necessary.

[0108] It should be noted that the process of adjusting the model parameters of the wheel-rail test model in this embodiment includes the process of correcting some model parameters and adding or deleting some model parameters.

[0109] In one embodiment, after adjusting the model parameters of the wheel-rail test model, the method may further include:

[0110] Based on the test data samples collected in real time, the wheel-rail test model after the model parameters are adjusted is retested to determine the second model error value;

[0111] Comparing the second model error value with the preset error threshold and the first model error value respectively to obtain a comparison result;

[0112] Based on the comparison results, the model optimization direction is determined.

[0113] It is understood that after the wheel-rail test model is revised, it needs to be verified and tested again to ensure that the revised model has better performance and accuracy. During the secondary testing process, this embodiment not only compares the second model error value with the preset error threshold, but also compares the second model error value with the first model error value. In other words, it also introduces a step of comparing the test results before and after the revision. This comparison can determine a more accurate model optimization direction, thereby improving model optimization efficiency.

[0114] In practical applications, the first model error value and the second model error value can be calculated by obtaining the mean square error or mean absolute error between the wheel-rail relationship prediction result and the wheel-rail relationship measurement result.

[0115] In one embodiment, determining a model optimization direction based on the comparison results includes:

[0116] On the one hand, if the comparison result is that the second model error value is smaller than the first model error value, and the second model error value is greater than the preset error threshold, it is determined that the model optimization direction is the same as the previous round of optimization direction.

[0117] In this case, the second model error value is smaller than the first model error value, indicating that the model performance has been improved to a certain extent, and the model optimization direction of the previous round is correct. However, the second model error value at this time is greater than the preset error threshold, indicating that the model needs further optimization to meet the application conditions. Therefore, it can be determined that the model optimization direction is the same as the optimization direction of the previous round.

[0118] On the other hand, if the comparison result is that the second model error value is greater than the first model error value, and the second model error value is greater than the preset error threshold, it is determined that the model optimization direction is opposite to the optimization direction of the previous round.

[0119] In this case, the second model error value is greater than the first model error value, indicating that the model performance has not been improved and the model optimization direction of the previous round is incorrect. At this time, the second model error value is greater than the preset error threshold, indicating that the model needs further optimization to meet the application conditions. Therefore, at this time, it can be determined that the model optimization direction is opposite to the optimization direction of the previous round in an attempt to improve the model performance.

[0120] On the other hand, if the comparison result shows that the second model error value is smaller than the first model error value, and the second model error value is smaller than the preset error threshold, the wheel-rail test model after the model parameters are adjusted is used as the test result.

[0121] In this case, the second model error value is smaller than the first model error value, indicating that the model performance has been improved to a certain extent, and the model optimization direction of the previous round is correct. In addition, the second model error value is smaller than the preset error threshold at this time, indicating that the optimization end condition is met. At this time, the current round of testing can be ended to obtain the optimized wheel-rail test model, that is, the test result.

[0122] In practical applications, the correction and optimization of the wheel-rail test model is an ongoing process. With the accumulation of test data and the addition of new data, the wheel-rail test model needs to be continuously updated and iterated to maintain its accuracy and reliability.

[0123] In one embodiment, obtaining real-time monitoring data of the current round of testing process includes:

[0124] The first step is to obtain the wheel-rail monitoring video collected in real time during the current round of testing.

[0125] In practical applications, wheel-rail monitoring videos can be captured by image acquisition equipment installed around the track. For example, high-definition cameras can be installed around the track to capture wheel-rail monitoring videos.

[0126] In the second step, key image points corresponding to the wheel-rail contact area in at least some key video frames of the wheel-rail monitoring video are extracted.

[0127] Since wheel-rail monitoring video can be understood as a collection of continuous video frames within a certain period of time, it may contain some video frames that are useless for subsequent data processing, such as video frames of a high-speed train at rest before it starts, or multiple video frames with the same track morphology and vehicle operating status. To ensure subsequent data processing efficiency and accuracy, key video frames can be extracted from the wheel-rail monitoring video. In this embodiment, the key video frames can be a preset number of video frames corresponding to different track morphologies and different vehicle operating states.

[0128] The third step is to determine the running status and wheel-rail interaction status of the high-speed train based on the key image points.

[0129] Key image points can be the image points within the wheel-rail contact area that can represent the interaction between the high-speed train wheels and the target track. By analyzing the position changes of key image points, the operating status of the high-speed train and the wheel-rail interaction status can be determined.

[0130] Specifically, the running status of a high-speed train can be determined by the position change relationship of the wheel image points in different video frames in the key image points. For example, the position change of the wheel image points in two adjacent frames can be calculated respectively, and the interval length between the two adjacent video frames can be determined. The obtained position change is divided by the interval length to obtain the speed of the high-speed train at the interval length. Based on the multiple speeds obtained, the acceleration of the high-speed train and other data that can characterize the running status can be further calculated.

[0131] The fourth step is to use the running status and wheel-rail interaction status of the high-speed train as real-time monitoring data.

[0132] In this embodiment, the real-time monitoring data may specifically include image data corresponding to key video frames and status data such as the running status and wheel-rail interaction status of the high-speed train.

[0133] In practical applications, after obtaining the wheel-rail test model, it can be used in wheel-rail relationship testing scenarios. The output wheel-rail relationship prediction results can help researchers understand the operation and performance of high-speed wheel-rail relationships, and thus optimize the performance and stability of high-speed wheel-rail relationships. For example, by adjusting the parameters in the vehicle dynamics model, the vehicle's operating performance can be improved, and by monitoring track wear, timely measures such as repair and replacement can be taken.

[0134] It should be noted that different modeling methods and algorithms can be used when establishing a wheel-rail test model, such as physical model-based modeling, statistical learning-based modeling, etc. Furthermore, different comparative analysis methods can be used when verifying the accuracy of the wheel-rail test model, such as mean square error (MSE) and mean absolute error (MAE).

[0135] The method for establishing a wheel-rail test model provided in this embodiment has at least the following advantages:

[0136] First, digital twin technology is used to model and simulate the railway system. The established wheel-rail test model can simulate various operating conditions and changes in the wheel-rail relationship, thereby providing accurate data support and reference basis for actual wheel-rail relationship testing and evaluation.

[0137] Second, combining real-time monitoring data to correct and optimize the model can improve test efficiency and accuracy, reduce test costs and risks, and has strong scalability and adaptability. At the same time, it can also improve test efficiency and accuracy, reduce test costs and risks.

[0138] Based on the same general inventive concept, the present invention also protects a rail wear diagnosis method and system. The rail wear diagnosis method and system provided by the present invention are described below. The rail wear diagnosis method and system described below and the method for establishing the wheel-rail test model described above can be referenced to each other.

[0139] Figure 2 The track wear diagnosis method provided by an embodiment of the present invention is exemplarily shown, including:

[0140] Step 210: Acquire wheel-rail relationship data between the high-speed train and the target track.

[0141] In this embodiment, the wheel-rail relationship data between the high-speed train and the target track may specifically include vehicle operation status data of the current test cycle, wheel-rail interaction force data of the previous test cycle, and track wear data of the previous test cycle.

[0142] Step 220: Input the wheel-rail relationship data into the wheel-rail test model to obtain the wheel-rail relationship prediction result output by the wheel-rail test model; wherein the wheel-rail test model is obtained based on the wheel-rail test model establishment method provided in the above embodiments.

[0143] In this embodiment, the wheel-rail relationship prediction result may specifically include the wheel-rail interaction force of the current test cycle and the track wear data of the current test cycle.

[0144] It can be understood that since this embodiment uses the wheel-rail test model to perform wheel-rail relationship testing, the wheel-rail relationship between the high-speed train and the target track can be predicted more conveniently and accurately, thereby achieving reliable testing of the wheel-rail relationship.

[0145] Step 230: Based on the wheel-rail relationship prediction result, wear diagnosis is performed on the target track to obtain a wear diagnosis result.

[0146] In this embodiment, the wheel-rail interaction force and the track wear condition can be comprehensively analyzed to achieve wear diagnosis of the target track.

[0147] In one embodiment, based on the wheel-rail relationship prediction result, wear diagnosis is performed on the target track to obtain a wear diagnosis result, which specifically includes:

[0148] Comparing the wheel-rail interaction force of the current test cycle with the safety range of the interaction force to obtain a first comparison result;

[0149] Comparing the rail wear data of the current test cycle with the wear safety range to obtain a second comparison result;

[0150] A wear diagnosis result is obtained based on the first comparison result and / or the second comparison result.

[0151] In a specific implementation, obtaining a wear diagnosis result based on the first comparison result and the second comparison result includes:

[0152] On the one hand, if the first comparison result is that the wheel-rail interaction force of the current test cycle is within the force safety range, and the rail wear data of the current test cycle is within the wear safety range, then the wear diagnosis result is normal wear.

[0153] On the other hand, if the first comparison result shows that the wheel-rail force of the current test cycle exceeds the force safety range, and the rail wear data of the current test cycle exceeds the wear safety range, the wear level is determined based on the wheel-rail force out-of-range value and the rail wear data out-of-range value, and the wear level is used as the wear diagnosis result.

[0154] Specifically, multiple wheel-rail force out-of-range levels and rail wear data out-of-range levels can be set, a first level corresponding to the wheel-rail force out-of-range value is determined from the wheel-rail force out-range levels, and a second level corresponding to the rail wear data out-of-range value is determined from the rail wear data out-range levels, and the wear level is determined based on the first level and the second level.

[0155] In practical applications, the wear level can be determined in a variety of ways. For example, the higher of the first and second levels can be used as the wear level. Another example is to average the first and second levels and use the average as the wear level. Another example is to take a weighted sum or weighted average of the first and second levels and use the weighted result as the wear level.

[0156] In some embodiments, after obtaining the wear diagnosis result, the rail wear diagnosis method may further include:

[0157] A safety assessment is performed on the target track according to the wear diagnosis result to obtain a safety assessment result.

[0158] In this embodiment, when the wear diagnosis result includes the wear level, multiple risk levels can be divided according to the wear level. For example, if the wear level is level one to level three, the corresponding risk level is level one risk, and if the wear level is level four to level six, the corresponding risk level is level two risk.

[0159] In this way, the risk level corresponding to the wear level can be further determined according to the wear level, and a safety assessment can be performed according to the risk level to obtain a safety assessment result.

[0160] In some embodiments, the security assessment result can be determined based on the risk level. Different risk levels correspond to different security assessment results, and the security assessment result can be low security risk, medium security risk, or high security risk. For example, a risk level of three or less can correspond to a low security risk; a risk level of four to five can correspond to a medium security risk; and a risk level of six or above can correspond to a high security risk.

[0161] This embodiment uses the above-mentioned safety assessment solution to reduce safety issues caused by track wear, thereby improving the operational safety and stability of the railway system.

[0162] Figure 3 Schematic diagram of the structure of the track wear diagnosis system provided in this embodiment.

[0163] See also Figure 3 The rail wear diagnosis system provided by the embodiment of the present invention specifically includes:

[0164] Data collection device 300; and

[0165] The electronic device 400 includes a memory, a processor, and a computer program stored in the memory and executable on the processor. The processor is connected to a data acquisition device. When the processor executes the program, the rail wear diagnosis method provided in the above embodiments is implemented.

[0166] In one embodiment, see Figure 3 The data acquisition device 300 includes: a sensor group 310 and an image acquisition device 320 .

[0167] The sensor group 310 is installed on the high-speed train and the target track respectively, and the image acquisition device 320 is installed around the target track.

[0168] The sensor group 310 may specifically include at least one of an accelerometer, a velocity sensor, and a force sensor.

[0169] In this embodiment, the acceleration level and speed sensor can collect the running speed and acceleration of the high-speed train, the force sensor can collect the wheel-rail force, and the image acquisition device can be a high-definition camera that can shoot wheel-rail monitoring video.

[0170] Figure 4 It is a structural diagram of an electronic device provided by an embodiment of the present invention.

[0171] like Figure 4As shown, the electronic device may include: a processor 410, a communication interface 420, a memory 430, and a communication bus 440, wherein the processor 410, the communication interface 420, and the memory 430 communicate with each other via the communication bus 440. The processor 410 may call the logic instructions in the memory 430 to execute the rail wear diagnosis method provided in the above embodiments.

[0172] In addition, the logic instructions in the above-mentioned memory 430 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when sold or used as an independent product. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0173] On the other hand, the present invention also provides a computer program product, which includes a computer program. The computer program can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the track wear diagnosis method provided in the above embodiments.

[0174] On the other hand, the present invention further provides a non-transitory computer-readable storage medium having a computer program stored thereon, which is implemented when the computer program is executed by a processor to perform the rail wear diagnosis method provided by the above embodiments.

[0175] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, i.e., they may be located in one location or distributed across multiple network units. Some or all of the modules may be selected based on actual needs to achieve the objectives of the present embodiment. Persons of ordinary skill in the art will be able to understand and implement the present invention without inventive effort.

[0176] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, or of course, by hardware. Based on this understanding, the essence of the above technical solution or the part that contributes to the existing technology can be embodied in the form of a software product. The computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, an optical disk, etc., and includes a number of instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or certain parts of the embodiments.

[0177] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A method for establishing a wheel-rail test model, characterized in that: include: Obtain track attribute data, vehicle motion attribute data, and wheel-rail relationship data of the railway system; Modeling the track attribute data to obtain a track data model of the target track; Modeling the vehicle motion attribute data to obtain a vehicle dynamics model of the high-speed train; Based on the wheel-rail relationship data and a pre-built track component library, digital twin modeling is performed on the track data model and the vehicle dynamics model to establish a wheel-rail test model; After establishing the wheel-rail test model, the method further includes: Inputting the test data samples collected in real time into the wheel-rail test model to obtain the wheel-rail relationship prediction result output by the wheel-rail test model; The wheel-rail relationship prediction result is compared with the wheel-rail relationship actual measurement result to determine a first model error value of the wheel-rail test model; if the first model error value is higher than a preset error threshold, the wheel-rail monitoring video collected in real time during the current round of testing is obtained; key image points corresponding to the wheel-rail contact area in at least some key video frames in the wheel-rail monitoring video are extracted, wherein the key video frames are a preset number of video frame information corresponding to different track shapes and different vehicle operating conditions, and the key image points are some image points in the wheel-rail contact area that characterize the interaction state between the wheels of the high-speed train and the target track; based on the key image points, the operating state of the high-speed train and the wheel-rail interaction state are determined, wherein the operating state of the high-speed train is determined by the position change relationship of the wheel image points in different video frames in the key image points; the operating state of the high-speed train and the wheel-rail interaction state are used as real-time monitoring data; Determine multiple key performance indicators of the wheel-rail test model and the error distribution data corresponding to each key performance indicator; Determine the root cause of the error based on measured monitoring data and the error distribution data corresponding to each key performance indicator. The key performance indicators of the wheel-rail test model are used to characterize the credibility of the wheel-rail test model and involve an indicator system at multiple levels: process, stage, activity, element, and feature. If the root cause of the error includes the accuracy of the model itself, determine at least one target indicator system with deficiencies based on the root cause of the error, determine the model parameters to be adjusted and the parameter adjustment direction corresponding to each target indicator system; and adjust the corresponding model parameters according to the parameter adjustment direction. If the root cause of the error also includes errors in the data acquisition or processing process, the data acquisition and processing links should be adjusted first. If the wheel-rail test model still has accuracy problems after the data acquisition and processing links are adjusted, the model parameters of the wheel-rail test model should be adjusted.

2. The method for establishing a wheel-rail test model according to claim 1, characterized in that: After adjusting the model parameters of the wheel-rail test model, the method further includes: Based on the test data samples collected in real time, the wheel-rail test model after the model parameters are adjusted is tested again to determine a second model error value; Comparing the second model error value with the preset error threshold and the first model error value respectively to obtain a comparison result; Based on the comparison results, the model optimization direction is determined.

3. The method for establishing a wheel-rail test model according to claim 2, characterized in that: Determining the model optimization direction based on the comparison results includes: If the comparison result shows that the second model error value is smaller than the first model error value, and the second model error value is greater than the preset error threshold, then it is determined that the model optimization direction is the same as the optimization direction of the previous round; If the comparison result shows that the second model error value is greater than the first model error value, and the second model error value is greater than the preset error threshold, then determining that the model optimization direction is opposite to the optimization direction of the previous round; If the comparison result is that the second model error value is smaller than the first model error value, and the second model error value is smaller than the preset error threshold, the wheel-rail test model after the model parameters are adjusted is used as the test result.

4. A rail wear diagnosis method, characterized in that: include: Obtain wheel-rail relationship data between the high-speed train and the target track; Inputting the wheel-rail relationship data into a wheel-rail test model to obtain a wheel-rail relationship prediction result output by the wheel-rail test model; wherein the wheel-rail test model is obtained based on the wheel-rail test model establishment method according to any one of claims 1 to 3; Based on the wheel-rail relationship prediction result, wear diagnosis is performed on the target track to obtain a wear diagnosis result.

5. The rail wear diagnosis method according to claim 4, characterized in that: The wheel-rail relationship prediction result includes the wheel-rail force of the current test cycle and the track wear data of the current test cycle; The step of performing wear diagnosis on the target track based on the wheel-rail relationship prediction result to obtain a wear diagnosis result specifically includes: Comparing the wheel-rail interaction force of the current test cycle with the safety range of the interaction force to obtain a first comparison result; Comparing the rail wear data of the current test cycle with the wear safety range to obtain a second comparison result; A wear diagnosis result is obtained based on the first comparison result and / or the second comparison result.

6. A rail wear diagnostic system, characterized in that: include: Data acquisition equipment; as well as An electronic device comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor is connected to the data acquisition device and implements the rail wear diagnosis method according to claim 4 or 5 when the processor executes the program.

7. The rail wear diagnostic system according to claim 6, characterized in that: The data acquisition device includes: a sensor group and an image acquisition device; The sensor groups are respectively installed on the high-speed train and the target track, and the image acquisition device is installed around the target track; The sensor group includes at least one of an accelerometer, a velocity sensor and a force sensor.

Images