Miniaturized high-precision measurement method and device for bleeding locking of continuously welded rail

Abstract

The invention provides a seamless track blow-off locking miniaturized high-precision measurement method and device, and the method comprises the steps: collecting the rail seam opening degree and steel rail temperature data in real time, and constructing a state input vector; calculating the strain and equivalent stress of the steel rail based on the state input vector, and introducing a structural position sensitive correction term to improve the accuracy of stress estimation; the locking state in the rail seam diffusion process is identified through the lightweight gated loop network model by utilizing the time sequence data of the strain and the corrected stress, and a state classification label is output; and according to the state classification label and the corrected stress value, generating a locking operation suggestion, and providing a visual prompt for construction personnel through terminal equipment. According to the method, a stress estimation method is combined with a time sequence analysis algorithm, so that state judgment is converted from a static index to a dynamic trend, and the locking decision-making efficiency and reliability in the construction process are remarkably improved.

Description

TECHNICAL FIELD

[0001] The present application belongs to the field of data measurement, and particularly relates to a seamless track diffusion locking miniaturized high-precision measurement method and device. BACKGROUND

[0002] As a key structure form in modern railway transportation, the seamless track has been widely applied in high-speed railway and heavy-load railway lines due to its excellent running stability and track continuity. However, the quality of the diffusion and locking links in the laying and maintenance process of the seamless track directly relates to the structural safety and service life of the track in the later period. The so-called "diffusion" refers to the process of releasing the temperature stress of the steel rail in the track laying process, and the "locking" refers to fixing the released steel rail in the appropriate state on the sleeper to ensure that the structure is not damaged due to thermal expansion and contraction in the running process. The traditional diffusion and locking operation relies on manual experience and limited instruments such as temperature sensors and stress meters to determine whether the diffusion is sufficient and the locking is safe. This method not only has a large subjective nature, but also has a large volume and is difficult to deploy, which cannot meet the current efficient, automatic and precise construction requirements. More prominent problems are that the existing method usually uses a static value judgment method, ignores the dynamic change law in the track joint diffusion process, and often relies on only a single physical quantity for discrimination, which is difficult to effectively reflect the overall situation of the track joint stress evolution. In fact, the factors affecting the track joint locking behavior include temperature gradient, strain change rate, tension response lag and other time-varying variables. The existing system lacks dynamic modeling capability for these continuous states, resulting in a lagging and unstable judgment result, which easily causes mislocking, wrong timing locking and other field risks. In addition, in the engineering field, the system deployment needs to consider space limitations, energy consumption control and operation simplicity, and the existing equipment is difficult to achieve unity among miniaturization, high integration and high reliability. SUMMARY

[0003] The purpose of the present application is to design a seamless track diffusion locking miniaturized high-precision measurement method and device, which can continuously track the evolution state of the diffusion process while maintaining the simplicity of deployment and provide timely and clear executable locking operation recommendations.

[0004] In order to achieve the above purpose, in the first aspect of the present application, a seamless track diffusion locking miniaturized high-precision measurement method is provided, which comprises: Collecting track joint opening and rail temperature data, and constructing a state input vector, wherein the track joint opening is measured by a laser ranging module in cooperation with an image auxiliary recognition component, and the measurement point is ensured to be located at the center of the track joint through an image calibration mechanism, and the rail temperature is collected by an infrared thermopile temperature array to collect the surface temperature of the rail; Based on the state input vector, the average strain of the rail is calculated according to the rail joint opening and the design length, the initial equivalent stress is calculated by a thermal-mechanical coupling model combined with the rail temperature, the elastic modulus and the thermal expansion coefficient of the material, and the initial equivalent stress is corrected by a perturbation function according to the structural position sensitivity to obtain the corrected equivalent stress; The history data of the strain and the corrected equivalent stress in a continuous time period are used to construct a time series, and a dynamic trend analysis is performed by a gated recurrent network model, which introduces a thermal-mechanical time consistency constraint and a tension stable region slope constraint in the training stage to improve the generalization ability to the non-ideal diffusion process, and finally outputs a classification label; According to the classification label and the corrected equivalent stress, it is judged whether the stress is in a preset safe tension interval, and an operation instruction of immediate locking, suggestion waiting or prohibition locking is generated combined with the state label, and real-time visualization and wireless prompt are provided to the construction personnel through the man-machine interaction module integrated in the miniaturized terminal device.

[0005] Further, the image calibration mechanism judges the laser landing position by visual assistance and automatically adjusts the direction of the distance measuring module when detecting deviation.

[0006] Further, at least two samplings are performed during the collection process of the rail joint opening, and the average value is taken as the final measurement result.

[0007] Further, the perturbation function is a function of the normalized position of the rail joint section, which has the maximum correction amplitude at the end of the rail joint area and no correction at the center of the rail joint.

[0008] Further, the thermal-mechanical coupling model constructs the equivalent stress by the following way: the rail strain is calculated based on the proportional relationship between the rail joint opening and the design length of the rail; the rail strain, the real-time collected rail temperature, and the elastic modulus and thermal expansion coefficient of the rail material changing with temperature are taken as input parameters; the initial equivalent stress of the rail is calculated by coupling the thermal elastic effect of the strain and the thermal expansion effect caused by temperature.

[0009] Further, the gated recurrent network model adopts a multi-layer GRU structure, the input is the time series of strain and corrected stress, and the output is the state classification label.

[0010] Further, the thermal-mechanical time consistency constraint is used to maintain the coordination of stress change and strain change, and the tension stable region slope constraint is used to suppress the interference of short-term tension fluctuation on the judgment.

[0011] Further, the classification label includes three types of non-locking, lockable and state abnormality.

[0012] Further, the man-machine interaction module provides at least one of visual prompt and wireless broadcast prompt.

[0013] A second aspect of the invention provides a miniaturized, high-precision measurement device for seamless line venting and locking, the device comprising: The data acquisition module is used to collect rail gap opening and rail temperature data and construct a state input vector. The rail gap opening is measured by a laser ranging module in conjunction with an image-assisted recognition component, and the image calibration mechanism ensures that the measuring point is located at the center of the rail gap. The rail temperature is collected by an infrared thermopile temperature array to collect the surface temperature of the rail. The data processing module is used to calculate the average strain of the rail based on the state input vector, according to the rail gap opening and design length, and calculate the initial equivalent stress through a thermo-mechanical coupling model by combining the rail temperature, material elastic modulus and thermal expansion coefficient, and perform structural position sensitive correction on the initial equivalent stress through a perturbation function to obtain the corrected equivalent stress. The state recognition module is used to construct a time series using historical data of strain and corrected equivalent stress over a continuous time period. Dynamic trend analysis is performed through a gated recurrent network model. The gated recurrent network model introduces thermodynamic time series consistency constraints and tension stability zone slope constraints during the training phase to improve the generalization ability for non-ideal release processes. Finally, it outputs classification labels. The decision output module is used to determine whether the stress is within the preset safe tension range based on the classification label and the corrected equivalent stress, and to generate operation instructions such as immediate locking, suggested waiting, or prohibited locking by combining the status label. It provides real-time visualization and wireless prompts to construction personnel through the human-computer interaction module integrated into a miniaturized terminal device.

[0014] The beneficial technical effects of the present invention are at least as follows: To address the aforementioned problems, this invention provides a miniaturized, high-precision measurement method and device for seamless track release and locking. It can construct a dynamic state sequence using two key variables—rail gap opening and rail temperature—without relying on stress gauges or multidimensional coupled models, thereby deriving strain, tension evolution trends, and locking windows. Ben's invention combines a stress estimation method based on a thermo-mechanical model correction with a time series analysis algorithm, transforming state judgment from static indicators to dynamic trends. Furthermore, it introduces multiple regularization constraints highly coupled with on-site engineering conditions into the model structure to enhance the system's explanatory power and stability for real rail gap behavior. Based on this, the invention also proposes a two-factor judgment mechanism for tension and trend, jointly analyzing the time series model output with the current stress state to generate three distinct types of construction instructions for on-site prompts and control. The overall system has a compact structure, simple variables, and traceable and clear judgment logic, significantly improving the efficiency and reliability of locking decisions during construction. It possesses good engineering applicability and reproducibility, representing a technologically innovative solution with complete closed-loop characteristics for seamless track release control problems in practical construction scenarios. Attached Figure Description

[0015] The present invention will be further described with reference to the accompanying drawings, but the embodiments in the drawings do not constitute any limitation on the present invention. For those skilled in the art, other drawings can be obtained based on the following drawings without creative effort.

[0016] Figure 1 This is a flowchart of the miniaturized, high-precision measurement method for seamless line release locking according to the present invention.

[0017] Figure 2 This is a frame diagram of the miniaturized, high-precision measurement device for seamless line release and locking according to the present invention. Detailed Implementation

[0018] Embodiments of the present invention are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and should not be construed as limiting the present invention.

[0019] In one or more embodiments, such as Figure 1 As shown, a miniaturized, high-precision measurement method for seamless line release locking is disclosed, the method comprising the following: S1: Collect rail gap opening and rail temperature data to construct a state input vector. The rail gap opening is measured by a laser ranging module in conjunction with an image-assisted recognition component, and the image calibration mechanism ensures that the measuring point is located at the center of the rail gap. The rail temperature is collected by an infrared thermopile temperature array to collect the surface temperature of the rail. Specifically, this step is used to collect the core input data required to determine whether the seamless track meets the locking conditions. The specific goal is to acquire the mechanical opening status of the rail joint and rail temperature changes in real time during rail laying to support subsequent stress state calculations and trend identification. Considering the complex construction site environment and limited working windows, this step specifically designed a simplified data acquisition method based on two types of highly sensitive physical quantities. This ensures that the measurement system is small in size, has a fast response time, and strong anti-interference capabilities, guaranteeing data quality and timeliness in subsequent processing stages from the source.

[0020] Furthermore, the rail gap opening Data acquisition is achieved through a combination of a laser ranging module and an image-assisted recognition component. Structurally, the laser ranging module integrates an infrared laser diode and a reflective ranging chip, positioned directly above the rail gap with the emission direction perpendicular to the gap plane. The detection point is located at the centerline between the two rails. To ensure ranging stability and effective landing point control, a 5-megapixel miniature camera is integrated next to the laser ranging module to assist in determining whether the laser point is centered in the gap. If a deviation occurs (e.g., due to construction shaking, wind pressure interference, etc.), the image module outputs a calibration signal, which adjusts the ranging module's direction via a servo motor, refocusing it on the gap's center point. For example, in a certain test section, if the laser ranging module initially deviates to the left rail, and the image module detects that the laser point is not in the gap's center area, the control module is triggered to fine-tune the positioning angle by approximately 2.5° until the laser point returns to the center.

[0021] The sampling frequency is set to once per second, and three consecutive samples are taken during each ranging process. The average value is taken as the rail gap opening at the current moment. ; in, Indicates at time The The laser ranging value is valid after the landing point is confirmed by image analysis. This value is obtained through a laser ranging module, and the unit is a length scalar, with measurement accuracy controlled at the millimeter level.

[0022] rail surface temperature Data acquisition is accomplished using a thermopile infrared temperature array. This temperature module is fixedly installed 15cm from the lower edge of the rail gap, facing the rail web to avoid direct interference from wind and sunlight. The sensor's acquisition area covers approximately 10cm x 10cm, and the average temperature of this area is calculated to represent the average thermal state of the rail section. The temperature sampling frequency is consistent with the rail gap opening sampling frequency to ensure alignment in the time dimension. They share the same timestamp. To improve the representativeness of the measured data for the overall thermal state of the rails, point-to-point thermal imager comparison experiments were conducted on different rail types in different seasons during the sensor calibration phase, confirming that the temperature difference between this arrangement and the actual rail temperature does not exceed 0.6.

[0023] All data acquisition and preliminary processing are performed on the local microcontroller, which outputs a state vector for each sampling period. and mark the sampling time. The data is stored in an on-chip buffer in time-series format for subsequent stress estimation and trend modeling.

[0024] The output of this step is the state input vector. ,in Indicates at time The mechanical opening at the midpoint of the rail gap is measured by a laser ranging module and after image calibration. This represents the average surface temperature of the rail at the corresponding moment, which is obtained by measuring the rail web using a thermopile infrared sensor.

[0025] S2: Based on the state input vector, calculate the average strain of the rail according to the rail gap opening and the design length, and calculate the initial equivalent stress through a thermo-mechanical coupling model by combining the rail temperature, material elastic modulus and thermal expansion coefficient, and perform structural position sensitive correction on the initial equivalent stress through a perturbation function to obtain the corrected equivalent stress. Specifically, this step involves the state input vector obtained in step one. Based on this, a strain-stress estimation mechanism for determining the locking state of seamless tracks is established. The goal is to derive the equivalent strain of the rail in real time, without the need for on-site deployment of strain gauges or force sensors, solely based on the rail gap opening and rail temperature. With stress Unlike traditional static tension discrimination methods, this step focuses on the thermodynamic characteristics of the rail gap release process and proposes a "geometric displacement-temperature offset coupled stress model". In addition, for the "temperature lag-force lag" characteristics in the continuous locking section of ballastless track, a position-sensitive tension fine-tuning correction term is introduced to make the derivation results more physically reasonable and predictively stable in the actual rail gap release section.

[0026] In this step, the rail gap structure is first simplified mechanically modeled, approximating the rail segment anchored at both ends and flared at the center as a thermally expanding rod structure with a uniform cross-section. This is done within the standard rail gap length. Within this range, a uniformly distributed elongation approximation is adopted. Based on the geometrically defined linear strain relationship, we can obtain: ; in, This represents the average strain generated in the rail within the locking section at the current moment. The track gap opening is measured by laser ranging. The design length of the rail gap is derived from the construction design drawings and is typically a standard segment length such as 12.5m or 25m. This derivation ignores the non-axial components caused by rail wobble or lateral offset, retaining only the contribution of the dominant tension, thus ensuring a simple formula and sparse variables.

[0027] After obtaining the strain, it is necessary to further consider the changes in the material's elastic modulus caused by temperature variations and the stress hysteresis behavior. In the traditional thermo-mechanical coupling model, the equivalent stress of the rail... This can be given by the following classic expression: ; in, For rail materials at temperature The elastic modulus under the given conditions is loaded offline onto the device using a temperature-modulus interpolation table. The coefficient of thermal expansion of the material. The design lock temperature is based on the definition of the regional rail temperature statistical average.

[0028] To overcome the problem of tension misjudgment caused by the lag in actual temperature response during seamless track construction, this invention further proposes a structural position-sensitive tension correction mechanism. Considering that in continuously locked sections, the region near the rail joint end exhibits different stress response characteristics than the middle section due to construction disturbances and track laying sequence, especially during high-temperature release where "displacement arrives before tension arrives," this step introduces a stress expression relative to the rail joint section into the traditional stress expression. (Normalization to) Related perturbation functions This results in the following modified stress expression: ; in, This represents the corrected equivalent stress; This is the tension adjustment factor, set based on engineering experience, for example, taking... ; Let be the perturbation function, defined as: ,when The correction range is greatest when the value is close to 0 (near the rail gap end); ,when That is, the center of the segment should not be modified.

[0029] This correction ensures that the tension value output by the system is higher during the initial stages of construction or when there are drastic environmental changes. It more closely reflects the actual evolution of rail gap tension, solving the problem that traditional models, which are based solely on thermo-mechanical calculations, cannot reflect the physical phenomenon of "stress hysteresis." It is especially suitable for high-temperature environments or multi-segment continuous laying operations.

[0030] In an actual track-laying test, when the rail gap was in (Near the locking end) , At that time, the result calculated by this modified model The difference between the value and the manually statically locked baseline is only 2.3%, which is far better than the 9.6% under the uncorrected model, showing good engineering matching and prediction stability.

[0031] S3: Construct a time series using historical data of strain and corrected equivalent stress over a continuous time period, and perform dynamic trend analysis using a gated recurrent network model. The gated recurrent network model introduces thermodynamic time series consistency constraints and tension stability zone slope constraints during the training phase to improve the generalization ability for non-ideal release processes, and finally outputs classification labels. Specifically, the goal of this step is to obtain the strain derived in step two. With corrected stress Based on this, a locking state identification mechanism suitable for the dynamic process of rail gap release is constructed. Since the release behavior is not completed instantaneously, but is a dynamic process under the influence of multiple factors such as rail temperature fluctuation, foundation disturbance, and rail gap non-uniformity over a period of time, this step extends the state variables into a time series structure, adopts the sliding window sequence modeling method, and combines it with the temperature-stress coupling constraint mechanism designed specifically for the rail gap release process to propose a tension-trend collaborative judgment scheme that can be deployed on the construction site.

[0032] This step involves constructing a continuous discrimination model with a fixed window length. Constructing a historical state sequence ,in Generally, a value of 30 to 50 is used to cover a 2 to 3 minute window of flare behavior.

[0033] First, a lightweight gated recurrent network (GRU) is adopted as the core structure in the modeling architecture, balancing timeliness and deployment efficiency. This network consists of three layers: Input layer: Accepts input sequences The input dimension is 2 at each time step; The intermediate layer consists of two stacked GRU unit groups, each containing 32 hidden state units, using the tanh activation function to capture the nonlinear evolution characteristics of the track gap state. Output layer: A single fully connected classification node outputs the state classification labels. .

[0034] Due to the physical lag characteristic of "strain responding first, stress responding later" during the gap release process, a thermo-mechanical timing consistency constraint term is introduced as a regularization in the model training to suppress overfitting of the model to short-period tension fluctuations. This regularization term takes the following form: ; in, , , which represents the rate of change of the current state variable; The response scaling factor is typically obtained through fitting prior data and has values ​​such as 0.8 or 1.2. This regularization term is used to limit the model output from dependence on high-frequency, small-amplitude fluctuations, emphasizing the thermo-strain co-current trend of the exothermic process. This design is one of the core innovations of the patent, with a clear physical background, high engineering feasibility, and can be directly deployed on edge devices to enhance the model's generalization ability to complex operating conditions.

[0035] Furthermore, to further improve the model's discriminative stability when approaching the lockable boundary state, this step introduces a slope constraint term in the tension stability region to impose gradient constraints on the model output within the structural boundary: ; in, , representing the acceleration term of the tension sequence; The maximum allowable fluctuation gradient threshold (e.g., 0.01) is used to prevent the judgment result from being misled by short-term drastic tension fluctuations, thus causing the locking decision to be made prematurely or delayed.

[0036] The final model outputs classification labels during the inference phase. Defined as: The current state is in an unstable discharge and cannot be locked. The current status indicates that the window has been locked and is open for access. If the status is abnormal or the trend is reversed, it is recommended to pause and prepare for locking down.

[0037] The output is generated by the final fully connected layer based on the inference results of the hidden state, and is combined with a regularization term applied during the model training phase to enhance the actual interpretability and reliability of the output.

[0038] S4: Based on the classification label and the corrected equivalent stress, determine whether the stress is within the preset safe tension range, and generate operation instructions such as immediately locking, suggesting waiting, or prohibiting locking by combining the status label. Provide real-time visualization and wireless prompts to construction personnel through the human-computer interaction module integrated into the miniaturized terminal device.

[0039] Specifically, this step is based on the state classification labels output in step three. The corrected stress value obtained in step two This generates locking operation suggestions suitable for the construction site. These suggestions, through matching analysis with the track tension design safety range and joint confirmation of the release trend, are translated into control instructions understandable to three types of personnel, driving the terminal prompt interface or on-site personnel to perform the track joint locking operation.

[0040] To determine whether the locking operation can be implemented, this step designs a tension-trend dual verification mechanism to ensure that both the trend of release is completed and the absolute tension is within a safe range. The safe range for construction tension is defined as follows: The value ranges from 180 MPa to 240 MPa. This range is determined by the rail type, fastener design, and rail temperature zone, and is preset in the system.

[0041] First, determine whether the stress value falls within this range by constructing a judgment function: ; Then, with status labels Union, used to generate lock instruction codes : ; in: Indicates "lock immediately"; This indicates that "the trend has been completed but the tension has not yet been reached; it is recommended to wait or make minor adjustments." This indicates "lockdown prohibited".

[0042] Specifically, in the test section in East China, the system determined... This indicates that the release is complete, and at the same time, the following measurements were taken: MPa, while the range is 180~240 MPa, then Final output The on-site terminal will display a flashing yellow message indicating "Tension critical, delayed locking recommended".

[0043] Instruction encoding The system will trigger the corresponding display module (such as RGB lighting or an OLED screen) and Bluetooth broadcast on the terminal device, allowing construction personnel to immediately determine whether the rail gap status can be locked or adjusted. The system automatically reassesses every 30 seconds until... Alternatively, manual confirmation can be performed to lock the device.

[0044] In one or more embodiments, such as Figure 2 As shown, a miniaturized, high-precision measuring device for seamless line release locking is disclosed, the device comprising: The data acquisition module is used to collect rail gap opening and rail temperature data and construct a state input vector. The rail gap opening is measured by a laser ranging module in conjunction with an image-assisted recognition component, and the image calibration mechanism ensures that the measuring point is located at the center of the rail gap. The rail temperature is collected by an infrared thermopile temperature array to collect the surface temperature of the rail. The data processing module is used to calculate the average strain of the rail based on the state input vector, according to the rail gap opening and design length, and calculate the initial equivalent stress through a thermo-mechanical coupling model by combining the rail temperature, material elastic modulus and thermal expansion coefficient, and perform structural position sensitive correction on the initial equivalent stress through a perturbation function to obtain the corrected equivalent stress. The state recognition module is used to construct a time series using historical data of strain and corrected equivalent stress over a continuous time period. Dynamic trend analysis is performed through a gated recurrent network model. The gated recurrent network model introduces thermodynamic time series consistency constraints and tension stability zone slope constraints during the training phase to improve the generalization ability for non-ideal release processes. Finally, it outputs classification labels. The decision output module is used to determine whether the stress is within the preset safe tension range based on the classification label and the corrected equivalent stress, and to generate operation instructions such as immediate locking, suggested waiting, or prohibited locking by combining the status label. It provides real-time visualization and wireless prompts to construction personnel through the human-computer interaction module integrated into a miniaturized terminal device.

[0045] It is worth noting that the specific workflow of the miniaturized high-precision measurement system for seamless line venting and locking provided in the embodiments of the present invention is the same as that of the miniaturized high-precision measurement method for seamless line venting and locking described in the above embodiments, and will not be repeated here.

[0046] This invention also provides a miniaturized high-precision measurement device for seamless line venting and locking, including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements the steps described in the embodiments of the miniaturized high-precision measurement method for seamless line venting and locking described above, for example... Figure 1 The steps S1 to S4 described above; or, when the processor executes the computer program, it implements the functions of each module in the above system embodiments.

[0047] For example, the computer program may be divided into one or more modules, which are stored in the memory and executed by the processor to complete the present invention. The one or more modules may be a series of computer program instruction segments capable of performing specific functions, which describe the execution process of the computer program in the seamless line release locking miniaturized high-precision measurement device.

[0048] The miniaturized high-precision measurement device for seamless line scattering and locking can be a computing device such as a desktop computer, laptop, handheld computer, or cloud server. This device may include, but is not limited to, a processor and memory. Those skilled in the art will understand that the device may also include input / output devices, network access devices, buses, etc.

[0049] The processor can be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor. The processor is the control center of the seamless line scattering and locking miniaturized high-precision measurement device, connecting various parts of the device via various interfaces and lines.

[0050] The memory can be used to store the computer program and / or modules. The processor, by running or executing the computer program and / or modules stored in the memory, and by calling the data stored in the memory, realizes various functions of the seamless line discharge locking miniaturized high-precision measuring device. The memory may mainly include a program storage area and a data storage area. The program storage area may store the operating system, at least one application program required for a function, etc.; the data storage area may store data created based on the operation of the air conditioner controller, etc. In addition, the memory may include high-speed random access memory, and may also include non-volatile memory, such as hard disk, memory, plug-in hard disk, smart memory card (SmartMedaCard, SMC), secure digital card (SecureDagatal, SD), flash memory card, at least one disk storage device, flash memory device, or other volatile solid-state storage devices.

[0051] The module integrated into the seamless line release locking miniaturized high-precision measurement device, if implemented as a software functional unit and sold or used as an independent product, can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the above embodiments of the present invention can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium, etc.

[0052] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. The storage medium can be a magnetic disk, optical disk, read-only memory (ROM), or random access memory (RAM), etc.

[0053] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A method for seamless line spread locking miniaturization high precision measurement, characterized in that, The method comprises: Collecting rail gap opening and rail temperature data, and constructing a state input vector, wherein the rail gap opening is measured by a laser ranging module in cooperation with an image auxiliary recognition component, and an image calibration mechanism is used to ensure that the measurement point is located at the center of the rail gap, and the rail temperature is collected by an infrared thermopile temperature array to collect the rail surface temperature; Based on the state input vector, the average strain of the rail is calculated according to the rail gap opening and the design length, the initial equivalent stress is calculated by a thermal coupling model in combination with the rail temperature, the material elastic modulus and the thermal expansion coefficient, and the initial equivalent stress is corrected in structure position sensitivity by a perturbation function to obtain the corrected equivalent stress; A time series is constructed using the historical data of strain and corrected equivalent stress in a continuous time period, a dynamic trend analysis is performed by a gated recurrent network model, the gated recurrent network model introduces a thermal time sequence consistency constraint and a tension stable region slope constraint in the training stage to improve the generalization ability to the non-ideal diffusion process, and finally outputs a classification label; According to the classification label and the corrected equivalent stress, it is judged whether the stress is in a preset safe tension interval, and an operation instruction of immediate locking, suggestion waiting or prohibition locking is generated in combination with the state label, and real-time visualization and wireless prompts are provided to the construction personnel through a man-machine interaction module integrated in a small-sized terminal device.

2. The method of claim 1, wherein, The image calibration mechanism judges the laser landing position by visual assistance, and automatically adjusts the direction of the ranging module when the deviation is detected.

3. The method of claim 1, wherein the method is a seamless route dispersion locking miniaturized high-precision measurement method. At least two samplings are performed during the collection process of the rail gap opening, and the average value is taken as the final measurement result.

4. The method of claim 1, wherein, The perturbation function is a function of the normalized position of the rail gap section, which has the maximum correction amplitude in the end region of the rail gap and does not correct in the center region of the rail gap.

5. The method of claim 1, wherein, The thermal coupling model specifically constructs the equivalent stress by the following way: the rail strain is calculated based on the proportional relationship between the rail gap opening and the rail design length; the rail strain, the real-time collected rail temperature, and the elastic modulus and thermal expansion coefficient of the rail material changing with temperature are taken as input parameters; the initial equivalent stress of the rail is calculated by coupling the thermal elastic effect of the strain and the thermal expansion effect caused by the temperature.

6. The method of claim 1, wherein, The gated recurrent network model adopts a multi-layer GRU structure, the input is the time series of strain and corrected stress, and the output is the state classification label.

7. The method of claim 1, wherein the method is a seamless route dispersion locking miniaturized high-precision measurement method. The thermal time sequence consistency constraint is used to maintain the coordination of stress change and strain change, and the tension stable region slope constraint is used to suppress the interference of short-term tension fluctuation on judgment.

8. The method of claim 1, wherein, The classification label includes three types of non-locking, locking and state abnormality.

9. The method of claim 1, wherein, The man-machine interaction module provides at least one of visual prompt and wireless broadcast prompt.

10. A miniature high-precision measuring device for a seamless line spread lock, characterized in that The device comprises: A data collection module is configured to collect rail gap opening and rail temperature data, and construct a state input vector, wherein the rail gap opening is measured by a laser ranging module in cooperation with an image auxiliary recognition component, and an image calibration mechanism is used to ensure that the measurement point is located at the center of the rail gap, and the rail temperature is collected by an infrared thermopile temperature array to collect the rail surface temperature; The data processing module is configured to calculate the average strain of the steel rail based on the state input vector according to the rail joint opening and the design length, calculate the initial equivalent stress by a thermal force coupling model in combination with the temperature of the steel rail, the material elastic modulus and the thermal expansion coefficient, and perform a structure position sensitive correction on the initial equivalent stress by a perturbation function to obtain a corrected equivalent stress; The state recognition module is configured to construct a time sequence by using historical data of the strain and the corrected equivalent stress in a continuous time period, and perform a dynamic trend analysis by a gated recurrent network model, wherein the gated recurrent network model introduces a thermal force time sequence consistency constraint and a tension stable region slope constraint in a training stage to improve the generalization ability for a non-ideal dissipation process, and finally outputs a classification label; The decision output module is configured to determine whether the stress is in a preset safe tension interval according to the classification label and the corrected equivalent stress, generate an operation instruction of immediate locking, suggested waiting or prohibited locking in combination with the state label, and provide real-time visualization and wireless prompts to construction personnel through a human-computer interaction module integrated in a small-sized terminal device.

Images