Rail transit contact line abrasion detection device and method
By using a rail transit contact line wear detection device, combined with intelligent supplementary lighting, pose compensation and multi-source sensor fusion technology, the problem of insufficient detection accuracy in existing technologies has been solved. This enables efficient and accurate wear detection and abnormal damage identification, supports multi-level early warning, and improves the scientific and economical nature of operation and maintenance management.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- GUANGDONG HUANENG ELECTROMECHANICAL GRP CO LTD
- Filing Date
- 2026-02-06
- Publication Date
- 2026-06-05
AI Technical Summary
Existing technologies are insufficient for efficiently and accurately detecting wear on rail transit contact lines under complex operating conditions, resulting in insufficient detection accuracy, poor robustness, and a high false alarm rate, which cannot meet the maintenance needs of high-density operating lines.
A rail transit contact line wear detection device is adopted, including an intelligent supplementary lighting module, a data acquisition module, a pose compensation module, an image processing module, and an early warning module. Through precise hardware configuration and multi-layer collaborative control, it realizes high frame rate image acquisition, adaptive lighting adjustment, pose compensation, and multi-source sensor data fusion, and combines a deep learning model for wear analysis and early warning.
It achieves high-precision detection of contact wire wear in complex environments, significantly improves the environmental adaptability and reliability of the detection system, provides a comprehensive assessment capability from conventional wear quantification to abnormal damage, supports multi-level early warning, and improves the scientific and economical nature of operation and maintenance management.
Smart Images
Figure CN122149313A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of rail transit technology, and in particular to a device and method for detecting wear on rail transit contact lines. Background Technology
[0002] Electrified rail transit systems (including high-speed railways, subways, and light rail) obtain electrical energy through sliding contact between a pantograph and an overhead contact line. During long-term operation, the contact line inevitably experiences wear due to mechanical friction from the pantograph's sliding contact plate, arc erosion, and environmental corrosion. This wear reduces its cross-sectional area, mechanical strength, and current-carrying capacity. When this wear exceeds safety limits, it can easily lead to wire breakage, arcing, or even pantograph-catenary accidents, seriously threatening train safety.
[0003] Currently, contact wire wear detection mainly relies on the following methods: 1. Manual inspection, which is carried out by maintenance personnel using calipers or portable thickness gauges to climb up and measure during the maintenance window, has problems such as low efficiency, limited coverage, large subjective error, and cannot achieve continuous monitoring; 2. Although contact-type online inspection vehicles can automatically measure, they require dedicated rail vehicles, which are costly and have a low inspection frequency, making it difficult to meet the maintenance needs of high-density operating lines. 3. Non-contact visual inspection methods, such as contour scanning using laser or image technology, while possessing online potential, still face significant challenges in practical applications: high-speed train operation causes image motion blur, alternating tunnel and open-air environments result in drastic changes in lighting, oil stains, oxidation, or reflections on the contact wire surface interfere with contour extraction, and vehicle vibration causes camera pose disturbances. These factors contribute to insufficient measurement accuracy, poor robustness, and high false alarm rates in existing systems under complex conditions. Therefore, we propose a device and method for detecting wear on rail transit contact wires. Summary of the Invention
[0004] The main objective of this invention is to provide a device and method for detecting wear of rail transit contact lines, which can effectively solve the problems in the background art.
[0005] To achieve the above objectives, the technical solution adopted by the present invention is as follows: A rail transit contact wire wear detection device, comprising: The equipment platform is used to stably integrate and reliably deploy various functional modules in testing scenarios; The intelligent supplemental lighting module is used to adaptively adjust the light source parameters according to the ambient light and the surface condition of the contact line, providing high-contrast and low-interference lighting conditions; The acquisition module is used to synchronously acquire high-resolution optical images of the contact wire during train operation, serving as the raw visual data source for wear analysis. The pose compensation module is used to sense the spatial pose and motion state of the device in real time, and to perform dynamic distortion correction and coordinate alignment on the acquired images to eliminate measurement errors caused by vibration and motion. The image processing module is used to extract contours, reconstruct three dimensions and calculate wear features from the acquired images, and output quantitative evaluation results such as the remaining thickness of the contact line and the cross-sectional area loss. The early warning module compares the wear data output by the image processing unit with a preset safety threshold. When excessive wear or abnormal damage is detected, it automatically triggers a graded early warning signal and sends alarm information to the operation and maintenance system.
[0006] As a further improvement to the above solution, in order to provide a high-precision hardware configuration foundation and ensure image acquisition quality and laser contour extraction accuracy, the equipment platform includes a mounting base, a line laser generator, a white LED array, and a high-speed industrial camera. The wavelength of the line laser generator is 635nm, the line width is less than 1mm, and the resolution of the high-speed industrial camera is not less than 2448×2048 and supports a frame rate greater than or equal to 90 fps.
[0007] Through the above improvements and precise hardware parameter configuration, the system can clearly capture millimeter-level minute wear on the contact wire, meeting the high-precision detection requirements of rail transit. At the same time, the high frame rate ensures image stability under high-speed operation conditions.
[0008] As a further improvement to the above solution, in order to achieve adaptive adjustment of lighting conditions and eliminate the impact of environmental changes on imaging quality, the intelligent supplementary lighting module includes: an ambient light sensing unit for real-time monitoring of the ambient illuminance of the detection area; a light source control unit for synchronously controlling the start-stop timing and output power of the line laser generator and the white LED array based on the output of the ambient light sensing unit and the exposure signal of the acquisition module; and a light field optimization unit for dynamically adjusting the focusing position and linewidth uniformity of the emitted light spot of the line laser generator.
[0009] Through the above improvements and the three-layer collaborative control, stable high-contrast images can be obtained in complex lighting environments such as tunnels, daytime, and nighttime, significantly improving the environmental adaptability and reliability of the detection system.
[0010] As a further improvement to the above scheme, in order to achieve precise closed-loop control of the light source output and maintain the stability of image brightness, the light source control unit is configured to run a PID control algorithm. It uses the real-time average gray value of the contact line area in the image as the feedback signal, and the preset target gray value as the set value. The control quantity is generated through proportional-integral-derivative operations, and the driving current of the line laser generator is adjusted accordingly to keep the imaging brightness stable within the optimal dynamic range. The integral term of the PID control algorithm is equipped with an anti-saturation limit, and the control cycle is synchronized with the camera frame rate.
[0011] Through the above improvements, illumination fluctuations are eliminated by PID closed-loop control, ensuring the consistency of image quality and providing a reliable data foundation for subsequent accurate measurements. At the same time, the anti-saturation design prevents control oscillations.
[0012] As a further improvement to the above solution, in order to eliminate the geometric distortion caused by train vibration and high-speed motion to image acquisition, the pose compensation module includes an inertial measurement unit, a displacement synchronization unit, and a spatiotemporal synchronization control unit; the inertial measurement unit senses the attitude disturbance of the device in real time; the displacement synchronization unit provides accurate displacement information along the track direction; the spatiotemporal synchronization control unit fuses the two to generate a six-degree-of-freedom pose sequence and strictly synchronizes it with the exposure time of the acquisition module.
[0013] Through the above improvements, high-precision pose compensation is achieved through multi-unit collaborative work, ensuring accurate correspondence between the image and the actual position, and significantly improving the accuracy of 3D reconstruction and wear calculation.
[0014] As a further improvement to the above scheme, in order to achieve accurate spatiotemporal alignment of multi-sensor data and provide high-precision pose information, the spatiotemporal synchronization control unit adopts a multi-source heterogeneous sensor fusion algorithm. Specifically, the multi-source heterogeneous sensor fusion algorithm is an extended Kalman filter algorithm. The extended Kalman filter algorithm fuses the high-frequency angular velocity and acceleration data output by the inertial measurement unit and the low-frequency displacement information provided by the displacement synchronization unit to estimate the six-degree-of-freedom pose of the device in the orbital coordinate system in real time. At the same time, the data of each sensor is timestamped based on the hardware trigger signal, and the corresponding accurate pose label is generated for each frame of image through IMU pre-integration technology.
[0015] Through the above improvements, the EKF algorithm effectively integrates sensor data of different frequencies, achieving sub-millimeter-level spatial positioning accuracy and providing a reliable guarantee for accurate measurement under high-speed operating conditions.
[0016] As a further improvement to the above solution, to achieve an automated processing flow from the original image to the final result, the image processing module includes an image preprocessing unit, a contour extraction unit, a 3D reconstruction unit, a wear analysis unit, and a data fusion and output unit connected in sequence. The image preprocessing unit performs noise reduction and correction on the original image; the contour extraction unit extracts the center line of the contact line laser light stripe; the 3D reconstruction unit generates a cross-sectional point cloud based on the principle of triangulation; the wear analysis unit calculates the remaining thickness and cross-sectional area loss; and finally, the data fusion and output unit generates a wear assessment report with location labels.
[0017] Through the above improvements, the modular design enhances processing efficiency and reliability, ensures full automation from image acquisition to wear assessment, and significantly improves the practicality and accuracy of the detection system.
[0018] As a further improvement to the above scheme, in order to achieve accurate wear quantification analysis and intelligent identification of abnormal damage, the wear analysis unit is configured to execute a standard cross-section template matching algorithm, align the reconstructed three-dimensional cross-section of the contact line with the preset new line model through the ICP algorithm, and calculate the minimum remaining thickness and cross-sectional area loss rate along the normal direction of the design working surface; the wear analysis unit also integrates a lightweight convolutional neural network to identify abnormal wear patterns that are difficult to identify by geometric methods, including grooves, cracks or arc ablation areas; the wear analysis unit adopts a hybrid analysis strategy: firstly, it calculates basic wear parameters based on the geometric template, and when it detects local curvature abnormalities or the wear difference between the left and right sides exceeds the threshold, it automatically calls a deep learning model for secondary diagnosis.
[0019] Through the above improvements, the geometric method ensures the accuracy of basic parameters, deep learning enhances the anomaly recognition capability, and the hybrid strategy improves the overall detection accuracy, providing more comprehensive and reliable technical support for operation and maintenance decisions.
[0020] A method for detecting wear on rail transit contact wires, implemented using the rail transit contact wire wear detection device described in this invention, includes the following steps: S1. Install the detection device on the top of the train or the trackside platform, and load the standard cross-section template, safety threshold and system initial parameters after powering on; S2. Real-time monitoring of lighting conditions via an ambient light sensor, with the light source control unit adaptively adjusting the intensity, angle, and triggering sequence of the linear laser and LED to provide optimal illumination; S3. During train operation, the high-speed camera is synchronously triggered to acquire images of the contact line, and the IMU and encoder data are fused to obtain the six-degree-of-freedom pose and track position at the corresponding moment. S4. Denoise and correct distortion of the image, extract the laser center line, and reconstruct the local three-dimensional cross-sectional morphology of the contact line based on triangulation or stereo vision. S5. Register the reconstructed cross section with the standard template, calculate the minimum remaining thickness, cross-sectional area loss rate and degree of wear, and identify abnormal damage by combining a deep learning model. S6. Based on the comparison between the wear results and the preset threshold, trigger a graded early warning, generate maintenance suggestions or alarm information, and bind and store data according to mileage to form an electronic file.
[0021] As a further improvement to the above scheme, in order to establish a scientific and reasonable hierarchical early warning system and achieve full coverage from normal monitoring to emergency intervention, in S6, when the remaining thickness is less than 55% of the design value or structural damage is detected, an emergency early warning is triggered; when the wear is in the range of 75%–85%, it is only recorded as a normal state.
[0022] Through the above improvements, precise threshold classification and graded response mechanisms have enabled differentiated risk management, avoiding resource waste caused by excessive early warnings and ensuring timely response to emergencies, thereby enhancing the scientific and economical nature of operation and maintenance management.
[0023] Compared with the prior art, the present invention has the following beneficial effects: 1. By setting up an intelligent light source control unit, the adaptability and stability of the imaging system in complex lighting environments are effectively improved. This unit can dynamically adjust the output parameters of the line laser and auxiliary light source according to changes in ambient light and image quality feedback, and precisely synchronize with camera exposure. Thus, even under interference conditions such as tunnels, strong light, rain, fog, or reflections from the contact line surface, it can still acquire high-contrast, low-noise contour images, providing a reliable data foundation for subsequent processing.
[0024] 2. By introducing a spatiotemporal synchronization control unit, the consistency and accuracy between pose information and image acquisition are significantly improved. This unit integrates inertial measurement and displacement sensing data, and uses a sensor fusion algorithm to achieve temporal alignment and spatial registration of multi-source heterogeneous signals. It also generates pose labels that strictly correspond to each frame of the image, effectively suppressing geometric distortions caused by high-speed train operation, vibration, or track irregularities, and ensuring the geometric accuracy of 3D cross-section reconstruction.
[0025] 3. By constructing a wear analysis unit, a comprehensive assessment capability has been achieved, ranging from the quantification of conventional wear to the identification of abnormal damage. This unit combines standard cross-section template matching with an intelligent recognition model, which can not only calculate engineering indicators such as remaining thickness and cross-sectional area loss, but also identify non-uniform or structural damage such as grooves, cracks, and uneven wear. It also supports a multi-level early warning mechanism, making the detection results more diagnostically valuable and helping to achieve condition-based repair and refined operation and maintenance of the overhead contact system. Attached Figure Description
[0026] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0027] Figure 1 This is a structural block diagram of a rail transit contact wire wear detection device according to the present invention; Figure 2 This is a structural block diagram of the intelligent supplementary lighting module of the present invention; Figure 3 This is a structural block diagram of the pose compensation module of the present invention; Figure 4 This is a structural block diagram of the image processing module of the present invention; Figure 5 This is a flowchart of the rail transit contact wire wear detection method of the present invention. Detailed Implementation
[0028] To make the technical means, creative features, objectives and effects of this invention easier to understand, the invention will be further described below in conjunction with specific embodiments.
[0029] In the description of this invention, it should be noted that the terms "upper," "lower," "inner," "outer," "front end," "rear end," "both ends," "one end," and "the other end," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing this invention and for simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention. Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.
[0030] In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "installed," "equipped with," "connected," etc., should be interpreted broadly. For example, "connection" can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium; it can be a connection within two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.
[0031] The technical solution of the present invention will be further described below with reference to the accompanying drawings.
[0032] Example: Figure 1-4 As shown, a rail transit contact wire wear detection device includes: The equipment platform is used to stably integrate and reliably deploy various functional modules in testing scenarios; The intelligent supplemental lighting module is used to adaptively adjust the light source parameters according to the ambient light and the surface condition of the contact line, providing high-contrast and low-interference lighting conditions; The acquisition module is used to synchronously acquire high-resolution optical images of the contact wire during train operation, serving as the raw visual data source for wear analysis. The pose compensation module is used to sense the spatial pose and motion state of the device in real time, and to perform dynamic distortion correction and coordinate alignment on the acquired images to eliminate measurement errors caused by vibration and motion. The image processing module is used to extract contours, reconstruct three dimensions and calculate wear features from the acquired images, and output quantitative evaluation results such as the remaining thickness of the contact line and the cross-sectional area loss. The early warning module compares the wear data output by the image processing unit with a preset safety threshold. When excessive wear or abnormal damage is detected, it automatically triggers a graded early warning signal and sends alarm information to the operation and maintenance system.
[0033] Preferably, the equipment platform includes a mounting base, a line laser generator, a white LED array, and a high-speed industrial camera. The line laser generator has a wavelength of 635nm and a line width of less than 1mm. The high-speed industrial camera has a resolution of not less than 2448×2048 and supports a frame rate of greater than or equal to 90 fps.
[0034] Preferably, the intelligent supplementary lighting module includes: an ambient light sensing unit for real-time monitoring of the ambient illuminance of the detection area; a light source control unit for synchronously controlling the start-stop timing and output power of the line laser generator and the white LED array based on the output of the ambient light sensing unit and the exposure signal of the acquisition module; and a light field optimization unit for dynamically adjusting the focusing position and linewidth uniformity of the emitted light spot of the line laser generator.
[0035] Preferably, the light source control unit is configured to run a PID control algorithm, using the real-time average gray value of the contact line area in the image as a feedback signal, and a preset target gray value as a set value. The control quantity is generated through proportional-integral-derivative operations, and the driving current of the line laser generator is adjusted accordingly to stabilize the imaging brightness within the optimal dynamic range. The integral term of the PID control algorithm is provided with an anti-saturation limit, and the control cycle is synchronized with the camera frame rate.
[0036] Preferably, the pose compensation module includes an inertial measurement unit, a displacement synchronization unit, and a spatiotemporal synchronization control unit; the inertial measurement unit senses the attitude disturbance of the device in real time; the displacement synchronization unit provides accurate displacement information along the track direction; and the spatiotemporal synchronization control unit fuses the two to generate a six-degree-of-freedom pose sequence, which is strictly synchronized with the exposure time of the acquisition module.
[0037] Preferably, the spatiotemporal synchronization control unit is implemented using a multi-source heterogeneous sensor fusion algorithm, specifically an extended Kalman filter algorithm. This extended Kalman filter algorithm fuses high-frequency angular velocity and acceleration data output by the inertial measurement unit and low-frequency displacement information provided by the displacement synchronization unit to estimate the six-degree-of-freedom pose of the device in the orbital coordinate system in real time. Simultaneously, it timestamps the data from each sensor based on hardware trigger signals and generates corresponding precise pose labels for each frame of image using IMU pre-integration technology.
[0038] Furthermore, the extended Kalman filter algorithm specifically includes: S1, predicting the current state based on the state estimate of the previous time step and the current IMU measurement value through a dynamic model; S2, calculating the state transition Jacobian matrix and propagating the error covariance of the previous time step to the current time step; S3, calculating the weighting coefficients for fusing the predicted value and the observed value based on the prediction covariance and the observation noise covariance; S4, comparing the predicted value with the actual observed value to obtain the residual, and correcting the predicted state using Kalman gain; S5, updating the error covariance matrix based on the Kalman gain to reflect the new uncertainty level after correction.
[0039] Preferably, the image processing module includes an image preprocessing unit, a contour extraction unit, a 3D reconstruction unit, a wear analysis unit, and a data fusion and output unit connected in sequence; the image preprocessing unit performs noise reduction and correction on the original image; the contour extraction unit extracts the center line of the contact line laser light stripe; the 3D reconstruction unit generates a cross-sectional point cloud based on the triangulation principle; the wear analysis unit calculates the remaining thickness and cross-sectional area loss; and finally, the data fusion and output unit generates a wear assessment report with location labels.
[0040] Preferably, the wear analysis unit is configured to execute a standard cross-section template matching algorithm to align the reconstructed three-dimensional cross-section of the contact line with a preset new line model using the ICP algorithm, and calculate the minimum remaining thickness and cross-sectional area loss rate along the normal direction of the design working surface; the wear analysis unit also integrates a lightweight convolutional neural network to identify abnormal wear patterns that are difficult to identify by geometric methods, including grooves, cracks, or arc ablation areas; the wear analysis unit adopts a hybrid analysis strategy: firstly, it calculates basic wear parameters based on a geometric template, and when it detects local curvature abnormalities or wear differences between the left and right sides exceeding a threshold, it automatically calls a deep learning model for secondary diagnosis.
[0041] Furthermore, the formula for the minimum remaining thickness is: ,in, The minimum remaining thickness (mm). Let be the remaining thickness (mm) at the i-th measurement point on the contact wire cross-section; the formula for the cross-sectional area loss rate is: ,in, The cross-sectional area loss rate (%) This represents the theoretical area (mm²) of the standard new line cross-section. The current remaining cross-sectional area (mm²) is calculated from the measured three-dimensional cross-section or two-dimensional profile.
[0042] like Figure 5 As shown, a method for detecting wear on rail transit contact wires, implemented using the rail transit contact wire wear detection device described in this invention, includes the following steps: S1. Install the detection device on the top of the train or the trackside platform, and load the standard cross-section template, safety threshold and system initial parameters after powering on; S2. Real-time monitoring of lighting conditions via an ambient light sensor, with the light source control unit adaptively adjusting the intensity, angle, and triggering sequence of the linear laser and LED to provide optimal illumination; S3. During train operation, the high-speed camera is synchronously triggered to acquire images of the contact line, and the IMU and encoder data are fused to obtain the six-degree-of-freedom pose and track position at the corresponding moment. S4. Denoise and correct distortion of the image, extract the laser center line, and reconstruct the local three-dimensional cross-sectional morphology of the contact line based on triangulation or stereo vision. S5. Register the reconstructed cross section with the standard template, calculate the minimum remaining thickness, cross-sectional area loss rate and degree of wear, and identify abnormal damage by combining a deep learning model. S6. Based on the comparison between the wear results and the preset threshold, trigger a graded early warning, generate maintenance suggestions or alarm information, and bind and store data according to mileage to form an electronic file.
[0043] Furthermore, when the remaining thickness is less than 55% of the design value or structural damage is detected, an emergency warning is triggered; when the wear is in the 75%–85% range, it is only recorded as a normal state. Through precise threshold division and graded response mechanism, differentiated risk management is achieved, which not only avoids the waste of resources caused by excessive warnings, but also ensures timely response to emergencies, and improves the scientificity and economy of operation and maintenance management.
[0044] In practical application, this embodiment effectively enhances the adaptability and stability of the imaging system under complex lighting conditions by incorporating an intelligent light source control unit. This unit dynamically adjusts the output parameters of the line laser and auxiliary light source based on changes in ambient light and image quality feedback, and precisely synchronizes with camera exposure. This allows for the acquisition of high-contrast, low-noise contour images even under interference conditions such as tunnels, strong light, rain, fog, or surface reflections on the contact line, providing a reliable data foundation for subsequent processing. Furthermore, the introduction of a spatiotemporal synchronization control unit significantly improves the consistency and accuracy between pose information and image acquisition. This unit integrates inertial measurement and displacement sensing data, employing sensor fusion... The algorithm achieves temporal alignment and spatial registration of multi-source heterogeneous signals and generates pose labels that strictly correspond to each frame of the image. This effectively suppresses geometric distortion caused by high-speed train operation, vibration, or track irregularities, ensuring the geometric accuracy of the 3D cross-section reconstruction. By constructing a wear analysis unit, it realizes a comprehensive evaluation capability from conventional wear quantification to abnormal damage identification. This unit, combined with standard cross-section template matching and intelligent recognition model, can not only calculate engineering indicators such as remaining thickness and cross-sectional area loss, but also identify non-uniform or structural damage such as grooves, cracks, and uneven wear. It also supports a multi-level early warning mechanism, making the detection results more diagnostically valuable and helping to achieve condition-based repair and refined operation and maintenance of the overhead contact system.
[0045] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of this invention is defined by the appended claims and their equivalents.
Claims
1. A device for detecting wear on rail transit contact lines, characterized in that, include: The equipment platform is used to stably integrate and reliably deploy various functional modules in testing scenarios; The intelligent supplemental lighting module is used to adaptively adjust the light source parameters according to the ambient light and the surface condition of the contact line, providing high-contrast and low-interference lighting conditions; The acquisition module is used to synchronously acquire high-resolution optical images of the contact wire during train operation, serving as the raw visual data source for wear analysis. The pose compensation module is used to sense the spatial pose and motion state of the device in real time, and to perform dynamic distortion correction and coordinate alignment on the acquired images to eliminate measurement errors caused by vibration and motion. The image processing module is used to extract contours, reconstruct three dimensions and calculate wear features from the acquired images, and output quantitative evaluation results such as the remaining thickness of the contact line and the cross-sectional area loss. The early warning module compares the wear data output by the image processing unit with a preset safety threshold. When excessive wear or abnormal damage is detected, it automatically triggers a graded early warning signal and sends alarm information to the operation and maintenance system.
2. The rail transit contact wire wear detection device according to claim 1, characterized in that: The equipment platform includes a mounting base, a line laser generator, a white LED array, and a high-speed industrial camera. The line laser generator has a wavelength of 635nm and a line width of less than 1mm. The high-speed industrial camera has a resolution of not less than 2448×2048 and supports a frame rate of greater than or equal to 90 fps.
3. The rail transit contact wire wear detection device according to claim 1, characterized in that: The intelligent supplementary lighting module includes: an ambient light sensing unit for real-time monitoring of the ambient illuminance of the detection area; a light source control unit for synchronously controlling the start-stop timing and output power of the line laser generator and the white LED array based on the output of the ambient light sensing unit and the exposure signal of the acquisition module; and a light field optimization unit for dynamically adjusting the focusing position and linewidth uniformity of the emitted light spot of the line laser generator.
4. The rail transit contact wire wear detection device according to claim 3, characterized in that: The light source control unit is configured to run a PID control algorithm, using the real-time average gray value of the contact line area in the image as the feedback signal, and a preset target gray value as the set value. The control quantity is generated through proportional-integral-derivative operations, and the driving current of the line laser generator is adjusted accordingly to stabilize the imaging brightness within the optimal dynamic range. The integral term of the PID control algorithm is equipped with an anti-saturation limit, and the control cycle is synchronized with the camera frame rate.
5. The rail transit contact wire wear detection device according to claim 1, characterized in that: The pose compensation module includes an inertial measurement unit, a displacement synchronization unit, and a spatiotemporal synchronization control unit; the inertial measurement unit senses the attitude disturbance of the device in real time; the displacement synchronization unit provides accurate displacement information along the track direction; the spatiotemporal synchronization control unit fuses the two to generate a six-degree-of-freedom pose sequence and strictly synchronizes it with the exposure time of the acquisition module.
6. The rail transit contact wire wear detection device according to claim 5, characterized in that: The spatiotemporal synchronization control unit is implemented using a multi-source heterogeneous sensor fusion algorithm, specifically an extended Kalman filter algorithm. This algorithm fuses high-frequency angular velocity and acceleration data output by the inertial measurement unit and low-frequency displacement information provided by the displacement synchronization unit to estimate the six-degree-of-freedom pose of the device in the orbital coordinate system in real time. Simultaneously, it timestamps the data from each sensor based on hardware trigger signals and generates corresponding precise pose labels for each frame of image using IMU pre-integration technology.
7. The rail transit contact wire wear detection device according to claim 1, characterized in that: The image processing module includes an image preprocessing unit, a contour extraction unit, a 3D reconstruction unit, a wear analysis unit, and a data fusion and output unit connected in sequence. The image preprocessing unit performs noise reduction and correction on the original image. The contour extraction unit extracts the center line of the laser light stripe on the contact line. The 3D reconstruction unit generates a cross-sectional point cloud based on the principle of triangulation. The wear analysis unit calculates the remaining thickness and cross-sectional area loss. Finally, the data fusion and output unit generates a wear assessment report with location labels.
8. The rail transit contact wire wear detection device according to claim 7, characterized in that: The wear analysis unit is configured to execute a standard cross-section template matching algorithm to align the reconstructed three-dimensional cross-section of the contact line with the preset new line model using the ICP algorithm, and calculate the minimum remaining thickness and cross-sectional area loss rate along the normal direction of the design working surface. The wear analysis unit also integrates a lightweight convolutional neural network to identify abnormal wear patterns that are difficult to identify by geometric methods, including grooves, cracks, or arc ablation areas. The wear analysis unit adopts a hybrid analysis strategy: first, it calculates basic wear parameters based on the geometric template; when it detects local curvature abnormalities or wear differences between the left and right sides exceeding a threshold, it automatically calls a deep learning model for secondary diagnosis.
9. A method for detecting wear on rail transit contact lines, implemented using the rail transit contact line wear detection device as described in any one of claims 1 to 8, characterized in that, Includes the following steps: S1. Install the detection device on the top of the train or the trackside platform, and load the standard cross-section template, safety threshold and system initial parameters after powering on; S2. Real-time monitoring of lighting conditions via an ambient light sensor, with the light source control unit adaptively adjusting the intensity, angle, and triggering sequence of the linear laser and LED to provide optimal illumination; S3. During train operation, the high-speed camera is synchronously triggered to acquire images of the contact line, and the IMU and encoder data are fused to obtain the six-degree-of-freedom pose and track position at the corresponding moment. S4. Denoise and correct distortion of the image, extract the laser center line, and reconstruct the local three-dimensional cross-sectional morphology of the contact line based on triangulation or stereo vision. S5. Register the reconstructed cross section with the standard template, calculate the minimum remaining thickness, cross-sectional area loss rate and degree of wear, and identify abnormal damage by combining a deep learning model. S6. Based on the comparison between the wear results and the preset threshold, trigger a graded early warning, generate maintenance suggestions or alarm information, and bind and store data according to mileage to form an electronic file.
10. The method for detecting wear of rail transit contact wires according to claim 9, characterized in that: In S6, an emergency warning is triggered when the remaining thickness is less than 55% of the design value or when structural damage is detected; when the wear is in the 75%–85% range, it is recorded as a normal state.