Wheel-rail contact load decoupling prediction method, electronic equipment and storage medium

Abstract

The invention provides a wheel-rail contact load decoupling prediction method, electronic equipment and a storage medium, and the prediction method comprises the steps: building a wheel-rail contact three-dimensional simulation model, and determining a strain collection radius and a collection point on a wheel; setting boundary conditions, running speed and traction torque of the wheel-rail contact three-dimensional simulation model; solving and calculating the wheel-rail contact three-dimensional simulation model to obtain a strain value and a contact force of each strain acquisition point of the wheel; inputting the obtained strain value and contact force of each strain acquisition point of the wheel into a neural network model for training, and establishing a wheel-rail contact neural network model of a strain-contact force relationship; unknown strain data are input into the wheel-rail contact neural network model of the strain-contact force relation, and wheel-rail contact force is obtained; according to the method, the wheel-rail contact force can be accurately identified and predicted.

Description

Technical Field

[0001] The present invention relates to the technical field of rail vehicles, and in particular to a wheel-rail contact load decoupling prediction method, electronic equipment and storage medium. Background Art

[0002] At present, the research on the identification of train wheel-rail contact force mostly adopts experimental test methods, such as the on-board test method based on strain gauges, the wheel-rail contact force test method based on optics and acoustics, and the wheel-rail contact force test method integrating multiple technologies. The above experimental test methods have high requirements on equipment integration and acquisition accuracy, and are limited by the limited experimental measurement points, making it difficult to achieve large-scale high-precision acquisition and real-time processing of a large amount of direct data. In addition, under the current conditions of mature artificial intelligence theory and technology, there has been no research on the combination of heavy-duty train wheel-rail contact force and such technology to achieve accurate identification of contact force.

[0003] Therefore, there is an urgent need for a wheel-rail contact load decoupling prediction method, electronic equipment and storage medium, which, combined with artificial intelligence methods, can accurately identify and predict the wheel-rail contact force. Summary of the invention

[0004] The purpose of the present invention is to provide a wheel-rail contact load decoupling prediction method, an electronic device and a storage medium, aiming to solve the technical problems of traditional train wheel-rail contact force testing methods, such as high requirements on equipment integration and acquisition accuracy, and difficulty in obtaining a large amount of high-precision data.

[0005] To achieve the above objectives, in a first aspect, the present invention provides a wheel-rail contact load decoupling prediction method, comprising:

[0006] S1. Establish a three-dimensional wheel-rail contact simulation model and determine the strain collection radius and collection points on the wheel;

[0007] S2, setting the boundary conditions, running speed and traction torque of the wheel-rail contact three-dimensional simulation model;

[0008] S3, solving and calculating the wheel-rail contact three-dimensional simulation model to obtain the strain value and contact force of each strain collection point of the wheel;

[0009] S4, taking a large number of strain values ​​and contact forces at each strain collection point of the wheel as sample data, inputting them into a neural network model for training, and establishing a wheel-rail contact neural network model of strain-contact force relationship;

[0010] S5. Input the unknown strain data into the wheel-rail contact neural network model of the strain-contact force relationship to obtain the corresponding wheel-rail contact force.

[0011] As a further improvement of the above solution, in step S1, the steps and methods for determining the strain collection radius and collection points on the wheel-rail are as follows:

[0012] S11, first obtain the strain values ​​of several nodes of the wheel-rail within the specified radius range under the specified lateral load and the specified vertical load, and perform comparative analysis to obtain a relationship diagram between the outer strain of the wheel spoke plate and the radius of the acquisition node;

[0013] S12, based on the relationship diagram between the outer side strain of the wheel spoke and the acquisition node radius, determining the non-contact strain acquisition radius C2 for lateral force decoupling identification and the non-contact strain acquisition radius C4 for vertical force decoupling identification, C2<C4;

[0014] S13. Based on the symmetrical four-group bridge layout method, at least eight strain collection points are selected on the circle corresponding to each radius.

[0015] As a further improvement of the above solution, in step S12, based on the obtained strain collection radii C2 and C4, at least three non-contact strain collection radii C1, C3 and C5 are selected, and C1< C2< C3< C4< C5.

[0016] As a further improvement of the above scheme, in step S2, a fixed constraint is applied to the two tracks, an angular velocity within a preset range is applied to the wheel axle, and a traction torque within a preset range is applied to the wheel axle;

[0017] The angular velocity within the preset range is stored as an angular velocity load array N, and the traction torque within the preset range is stored as a traction torque load array M.

[0018] As a further improvement of the above scheme, in step S2, when setting the boundary conditions, running speed and traction torque of the wheel-rail contact three-dimensional simulation model,

[0019] Batch modify model boundary conditions, running speed and traction torque and generate corresponding inp files.

[0020] As a further improvement of the above scheme, in step S3, when solving and calculating the wheel-rail contact three-dimensional simulation model,

[0021] Apply each angular velocity value in the angular velocity load array, traverse all traction torque values ​​in the traction torque load array, and obtain the strain value and contact force value of each strain collection point of the wheel under the corresponding load;

[0022] Apply each traction torque value in the traction torque load array, traverse all angular velocity values ​​in the angular velocity load array, and obtain the strain value and contact force value of each strain collection point of the wheel under the corresponding load;

[0023] A total of N*M groups of strain values ​​and contact force values ​​at each strain collection point of the wheel are obtained.

[0024] As a further improvement of the above solution, the strain value includes strain data in three directions, namely, normal strain in the x-direction, normal strain in the y-direction, and xy shear strain.

[0025] As a further improvement of the above scheme, in step S3, after all load conditions are calculated, the results of all odb result files are extracted and stored using a script for subsequent result analysis and neural network training.

[0026] As a further improvement of the above scheme, in step S4, the coefficient of determination (Coefficient of Determination, R 2 ) to evaluate the model effect of the wheel-rail contact neural network model of the strain-contact force relationship constructed;

[0027] The calculation formula of the determination coefficient is shown as follows:

[0028]

[0029] Where N is the number of samples, is the predicted value of the output, is the true value of the output, and are the true value and predicted value of the i-th sample respectively.

[0030] As a further improvement of the above solution, in step S4, the neural network model includes but is not limited to a feedforward neural network model.

[0031] In a second aspect, the present invention further provides a device, comprising a memory for storing computer program instructions and a processor for executing computer program instructions, wherein when the computer program instructions are executed by the processor, the device is triggered to execute a wheel-rail contact load decoupling prediction method provided in the first aspect.

[0032] In a third aspect, the present invention further provides a storage medium having a computer program stored thereon, wherein when the program is executed, the device where the storage medium is located is controlled to execute a wheel-rail contact load decoupling prediction method provided in the first aspect.

[0033] Since the present invention adopts the above technical solution, the beneficial effects of this application are:

[0034] The present invention provides a wheel-rail contact load decoupling prediction method, which first establishes a three-dimensional wheel-rail contact simulation model, and determines the strain collection radius and collection points on the wheel; then sets the boundary conditions, running speed and traction torque of the three-dimensional wheel-rail contact simulation model, and then solves based on the boundary conditions, different running speeds and different traction forces to obtain a large amount of strain values ​​and contact force data of each strain collection point of the wheel, and uses these data as sample data to input into a neural network model for training, establishes a wheel-rail contact neural network model of strain-contact force relationship, and finally inputs unknown strain data into the wheel-rail contact neural network model of strain-contact force relationship. Network model, wheel-rail contact force is obtained; in the present invention, a large amount of sample data is obtained through simulation, and then these sample data are input into the neural network model for training, so as to obtain the wheel-rail contact neural network model of strain-contact force relationship, and then the wheel-rail contact force can be predicted by the wheel-rail contact neural network model of strain-contact force relationship, that is, the present invention realizes the decoupling and identification of wheel-rail contact force through neural network; the present invention overcomes the difficulties that the traditional experimental test method has high requirements for equipment integration and acquisition accuracy, and is limited by the limited experimental measurement points, and it is difficult to realize large-scale high-precision acquisition and real-time processing of a large amount of direct data. The wheel-rail contact force prediction method proposed by the present invention greatly reduces the requirements for manpower, equipment and environment, and has significant advantages in the extraction and processing of required data, thereby improving data utilization and work efficiency. At the same time, combined with artificial intelligence technology, the prediction of the contact force of the contact points on both sides of the wheel and rail is realized based on the training of the neural network model, and the prediction accuracy is high. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the structures shown in these drawings without paying any creative work.

[0036] Figure 1 A schematic diagram of a flow chart of a wheel-rail contact load decoupling prediction method disclosed in the present invention;

[0037] Figure 2 The present invention is a schematic diagram of the three-dimensional modeling of rails determined according to the standard and the wheel-rail CAD model established, wherein FIG2(a) shows the dimensions of the RF2 wheelset and the JM3 tread; Figure 2 (b) 75kg rail dimensions; Figure 2 (c) The wheel-rail contact model established in Solidworks based on standard dimensions;

[0038] Figure 3The curve of the strain variation of the outer side of the wheel spoke with the node radius and the schematic diagram of the sampling radius and node distribution are shown in Figure 2. Figure 3 (a) is the curve of the strain on the outer side of the wheel spoke changing with the node radius; Figure 3 (b) is a schematic diagram of wheel sampling radius and node distribution;

[0039] Figure 4 Neural network training set, validation set and total set result curves for the lateral force and vertical force at the contact points of the left and right wheels on both sides.

[0040] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with the implementation methods and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0041] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0042] It should be noted that all directional indications (such as up, down, etc.) in the embodiments of the present invention are only used to explain the relative position relationship, movement status, etc. between the components under a certain specific posture (as shown in the accompanying drawings). If the specific posture changes, the directional indication will also change accordingly.

[0043] In addition, in the present invention, the descriptions such as "first", "second", etc. are only used for descriptive purposes and cannot be understood as indicating or implying their relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" or "second" may explicitly or implicitly include at least one of the features.

[0044] Furthermore, the technical solutions between the various embodiments of the present invention may be combined with each other, but this must be based on the fact that they can be implemented by ordinary technicians in the field. When the combination of technical solutions is mutually contradictory or cannot be implemented, it should be deemed that such combination of technical solutions does not exist and is not within the scope of protection required by the present invention.

[0045] Example 1

[0046] See also Figure 1-Figure 4 The present invention provides a wheel-rail contact load decoupling prediction method, comprising:

[0047] S1. Establish a three-dimensional wheel-rail contact simulation model and determine the strain collection radius and collection points on the wheel. Specifically, when establishing the three-dimensional wheel-rail contact simulation model, the wheel-rail multi-point contact system of the heavy-duty train in the existing line is adopted, and a wheel-rail model containing a complete wheelset and rails is established according to the literature reports and national standards, such as Figure 2 As shown. Specifically, according to TB1010-2016 "Railway Vehicle Wheelsets and Bearing Types and Basic Dimensions", the present invention selects the RF2 wheelset consisting of HFS wheels and RF2 axles, wherein the wheel tread model is the JM3 tread dedicated to heavy-duty trains, and the size is determined according to the national standard TB / T449-2003 "Road Vehicle Wheel Flange Tread Shape". Further, according to GB / T2582-2021 "Hot-rolled Rails for Railways", the rail model is determined to be a 75kg rail for heavy-duty trains. Finally, after determining the models and specifications of each component of the model in the numerical simulation, the 3D modeling software Solidworks is used to establish and assemble the model. The specific dimensional parameters in the model are given by the standard documents, and the detailed dimensional data of the axles and wheels not marked in some figures are given in Tables 1 and 2.

[0048]

[0049]

[0050] S2, setting the boundary conditions, running speed and traction torque of the wheel-rail contact three-dimensional simulation model;

[0051] In this embodiment, for the boundary conditions, the value of the vertical load is fixed and added to the central coupling reference points on both sides of the wheel axle, and its value is 250.0kN, the axle weight of the existing heavy-duty railway line. It can be seen from the "Heavy-Duty Railway Design Code" that the maximum traction force of a heavy-duty train is Fmax=570.0kN. The interval from 120.0kN to the maximum traction force is equally divided into six equal parts, and the traction torque is applied through the coupling reference point of the wheel axle center. The converted traction torque interval is listed in Table 3. The train running speed interval is divided into five equal parts with the existing line straight-line running speed of 80.0km / h as the upper limit and 16.0km / h as the lower limit. It is converted into the wheel rotation angular velocity and applied to the wheel through the central coupling reference point. The specific values ​​are listed in Table 4. A total of 30 groups of load conditions are formed by changing the traction force and the driving speed.

[0052]

[0053]

[0054] S3, solving and calculating the wheel-rail contact three-dimensional simulation model to obtain the strain value and contact force of each strain collection point of the wheel;

[0055] S4, taking a large number of strain values ​​and contact forces at each strain collection point of the wheel as sample data, inputting them into a neural network model for training, and establishing a wheel-rail contact neural network model of strain-contact force relationship;

[0056] S5, inputting the unknown strain data into the wheel-rail contact neural network model of the strain-contact force relationship to obtain the corresponding wheel-rail contact force;

[0057] In the present invention, a large amount of sample data is obtained through simulation, and then these sample data are input into the neural network model for training, so as to obtain the wheel-rail contact neural network model of strain-contact force relationship, and then the wheel-rail contact force can be predicted by the wheel-rail contact neural network model of strain-contact force relationship, that is, the present invention realizes the decoupling and identification of wheel-rail contact force through neural network; the present invention overcomes the difficulties of traditional experimental testing methods in terms of high requirements for equipment integration and acquisition accuracy, and is limited by the limited experimental measurement points, making it difficult to achieve large-scale high-precision acquisition and real-time processing of a large amount of direct data. The wheel-rail contact force prediction method proposed by the present invention greatly reduces the requirements for manpower, equipment and environment, and has significant advantages in the extraction and processing of required data, thereby improving data utilization and work efficiency. At the same time, combined with artificial intelligence technology, the prediction of the contact force of the contact points on both sides of the wheel and rail is realized based on the training of the neural network model, and the prediction accuracy is high.

[0058] As a preferred embodiment, considering the complex distribution of strain conditions at different positions on the wheel circumference, it is very important to select appropriate strain collection points. The present invention determines a reasonable strain collection radius and collection points through numerical simulation calculations; specifically, in step S1, the steps and methods for determining the strain collection radius and collection points on the wheel rail are as follows:

[0059] S11. First, obtain the strain values ​​of several nodes of the wheel-rail within the specified radius range under the specified lateral load and the specified vertical load, and perform comparative analysis to obtain the relationship diagram between the strain of the outer side of the wheel spoke and the radius of the acquisition node; specifically, considering that under normal circumstances, the strain value of the radial node element where the external load loading point is located on the same radius is larger than the strain value corresponding to other node elements. The present invention selects the nodes on the radial direction where the external load loading point is located, with a radius of 155.0mm-407.5mm, and numerically calculates their strain distribution under a lateral load of 10.0kN and a vertical load of 10.0kN, respectively, and performs comparative analysis to obtain the relationship between the strain of the outer side of the wheel spoke and the radius of the acquisition node as shown in the figure. Figure 3 (a)

[0060] S12, based on the relationship diagram between the outer side strain of the wheel spoke and the acquisition node radius, determining the non-contact strain acquisition radius C2 for lateral force decoupling identification and the non-contact strain acquisition radius C4 for vertical force decoupling identification, C2<C4;

[0061] On the basis of the obtained strain collection radii C2 and C4, at least three non-contact strain collection radii C1, C3 and C5 are selected, and C1< C2< C3< C4< C5;

[0062] S13, based on the symmetrical four-group bridge method, at least eight strain collection points are selected on each circle corresponding to the radius;

[0063] Specifically, in this embodiment, the principle for selecting the best strain collection point on the wheel is: the self-disturbance is as large as possible and the crosstalk is as small as possible. Figure 3 From the results in (a), it can be seen that among the positions on the wheelset surface where strain can be collected, the node strain at a radius of 297.0 mm is most affected by the lateral force, and this can be used as the non-contact strain collection radius for lateral force decoupling identification. Similarly, the strain of the node at a radius of 324.0 mm is most affected by the vertical force, so this radius can be used as the non-contact strain collection radius for vertical force decoupling identification. Furthermore, considering the optimization of the strain data sample, the present invention additionally selects three radii outside the optimal strain collection radius for strain data collection: ultimately, strain collection points are arranged on a total of five radii with radii of 284.0 mm, 297.0 mm, 310.0 mm, 324.0 mm and 337.0 mm. Regarding the selection of point locations on the collection radius, the present invention adopts a better point distribution method: a symmetrical four-group bridge method, that is, the angles of the strain gauges are (0.0°, 36.0°, 60.0°, 96.0°), and considering the symmetry, at least eight strain collection points are selected on each radius; considering that numerical simulation is a virtual experimental test method, it is more convenient than experimental testing in terms of point selection and data extraction, and also considering that in order to expand the data sample to make the subsequent neural network training more sufficient, for each data collection radius, the strain collection points of the present invention select a total of 30 uniform collection points including the above-mentioned angles. Therefore, finally, strain data in three directions of 150 collection points are extracted from each wheel: E11, E22, and E12, representing the normal strain in the x-direction and y-direction and the xy shear strain, respectively, for subsequent neural network training, and the strain collection points are distributed as follows: Figure 3 (b) as shown.

[0064] As a preferred embodiment, in step S2, fixed constraints are applied to the two tracks, an angular velocity within a preset range is applied to the wheel axle, and a traction torque within a preset range is applied to the wheel axle;

[0065] The angular velocity within the preset range is stored as an angular velocity load array N, and the traction torque within the preset range is stored as a traction torque load array M;

[0066] Specifically, in this embodiment, the angular velocity load array N includes five groups of angular velocity values, which are 3.0 rad / s, 9.3 rad / s, 15.7 rad / s, 22.0 rad / s and 28.3 rad / s respectively; the traction torque load array M includes six groups of traction torque values, which are 9.7E+6 N·mm, 1.2E+7 N·mm, 1.5E+7 N·mm, 1.7E+7 N·mm, 1.9E+7 N·mm and 2.2E+7 N·mm respectively; in the present invention, the above-mentioned different angular velocity values ​​and different traction torque values ​​are stored in corresponding arrays, and when loading, the three-dimensional model is loaded in batches, so that the calculation and solution of the model is convenient and fast; specifically, when setting the boundary conditions, running speed and traction torque of the wheel-rail contact three-dimensional simulation model, the model boundary conditions, running speed and traction torque are modified in batches and the corresponding inp files are generated; then the program is used to submit, calculate and monitor the inp files. After all load conditions are calculated, the script is used to extract and store the results of all odb result files for subsequent result analysis and neural network training. In the entire calculation and solution process, the program script is used to load files and batch process result files, which can reduce manual intervention and easily obtain a large amount of sample data, saving time and effort compared with traditional experimental measurements.

[0067] As a preferred embodiment, in step S3, when solving and calculating the wheel-rail contact three-dimensional simulation model,

[0068] Apply each angular velocity value in the angular velocity load array, traverse all traction torque values ​​in the traction torque load array, and obtain the strain value and contact force value of each strain collection point of the wheel under the corresponding load;

[0069] Apply each traction torque value in the traction torque load array, traverse all angular velocity values ​​in the angular velocity load array, and obtain the strain value and contact force value of each strain collection point of the wheel under the corresponding load;

[0070] A total of N*M groups of strain values ​​and contact force values ​​of each strain collection point of the wheel are obtained;

[0071] In this embodiment, the angular velocity load array N includes five groups of angular velocity values, and the traction torque load array M includes six groups of traction torque values. A total of 30 groups of strain values ​​and contact force values ​​of each strain collection point of each wheel can be obtained at each collection point. Three-directional strain data of 150 collection points are extracted from each wheel, and finally 13,500 sample data can be obtained for each wheel. A large amount of sample data can be easily and efficiently obtained through the data simulation method of the present invention, which effectively overcomes the high requirements of traditional experimental testing methods on equipment integration and collection accuracy, and is limited by the limited experimental measurement points.

[0072] As a preferred embodiment, in step S4, the coefficient of determination (Coefficient of Determination, R 2 ) to evaluate the model effect of the wheel-rail contact neural network model of the strain-contact force relationship constructed;

[0073] The calculation formula of the determination coefficient is shown as follows:

[0074]

[0075] Where N is the number of samples, is the predicted value of the output, is the true value of the output, and are the true value and predicted value of the i-th sample respectively;

[0076] For details, see Figure 4 ,in Figure 4 (a) Fitting curves of the training set, validation set and total set for the lateral force of the left wheel contact point; Figure 4 (b) is the training fitting curve for the lateral force at the right wheel contact point; Figure 4 (c) is the training fitting curve for the vertical force of the left wheel contact point; Figure 4 (d) is the training fitting curve for the vertical force of the right wheel contact point; the regression value results of each total set show R 2 The values ​​are all greater than 0.92, indicating that the neural network prediction results of the wheel-rail contact neural network model of the strain-contact force relationship can achieve the expected effect.

[0077] As a preferred embodiment, in step S4, the neural network model includes but is not limited to a feedforward neural network model; in the present application, a feedforward neural network model is used, including an input layer, a hidden layer and an output layer. In a feedforward neural network, data is transferred from the input layer to the hidden layer, and then transferred to the output layer after being processed by the hidden layer, forming a unidirectional information flow. This structure enables the feedforward neural network to handle various complex nonlinear problems and learn the mapping relationship between input data and output targets through training.

[0078] Embodiment 2:

[0079] The present invention also provides a device, comprising a memory for storing computer program instructions and a processor for executing the computer program instructions, wherein when the computer program instructions are executed by the processor, the device is triggered to execute some or all of the steps in Embodiment 1;

[0080] The processor may include one or more processing units, for example, the processor may include an application processor (AP), a modem processor, a graphics processor (GPU), an image signal processor (ISP), a controller, a video codec, a digital signal processor (DSP), a baseband processor, and / or a neural-network processing unit (NPU), etc. Different processing units may be independent devices or integrated into one or more processors.

[0081] The controller can be the nerve center and command center of the electronic device. The controller can generate operation control signals according to the instruction operation code and timing signal to complete the control of fetching and executing instructions.

[0082] A memory may also be provided in the processor for storing instructions and data. In some embodiments, the memory in the processor is a cache memory. The memory may store instructions or data that the processor has just used or is cyclically used. If the processor needs to use the instruction or data again, it may be directly called from the memory. This avoids repeated access, reduces the waiting time of the processor, and thus improves the efficiency of the system.

[0083] Embodiment 3:

[0084] The present invention also provides a storage medium having a computer program stored thereon, wherein when the program is executed, the device where the storage medium is located is controlled to execute part or all of the steps in Example 1.

[0085] The storage medium may include a high-speed RAM memory, and may also include a nonvolatile memory (nonvolatile memory), such as at least one disk memory. It is understandable that the storage medium can be a random access memory (Random Access Memory, RAM), a disk, a hard disk, a solid state disk (solid state disk, SSD) or a nonvolatile memory, etc., which can store program codes and other machine-readable media.

[0086] Those skilled in the art will appreciate that the embodiments of the present invention may be provided as methods or storage media. Therefore, the embodiments of the present invention may take the form of complete hardware embodiments, complete software embodiments, or embodiments combining software and hardware. Moreover, the embodiments of the present invention may take the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0087] The above are only preferred embodiments of the present invention, and are not intended to limit the patent scope of the present invention. All equivalent structural changes made using the contents of the present invention's specification and drawings, or directly / indirectly applied in other related technical fields, are included in the patent protection scope of the present invention.

Claims

1. A wheel-rail contact load decoupling prediction method, characterized in that: include: S1. Establish a three-dimensional wheel-rail contact simulation model and determine the strain collection radius and collection points on the wheel; S2, setting the boundary conditions, running speed and traction torque of the wheel-rail contact three-dimensional simulation model; S3, solving and calculating the wheel-rail contact three-dimensional simulation model to obtain the strain value and contact force of each strain collection point of the wheel; S4, taking a large number of strain values ​​and contact forces at each strain collection point of the wheel as sample data, inputting them into a neural network model for training, and establishing a wheel-rail contact neural network model of strain-contact force relationship; S5. Input the unknown strain data into the wheel-rail contact neural network model of the strain-contact force relationship to obtain the corresponding wheel-rail contact force.

2. A wheel-rail contact load decoupling prediction method according to claim 1, characterized in that: In step S1, the steps and methods for determining the strain collection radius and collection points on the wheel-rail are as follows: S11, first obtain the strain values ​​of several nodes of the wheel-rail within the specified radius range under the specified lateral load and the specified vertical load, and perform comparative analysis to obtain a relationship diagram between the outer strain of the wheel spoke plate and the radius of the acquisition node; S12, based on the relationship diagram between the outer side strain of the wheel spoke and the acquisition node radius, determining the non-contact strain acquisition radius C2 for lateral force decoupling identification and the non-contact strain acquisition radius C4 for vertical force decoupling identification, C2<C4; S13. Based on the symmetrical four-group bridge layout method, at least eight strain collection points are selected on the circle corresponding to each radius.

3. A wheel-rail contact load decoupling prediction method according to claim 2, characterized in that: In step S12, based on the obtained strain collection radii C2 and C4, at least three non-contact strain collection radii C1, C3 and C5 are selected, and C1< C2< C3< C4< C5.

4. A wheel-rail contact load decoupling prediction method according to any one of claims 1 to 3, characterized in that: In step S2, a fixed constraint is applied to the two tracks, an angular velocity within a preset range is applied to the wheel axle, and a traction torque within a preset range is applied to the wheel axle; The angular velocity within the preset range is stored as an angular velocity load array N, and the traction torque within the preset range is stored as a traction torque load array M.

5. A wheel-rail contact load decoupling prediction method according to claim 4, characterized in that: In step S2, when setting the boundary conditions, running speed and traction torque of the wheel-rail contact three-dimensional simulation model, Batch modify model boundary conditions, running speed and traction torque and generate corresponding inp files.

6. A wheel-rail contact load decoupling prediction method according to claim 4, characterized in that: In step S3, when solving and calculating the wheel-rail contact three-dimensional simulation model, Apply each angular velocity value in the angular velocity load array, traverse all traction torque values ​​in the traction torque load array, and obtain the strain value and contact force value of each strain collection point of the wheel under the corresponding load; Apply each traction torque value in the traction torque load array, traverse all angular velocity values ​​in the angular velocity load array, and obtain the strain value and contact force value of each strain collection point of the wheel under the corresponding load; A total of N*M groups of strain values ​​and contact force values ​​at each strain collection point of the wheel are obtained.

7. A wheel-rail contact load decoupling prediction method according to any one of claims 1 to 3, characterized in that: The strain value includes strain data in three directions, namely, normal strain in the x-direction, normal strain in the y-direction, and xy shear strain.

8. A wheel-rail contact load decoupling prediction method according to any one of claims 1 to 3, characterized in that: In step S3, after all load conditions are calculated, the results of all odb result files are extracted and stored using a script for subsequent result analysis and neural network training.

9. A device, characterized in that: The invention comprises a memory for storing computer program instructions and a processor for executing the computer program instructions, wherein when the computer program instructions are executed by the processor, the device is triggered to execute a wheel-rail contact load decoupling prediction method according to any one of claims 1 to 8.

10. A storage medium, characterized in that: A computer program is stored thereon, wherein when the computer program is running, the device where the storage medium is located is controlled to execute a wheel-rail contact load decoupling prediction method as claimed in any one of claims 1 to 8.

Images