Visual measurement device for wear elongation of chain ring

Abstract

The invention provides a chain ring wear elongation visual measurement device, and relates to the technical field of coal mine machinery intelligent monitoring, and the chain ring wear elongation visual measurement device is characterized in that an environment sensing system collects environment parameters of a coal mine well where a scraper conveyor is located in real time, and uploads the collected environment parameters to an industrial personal computer; the industrial personal computer determines an environment state according to the uploaded environment parameters and performs light supplement mode switching and light supplement power adjustment on the self-adaptive light supplement system according to the environment state so as to adjust the environment state of the underground coal mine; the visual measurement system performs visual image acquisition on the chain ring and uploads acquired image data to the industrial personal computer, and the industrial personal computer processes the image data to obtain chain ring positioning data; and the health diagnosis system analyzes and judges the wear state of the chain ring according to the chain ring positioning data in combination with an accumulated elongation analysis principle of the multi-section chain ring, and performs early warning operation according to an analysis and judgment result. According to the scheme, the accuracy of measuring the elongation of the chain ring in a low-illumination and high-dust environment can be improved.

Description

Technical Field

[0001] This invention relates to the field of intelligent monitoring technology for coal mining machinery, and in particular to a visual measurement device for chain link wear elongation rate. Background Technology

[0002] Scraper conveyors are core equipment in coal mining. Their chains are subjected to high-load impacts over long periods. Excessive cumulative elongation due to chain link wear is a major cause of chain breakage accidents. Therefore, monitoring the chain link wear elongation rate is of great significance. Traditional monitoring methods rely on contact-type devices such as tension sensors and proximity switches, which suffer from problems such as easy clogging by coal dust, signal distortion, and inability to cover the entire chain, resulting in a high rate of missed detections. Manual inspection methods, on the other hand, lack real-time performance and cannot identify scraper offset.

[0003] Currently, while machine vision systems exist for measurement, their dynamic tracking accuracy is insufficient in low-light and high-dust environments, resulting in high false alarm rates. For example, the extraction of the ROI region of a chain link relies on manual annotation, which cannot adaptively adjust in real time according to the chain's high-speed movement and random swaying, leading to dynamic tracking failure and target loss, thus failing to meet real-time monitoring requirements. Furthermore, high dust and oil deposits in coal mines severely interfere with imaging of the chain link's inner holes; existing algorithms struggle to effectively distinguish between true contours and artifacts, resulting in large errors in chain link center positioning and directly impacting the accuracy of wear calculations. Additionally, judging faults solely based on changes in the length of a single chain link ignores the cumulative effect of chain wear, leading to high false alarm rates. Moreover, in complex environments such as low light and sudden increases in dust, existing systems lack dynamic supplemental lighting and environmental adaptation mechanisms, resulting in unstable imaging quality. Additionally, traditional fixed installation methods are susceptible to image blurring due to equipment vibration. Summary of the Invention

[0004] In view of this, it is necessary to propose a visual measurement device for chain link wear elongation to solve at least one of the above-mentioned shortcomings, thereby improving the accuracy of chain link elongation measurement in low-light and high-dust environments.

[0005] This invention provides a visual measurement device for chain link wear elongation, comprising: a support assembly, an environmental sensing system, an adaptive supplementary lighting system, a visual measurement system, a health diagnosis system, and an industrial control computer; the support assembly is installed on the tail coal retaining plate of the scraper conveyor, the environmental sensing system, the adaptive supplementary lighting system, and the visual measurement system are installed on the support assembly and are all electrically connected to the industrial control computer; the health diagnosis system is deployed in the industrial control computer;

[0006] The environmental sensing system is used to collect environmental parameters of the coal mine where the scraper conveyor is located in real time, and to upload the collected environmental parameters to the industrial control computer; wherein, the environmental parameters include ambient light intensity, ambient temperature and dust concentration.

[0007] The industrial control computer is used to determine the environmental state based on the uploaded environmental parameters, and to switch the supplementary lighting mode and adjust the supplementary lighting power of the adaptive supplementary lighting system according to the environmental state, so as to adjust the current environmental state in the coal mine.

[0008] The vision measurement system is used to acquire visual images of the chain links and upload the acquired image data to an industrial control computer, which then processes the image data to obtain chain link positioning data.

[0009] The health diagnostic system is used to analyze and judge the wear status of the chain links based on the chain link positioning data and the cumulative elongation rate analysis principle of multiple chain links, and to execute early warning operations based on the analysis and judgment results.

[0010] Preferably, the support assembly includes: a support frame fixing plate, a camera vertical support frame, a camera horizontal support frame, a shock absorber, a universal joint, and a camera mounting bracket; the support frame fixing plate has a U-shaped structure and is installed on the tail coal retaining plate of the scraper conveyor; one end of the camera horizontal support frame is fixedly installed on the outside of the support frame fixing plate, and the camera vertical support frame is installed on the camera horizontal support frame in a vertically intersecting manner and is adjustable in the horizontal direction; the upper end of the shock absorber is installed on the camera vertical support frame and is adjustable in the vertical direction; the upper end of the universal joint is installed and connected to the lower end of the shock absorber, and the lower end is installed and connected to the upper end of the camera mounting bracket; the environmental perception system and the adaptive lighting system are installed on the camera horizontal support frame, and the visual measurement system is installed on the camera mounting bracket.

[0011] Preferably, the environmental sensing system includes: a light sensor, a dust sensor, and a temperature sensor;

[0012] The light sensor is used to collect the ambient light intensity at the current location of the coal mine.

[0013] The dust sensor is used to detect the concentration of particulate matter in the air;

[0014] The temperature sensor is used to detect the ambient temperature at the current location in the coal mine.

[0015] Preferably, the industrial control computer is equipped with a filtering module, a data normalization module, and an environmental comprehensive score calculation module;

[0016] The filtering module is used to perform noise reduction and smoothing processing on the environmental parameters uploaded by the environmental perception system using Kalman filtering to eliminate instantaneous interference and obtain filtered data.

[0017] The data normalization module is used to normalize the ambient light intensity, ambient temperature and dust concentration in the filtered data to obtain normalized data.

[0018] The environmental comprehensive score calculation module is used to perform weighted fusion calculation on the normalized values ​​of light intensity, dust concentration and temperature in the normalized data to obtain the environmental comprehensive score, and then send the comprehensive score to the adaptive supplementary lighting system.

[0019] Preferably, the environmental comprehensive score calculation module calculates the environmental comprehensive score based on the following formula: ;

[0020] in, For comprehensive environmental scoring, , , These are the weighted values ​​for ambient light intensity, dust concentration, and ambient temperature, respectively. This is the normalized value of light intensity. This is the normalized value of dust concentration. This is the normalized temperature value.

[0021] Preferably, the adaptive supplemental lighting system includes: a polarized LED module, a near-infrared pulse module, a laser scanning module, and a drive controller;

[0022] The drive controller is used to receive the comprehensive environmental score and analyze and judge the comprehensive environmental score; if the comprehensive environmental score is not less than a first preset score threshold, the polarized LED module is activated for conventional supplemental lighting; if the comprehensive environmental score is less than the first preset score threshold but not less than a second preset score threshold, the near-infrared pulse module is activated for supplemental lighting; if the comprehensive environmental score is less than the second preset score threshold, the laser scanning module is activated for supplemental lighting; wherein, the first preset score threshold is greater than the second preset score threshold.

[0023] Preferably, the drive controller is further configured to increase the supplementary light power by a first preset threshold when the comprehensive environmental score decreases by a third preset threshold, based on the upper limit of the comprehensive environmental score for each supplementary light mode.

[0024] Preferably, the vision measurement system includes: an explosion-proof industrial camera and an encoder electrically connected thereto;

[0025] The explosion-proof industrial camera is equipped with a lens with an anti-fouling coating to reduce the interference of dust and oil on imaging.

[0026] The encoder is used to capture chain displacement signals in real time, and when the chain link moves to the vicinity of the center of the field of view of the explosion-proof industrial camera, it triggers the explosion-proof industrial camera to acquire images, and the explosion-proof industrial camera uploads the acquired images to the industrial control computer.

[0027] The industrial control computer is also used to quickly locate chain links and determine the ROI region based on the YOLOv7-tiny model, and to locate chain links through an edge detection algorithm.

[0028] Preferably, the health diagnosis system includes: an edge computing module, an early warning actuator, and a communication module;

[0029] The edge computing module is used to periodically calibrate the reference length based on the chain link positioning data, and analyze the wear trend by calculating the cumulative elongation rate of multiple chain links; and to trigger the early warning actuator to perform corresponding early warning operations based on the cumulative elongation rate analysis results.

[0030] The communication module is used to synchronously transmit the early warning information to the back-end system.

[0031] Preferably, the process by which the edge computing module calculates the cumulative elongation of multiple chain links includes:

[0032] Determine the initial reference length; wherein, after each measurement of a detection unit is completed, the average pitch of the chain links of the previous detection unit is used as the reference length of the next detection unit; one detection unit contains N consecutive chain links;

[0033] Based on the chain link positioning data, determine the chain link spacing of each chain link in the current detection unit;

[0034] Calculate the elongation rate of a single link based on the link spacing of each link in the current detection unit;

[0035] Determine the attenuation weight of each link in the chain;

[0036] Based on the elongation rate of each link and the attenuation weight of each link, the weighted average of the elongation rate of the current detection unit is calculated to obtain the cumulative elongation rate.

[0037] As described above, the visual measurement device for chain link wear elongation provided by this invention includes a support assembly, an environmental sensing system, an adaptive supplementary lighting system, a visual measurement system, a health diagnosis system, and an industrial control computer. The support assembly is installed on the tail coal retaining plate of the scraper conveyor. The environmental sensing system, adaptive supplementary lighting system, and visual measurement system are all installed on the support assembly and electrically connected to the industrial control computer. The health diagnosis system is deployed within the industrial control computer. The environmental sensing system can collect environmental parameters such as light intensity, temperature, and dust concentration in the coal mine where the scraper conveyor is located in real time and upload them to the industrial control computer. The industrial control computer can determine the environmental state based on the uploaded environmental parameters and adjust the supplementary lighting mode and power of the adaptive supplementary lighting system accordingly, thereby adjusting the current environmental state in the coal mine. The visual measurement system can acquire visual images of the chain link and upload the acquired image data to the industrial control computer, which processes the image data to obtain chain link positioning data. The health diagnostic system can analyze and judge the wear status of chain links based on the cumulative elongation rate analysis principle of multiple chain links, using chain link positioning data, and execute early warning operations based on the analysis results. Therefore, this solution can collect environmental parameters such as light intensity, temperature, and dust concentration in coal mines through an environmental sensing system, and adjust the environmental conditions in the coal mine through an adaptive supplementary lighting system based on the environmental conditions determined by the collection results. This allows for dynamic tracking of the complex environment in coal mines, achieving image acquisition and recognition in low-light, high-dust environments, thereby improving the measurement accuracy of chain link wear elongation rate. Moreover, this solution analyzes and judges the wear status of chain links based on the cumulative elongation rate analysis principle of multiple chain links, fully considering the cumulative effect of chain link wear, and has higher measurement accuracy compared to single-link measurement methods. Attached Figure Description

[0038] Figure 1 This is a schematic diagram of a visual measurement device for chain link wear elongation provided in an embodiment of the present invention.

[0039] Figure 2 This is a schematic diagram of a support component provided in an embodiment of the present invention.

[0040] Figure 3 This is a schematic diagram of an environmental sensing system provided in an embodiment of the present invention.

[0041] Figure 4 This is a schematic diagram of an adaptive supplementary lighting system provided in an embodiment of the present invention.

[0042] Figure 5 This is a schematic diagram of a vision measurement system provided in an embodiment of the present invention.

[0043] Figure 6This is a schematic diagram of a health diagnosis system provided in an embodiment of the present invention.

[0044] Figure 7 This is a schematic diagram of an industrial control computer provided in an embodiment of the present invention.

[0045] The diagram shows: Support component 10, support frame fixing plate 11, camera vertical support frame 12, camera horizontal support frame 13, shock absorber 14, universal joint 15, camera mounting bracket 16, environmental sensing system 20, light sensor 21, dust sensor 22, temperature sensor 23, adaptive lighting system 30, polarized LED module 31, near-infrared pulse module 32, laser scanning module 33, drive controller 34, vision measurement system 40, explosion-proof industrial camera 41, encoder 42, health diagnosis system 50, edge computing module 51, early warning actuator 52, communication module 53, industrial control computer 60, filtering module 61, data normalization module 62, and environmental comprehensive scoring calculation module 63. Detailed Implementation

[0046] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0047] like Figure 1 As shown, the present invention provides a visual measurement device for chain link wear elongation rate, comprising: a support assembly 10, an environmental sensing system 20, an adaptive supplementary lighting system 30, a visual measurement system 40, a health diagnosis system 50, and an industrial control computer 60; the support assembly 10 is installed on the tail coal retaining plate of the scraper conveyor, the environmental sensing system 20, the adaptive supplementary lighting system 30, and the visual measurement system 40 are installed on the support assembly 10, and are all electrically connected to the industrial control computer 60; the health diagnosis system 50 is deployed in the industrial control computer 60;

[0048] The environmental sensing system 20 is used to collect environmental parameters of the coal mine where the scraper conveyor is located in real time, and to upload the collected environmental parameters to the industrial control computer 60; wherein, the environmental parameters include ambient light intensity, ambient temperature and dust concentration.

[0049] The industrial control computer 60 is used to determine the environmental state based on the uploaded environmental parameters, and to switch the supplementary light mode and adjust the supplementary light power of the adaptive supplementary light system 30 according to the environmental state, so as to adjust the current environmental state in the coal mine.

[0050] The vision measurement system 40 is used to acquire visual images of the chain link and upload the acquired image data to the industrial control computer 60, so that the industrial control computer 60 can process the image data to obtain the chain link positioning data.

[0051] The health diagnostic system 50 is used to analyze and judge the wear status of the chain links based on the chain link positioning data and the cumulative elongation rate analysis principle of multiple chain links, and to execute early warning operations based on the analysis and judgment results.

[0052] In this embodiment, the environmental sensing system 20 collects environmental parameters such as light intensity, temperature, and dust concentration in the coal mine for visual acquisition. Based on the environmental conditions determined by the acquisition results, the adaptive supplementary lighting system 30 adjusts the environmental conditions in the coal mine. This allows for dynamic tracking of the complex environment in the coal mine, enabling image acquisition and recognition in low-light, high-dust environments, thereby improving the measurement accuracy of chain link wear elongation rate. Furthermore, this solution analyzes and judges the wear state of the chain links based on the cumulative elongation rate analysis principle of multiple chain links, fully considering the cumulative effect of chain link wear, and has higher measurement accuracy compared to single-link measurement methods.

[0053] For support component 10, such as Figure 2 As shown, it may specifically include: a support frame fixing plate 11, a camera vertical support frame 12, a camera horizontal support frame 13, a shock absorber 14, a universal joint 15, and a camera mounting bracket 16; the support frame fixing plate 11 has a U-shaped structure and is installed on the coal retaining plate at the tail of the scraper conveyor; one end of the camera horizontal support frame 13 is fixedly installed on the outside of the support frame fixing plate 11, and the camera vertical support frame 12 is installed on the camera horizontal support frame 13 in a vertically intersecting manner and can be slidably adjusted in the horizontal direction; the upper end of the shock absorber 14 is installed on the camera vertical support frame 12 and can be slidably adjusted in the vertical direction; the upper end of the universal joint 15 is installed and connected to the lower end of the shock absorber 14, and the lower end is installed and connected to the upper end of the camera mounting bracket 16; the environmental perception system 20 and the adaptive supplementary lighting system 30 are installed on the camera horizontal support frame 13, and the visual measurement system 40 is installed on the camera mounting bracket 16.

[0054] In this embodiment, the environmental perception system 20 and the adaptive lighting system 30 are mounted on the upper camera horizontal support frame 13, while the visual measurement system 40 is mounted on the lower camera mounting frame 16. This arrangement allows the visual measurement system 40 to be closer to the chain links, thereby improving the accuracy and clarity of image data acquisition and further enhancing the accuracy of chain link wear elongation measurement. The adaptive lighting system 30, located above, can better supplement the lighting of the visual measurement system 40 according to environmental conditions, improving the measurement accuracy of the visual measurement system 40.

[0055] In one embodiment, the support frame fixing plate 11 can be made of Q235 steel plate with a thickness of 12mm. Its U-shaped structure can be clipped onto the coal retaining plate at the tail of the scraper conveyor, and it is fixed to the pre-set mounting holes by four sets of M16 expansion bolts. A 5mm thick rubber gasket can also be added to the contact surface between the support frame fixing plate 11 and the coal retaining plate to reduce vibration transmission. The camera horizontal support frame 13 can be made of aluminum alloy profile, with one end bolted to the support frame fixing plate 11 via a flange. The camera vertical support frame 12 is fixed to the camera horizontal support frame 13 by right-angle connectors, and the camera vertical support frame 12 can be adjusted horizontally along the camera horizontal support frame 13 to adapt to different detection positions. Simultaneously, the camera vertical support frame 12 can also slide vertically relative to the camera horizontal support frame 13 to adapt to different detection spaces. For example, sliding grooves can be provided on both the camera horizontal support frame 13 and the camera vertical support frame 12, allowing for position adjustment along the grooves. After the position is adjusted, it can be fixed with fasteners to prevent position changes during detection, which could affect measurement accuracy. The shock absorber 14 can be a hydraulic shock absorber with a rated load of 50 kg and an adjustable damping coefficient. It is mounted on the camera vertical support frame 12 to directly absorb equipment vibration. The universal joint 15 can be a ball joint type with a rotation angle range of 30 degrees. One end is connected to the shock absorber 14, and the other end is connected to the camera mounting bracket 16. It is used to fine-tune the camera angle to compensate for installation deviations. Each device in the environmental perception system 20, adaptive lighting system 30, and vision measurement system 40 can be fixed to its corresponding position using bolts.

[0056] The environmental sensing system 20 may specifically include a light sensor 21, a dust sensor 22, and a temperature sensor 23. The light sensor 21 is used to collect the ambient light intensity at the current location in the coal mine. The dust sensor 22 is used to capture the concentration of particulate matter in the air. The temperature sensor 23 is used to detect the ambient temperature at the current location in the coal mine. The light sensor 21 can be a digital light sensor with a measurement range of 0-10000 lux and an accuracy of ±5%. It is mounted on the camera's horizontal support frame 13, facing the chain running area to avoid direct interference from supplementary lighting. The dust sensor 22 can be a laser scattering sensor with a measurement range of 0-1000 μg / m³ and a response time ≤10 ms. It is mounted on the camera's horizontal support frame 13, with the air inlet facing 5 cm above the chain to avoid direct coal dust coverage. Temperature sensor 23 can be a PT100 platinum resistance sensor with a measurement range of -20℃ to 80℃ and an accuracy of ±0.5℃. It is mounted on the camera cross support 13 and close to the chain to monitor the ambient temperature.

[0057] In one embodiment, the industrial control computer 60 may be configured with a filtering processing module 61, a data normalization module 62, and an environmental comprehensive score calculation module 63;

[0058] The filtering module 61 is used to perform noise reduction and smoothing processing on the environmental parameters uploaded by the environmental perception system 20 through Kalman filtering to eliminate instantaneous interference and obtain filtered data.

[0059] The data normalization module 62 is used to normalize the ambient light intensity, ambient temperature and dust concentration in the filtered data to obtain normalized data.

[0060] The environmental comprehensive score calculation module 63 is used to perform weighted fusion calculation on the normalized values ​​of light intensity, dust concentration and temperature in the normalized data to obtain the environmental comprehensive score, and then send the comprehensive score to the adaptive supplementary lighting system 30.

[0061] In this embodiment, the system starts up immediately after power-on and continuously collects environmental parameters at 50ms intervals. Specifically, light sensor 21 monitors ambient light intensity, dust sensor 22 captures the concentration of particulate matter in the air, and temperature sensor 23 records the ambient temperature. The raw data collected by the sensors is first processed by Kalman filtering for noise reduction and smoothing to eliminate transient interference. Subsequently, the processed data is input into a weighted fusion model to calculate the comprehensive environmental score Es. This model first normalizes each parameter: the normalized value of light intensity L... norm = (L - L min ) / (L max - L min ), where L min and L max The preset minimum and maximum light intensity; the normalized dust concentration value D. norm = (D - D min ) / (D max -D min ), where D min and D max The preset minimum and maximum dust concentrations; temperature normalization value T norm = (T - T min ) / (T max - T min ), where T min and T max These are the preset minimum and maximum temperatures. The formula for calculating the comprehensive environmental score Es is: ,in, For comprehensive environmental scoring, , , These are the weighted values ​​for ambient light intensity, dust concentration, and ambient temperature, respectively, and the sum of the weighted values ​​is 1. This is the normalized value of light intensity. This is the normalized value of dust concentration. This is the normalized temperature value. (1 - D) norm ) and (1 - T norm The value of 0 indicates that dust concentration and temperature have a negative impact on system operation. The environmental comprehensive score Es ranges from 0 to 1, with a higher value indicating better environmental conditions. As a core decision-making indicator, it provides a direct and quantitative basis for the mode selection and power adjustment of the adaptive supplementary lighting system 30. A high score indicates a good environment with low supplementary lighting requirements; a low score indicates a harsh environment (such as low illuminance and high dust) requiring enhanced supplementary lighting.

[0062] The adaptive lighting system 30 may include: a polarized LED module 31, a near-infrared pulse module 32, a laser scanning module 33, and a drive controller 34;

[0063] The drive controller 34 is used to receive the comprehensive environmental score and analyze and judge the comprehensive environmental score; if the comprehensive environmental score is not less than a first preset score threshold, the polarized LED module 31 is activated for conventional supplemental lighting; if the comprehensive environmental score is less than the first preset score threshold but not less than a second preset score threshold, the near-infrared pulse module 32 is activated for supplemental lighting; if the comprehensive environmental score is less than the second preset score threshold, the laser scanning module 33 is activated for supplemental lighting; wherein, the first preset score threshold is greater than the second preset score threshold. Furthermore, the drive controller 34 is also used to increase the supplemental lighting power by the first preset threshold as a basis, based on the upper limit of the comprehensive environmental score for each supplemental lighting mode, when the comprehensive environmental score decreases by a third preset score threshold.

[0064] The polarization LED module 31 can consist of eight high-brightness LEDs with a wavelength of 550nm and adjustable power from 0-20W. It is mounted on both sides of the camera lens, with the polarization direction aligned with the lens polarizer to reduce glare. The near-infrared pulse module 32 can use an 850nm near-infrared LED with an adjustable pulse frequency of 100-1000Hz, a pulse width of 10-100μs, and a power of 0-30W. It is mounted on the front of the camera's horizontal support frame 13, with an illumination angle of 30°, suitable for high-speed motion scenarios. The laser scanning module 33 can use a 650nm semiconductor laser with an adjustable power of 0-50W, a scanning frequency of 50Hz, and a scanning angle of ±10°. It focuses the beam through an optical lens to enhance penetration in dusty environments. The drive controller 34 can use an STM32 microcontroller as the control core, integrating a PWM power adjustment module. It communicates with the industrial computer 60 via a CAN bus, with a response time ≤20ms.

[0065] In this embodiment, the adaptive supplementary lighting system 30 includes three supplementary lighting devices: a polarized LED module 31, a near-infrared pulse module 32, and a laser scanning module 33. Mode switching and power adjustment are achieved by a drive controller 34. This system is deeply integrated with the environmental perception system 20, constructing a decision-making mechanism based on a comprehensive environmental score Es and specific sensor thresholds. Its supplementary lighting mode is intelligently selected according to Es: when Es ≥ 0.7, the environment is good, and the polarized LED module 31 is activated for regular supplementary lighting; when 0.5 ≤ Es < 0.7, the environment is moderate, and the near-infrared pulse module 32 is activated to cope with low illumination or high chain speeds; when Es < 0.5, the environment is harsh, and the laser scanning module 33 with strong penetration is activated. Regarding the adjustment of supplementary lighting intensity, the drive controller 34 sets the initial power based on the upper limit of the comprehensive environmental score for each supplementary lighting mode. Within each supplementary lighting mode, for every 0.1 decrease in Es compared to the benchmark for that mode, the supplementary lighting power increases by 10%. Simultaneously, the system receives image contrast data from the vision measurement system 40 in real time. If the contrast is below a set threshold, the drive current is increased; otherwise, it is dynamically reduced, thus forming a closed-loop control of environmental perception, supplementary lighting adjustment, and image feedback. Furthermore, the system also includes a special scene handling mechanism: when the dust concentration is higher than D... max 80% of the temperature is forcibly switched to laser scanning mode; when the temperature is higher than T max When the contrast ratio reaches 80%, the supplementary light power is reduced, for example, by 30%, to prevent overheating and simultaneously simplify the vision algorithm, thereby prioritizing system stability and safety under extreme conditions. Furthermore, intensity closed-loop adjustment can be set. For example, the vision measurement system 40 provides real-time feedback on image contrast (target value 120-180). If the contrast ratio is <120, the drive controller 34 increases the PWM duty cycle, increasing power by 10% for every 10 units decrease in contrast ratio; if the contrast ratio is >180, the PWM duty cycle is reduced to ensure clear imaging without overexposure.

[0066] The vision measurement system 40 may include: an explosion-proof industrial camera 41 and an encoder 42 electrically connected thereto;

[0067] The explosion-proof industrial camera 41 is equipped with a dirt-resistant coated lens to reduce the interference of dust and oil on imaging.

[0068] The encoder 42 is used to capture chain displacement signals in real time, and when the chain link moves to the vicinity of the center of the field of view of the explosion-proof industrial camera 41, it triggers the explosion-proof industrial camera 41 to acquire images, and the explosion-proof industrial camera 41 uploads the acquired images to the industrial control computer 60.

[0069] The industrial control computer 60 is also used to quickly locate chain links and determine the ROI region based on the YOLOv7-tiny model, and to locate chain links through an edge detection algorithm.

[0070] In this embodiment, the vision measurement system 40 uses an explosion-proof industrial camera 41 as the core acquisition device, and is equipped with an anti-fouling coated lens to reduce the interference of dust and oil on imaging. The encoder 42 captures the chain displacement signal in real time to achieve precise synchronization of measurement triggering. Its core function is to complete the acquisition and processing of chain link images. When encoder 42 detects that the chain link has moved to the vicinity of the center of the field of view, it triggers the industrial camera to acquire images, and the exposure time is dynamically adjusted according to the chain speed to ensure image clarity. In the chain link positioning and enhancement stage, the first frame uses the YOLOv7-tiny model to quickly locate the chain link and determine the ROI region. Subsequent frames use optical flow to track displacement to compensate for the deviation caused by chain swaying. At the same time, multi-scale algorithms are used to enhance image contrast and improve texture details. In the contour detection and center positioning stage, the inner hole contour of the chain link is extracted and oil artifacts are removed by improving the image processing algorithm of Canny operator and morphological reconstruction. Then, the constrained RANSAC algorithm is used to fit the contour points into an elliptical model. Finally, sub-pixel compensation is performed by combining image gradient information and ellipse fitting residuals, which greatly improves the center positioning accuracy in high dust and oil environments and provides accurate image data support for wear calculation.

[0071] Specifically, the explosion-proof industrial camera 41 can be a 1.3-megapixel CMOS camera with a resolution of 1280×1024, a frame rate of ≥30fps, and an explosion-proof rating of ExdIICT6. It is mounted on the camera mounting bracket 16, with the lens facing the chain running direction and a field of view of 100×80mm. The anti-fouling coated lens can be an 8mm fixed-focus lens with a magnesium fluoride coating, which is waterproof and oil-proof, has a fixed focal length, and a depth of field of 50-200mm, ensuring clear focus throughout the chain's operation. The encoder 42 can be an incremental encoder with a resolution of 1000 lines, mounted on the driven wheel shaft of the scraper conveyor and fixed by a coupling. It collects chain displacement signals in real time; for example, it outputs 1000 pulses per revolution, corresponding to a chain movement of 1 pitch.

[0072] During image acquisition and processing, encoder 42 sends a trigger signal to the camera every 100 pulses, and the camera simultaneously acquires images. The exposure time is dynamically adjusted according to the chain speed; for example, the exposure time is 100μs when the chain speed is 1m / s, and 30μs when the chain speed is 3m / s. When performing adaptive ROI extraction, industrial computer 60 uses the YOLOv7-tiny model to detect the first frame image and outputs the chain loop target bounding box, such as the coordinate format x1, y1, x2, y2, which is automatically cropped into the ROI region, replacing manual annotation. For subsequent frame tracking, the LK optical flow method is used to calculate the displacement of feature points within the ROI region of the previous frame, such as selecting 50 corner points. The coordinates of the ROI region in the current frame are updated in real time to compensate for chain sway; for example, if the sway amplitude is ≤10mm, the tracking error is ≤1 pixel. During multi-scale enhancement, the ROI region image undergoes three levels of scale transformation (1×, 0.8×, 0.6×), and the contrast is enhanced before fusion to improve the distinction between the chain loop and the background. For contour extraction, an improved Canny operator is used to extract the initial contour. Morphological reconstruction is then used to remove oil and dust artifacts, preserving the true contour of the chain link's inner hole. For center location, a constrained RANSAC algorithm is employed, randomly selecting five contour points to fit an ellipse, iterating 1000 times, with an interior point threshold of 2 pixels to eliminate the influence of abnormal contour points. Furthermore, based on image gradient information, sub-pixel interpolation is performed on the ellipse center coordinates, achieving a positioning accuracy of 0.1 pixels.

[0073] The health diagnostic system 50 may include: an edge computing module 51, an early warning actuator 52, and a communication module 53;

[0074] The edge computing module 51 is used to periodically calibrate the reference length based on the chain link positioning data, and analyze the wear trend by calculating the cumulative elongation rate of multiple chain links; and to trigger the early warning actuator 52 to perform corresponding early warning operations based on the cumulative elongation rate analysis results.

[0075] The communication module 53 is used to synchronously transmit the early warning information to the background system.

[0076] The process by which the edge computing module 51 calculates the cumulative elongation of multiple chain links may include:

[0077] Determine the initial reference length; wherein, after each measurement of a detection unit is completed, the average pitch of the chain links of the previous detection unit is used as the reference length of the next detection unit; one detection unit contains N consecutive chain links;

[0078] Based on the chain link positioning data, determine the chain link spacing of each chain link in the current detection unit;

[0079] Calculate the elongation rate of a single link based on the link spacing of each link in the current detection unit;

[0080] Determine the attenuation weight of each link in the chain;

[0081] Based on the elongation rate of each link and the attenuation weight of each link, the weighted average of the elongation rate of the current detection unit is calculated to obtain the cumulative elongation rate.

[0082] In this embodiment, the health diagnosis system 50 integrates an edge computing module 51, an early warning actuator 52, and a communication module 53, undertaking the functions of analyzing, judging, and issuing early warnings regarding the wear status of chain links. The edge computing module 51 is responsible for core algorithm calculations, periodically calibrating the reference length based on the chain link positioning data output by the vision measurement system 40, and analyzing wear trends by calculating the cumulative elongation rate of multiple chain links. The system selects 10 consecutive chain links as a detection unit, assigning a decay weight to the elongation rate of each link; the closer the link is to the current detection point, the higher the weight. This weighted average cumulative elongation rate is then calculated, and this model better reflects the overall and recent wear trends of the chain. Simultaneously, the system dynamically updates the reference length: the average pitch calculated by the previous detection unit is used as the reference for the next detection unit, forming a sliding window to eliminate system errors and track changes in the overall chain status. Based on the value and changes in the cumulative elongation rate, the early warning actuator 52 is triggered to perform different levels of early warning operations, including audible and visual alarms, speed reduction control, and emergency shutdown, and the early warning information is synchronously transmitted to the backend system via the communication module 53.

[0083] In addition, the system can activate emergency handling mechanisms for extreme conditions such as excessively high dust concentration, power fluctuations, and communication interruptions. These mechanisms include switching adaptation algorithms, activating supercapacitors to maintain operation, and storing raw data locally. Simultaneously, it achieves coordinated response with other systems through subsystem linkage mechanisms. For example, when the deviation of a single link is large, the reference calibration cycle is shortened; when the temperature is too high, the adaptive supplementary lighting system is linked to reduce power and simplify the vision algorithm, ensuring the safety and data integrity of the system under fault conditions.

[0084] The edge computing module 51 can use an NVIDIA Jetson Nano development board, integrating a GPU acceleration module to run wear analysis algorithms with a computation latency of ≤100ms. The early warning actuator 52 may include an audible and visual alarm (≥85dB, red light flashing frequency 2Hz) and a relay module (to control the conveyor to slow down or stop), installed in the conveyor control cabinet. The communication module 53 can adopt a dual backup of 5G + wired Ethernet; the 5G module supports industrial-grade communication (latency ≤50ms), and the Ethernet transmission rate is 100Mbps, transmitting early warning information to the coal mine monitoring center.

[0085] During the initial installation and commissioning of the system, 10 brand-new, unworn chain links or manually calibrated standard chain links are selected. The actual pitch of each chain link is collected using a vision measurement system 40, and the average pitch of the 10 links is calculated as the initial reference length L. bench0 ;in, , This is the pitch measurement value of the i-th new chain link. Furthermore, after completing the measurement of each detection unit (10 chain links), the average pitch of the 10 chain links of the previous detection unit is used as the reference length Lbench of the next detection unit, forming a "sliding reference" to eliminate the influence of the overall chain elongation on the measurement of a single link.

[0086] Furthermore, when determining the pitch measurement value of a single chain link, encoder 42 first triggers the camera. When the first chain link enters the center of the field of view, three frames are acquired (to avoid single-frame blurring errors), and the frame with the highest clarity is selected for calculation. Then, the coordinates of the center O1 of the inner hole of the chain link are located using the Canny operator, morphological reconstruction, constrained RANSAC elliptic fitting, and subpixel compensation. Similarly, the coordinates of the center O2 of the inner hole of the second chain link are located. Based on the pixel-to-physical-size conversion relationship, the distance between the two centers is calculated, which is the pitch L1 of the first chain link. Repeating the above process sequentially obtains the pitches L1, L2, ..., L10 of the 10 chain links within the current detection unit.

[0087] Further, the elongation of a single chain link is calculated. The specific formula used is... Calculate the elongation of the i-th link. Then, assign weights to each link according to their distance from the current monitoring point. For example, as L1, L2, ..., L10 approach the current monitoring point, their corresponding weights are... The values ​​are 0.1, 0.2, ..., 1.0 in sequence. Then, the formula is used... The weighted elongation ratio can then be calculated. Further dividing this weighted elongation ratio by the sum of the weights (5.5) yields the cumulative elongation ratio. .

[0088] When issuing early warnings, a three-level early warning mechanism can be adopted, specifically as follows:

[0089] Yellow alert: If the percentage is ≥1% and <3%, an audible and visual alarm will be triggered, the green light will flash, the buzzer will sound intermittently, and a warning message will be sent to the monitoring center.

[0090] Red alert + speed reduction: For values ​​≥3% and <5%, the audible and visual alarm will have a constantly lit red light and a continuous buzzer, and the relay will control the conveyor to slow down by 50% while simultaneously uploading data.

[0091] Emergency shutdown + notification: If the value is ≥5%, the relay will cut off the power to the conveyor, the audible and visual alarm will continue to sound, and a text message notification will be sent to the manager's mobile phone via the 5G module.

[0092] Emergency handling measures can also be considered. For example, if the dust concentration is too high (D≥400μg / m³), switch to a simplified contour detection algorithm, skipping the morphological reconstruction step to improve processing speed. During power fluctuations (voltage ±10%), activate a supercapacitor (1000F) to maintain system operation for ≥5 minutes to ensure data storage. In the event of communication interruption, store the original image and computational data on a local SD card (32GB capacity), and automatically upload them after communication is restored.

[0093] The process flow of the visual measurement device for chain link wear elongation provided by this invention may include: after the system is powered on, the environmental sensing system 20 starts first and runs continuously with a period of 50ms. The light sensor 21 monitors the ambient light conditions, the dust sensor 22 monitors the particulate matter concentration, and the temperature sensor 23 measures the ambient temperature. After Kalman filtering and noise reduction, the data from these sensors are used to calculate the overall environmental score.

[0094] Next, the adaptive supplementary lighting decision-making and execution phase begins. When the overall score, illuminance, dust concentration, or temperature reaches specific conditions, the system will activate the laser scanning module 33, the near-infrared pulse module 32, or the polarized LED module 31 accordingly, based on factors such as temperature, dust concentration, and chain speed. Simultaneously, the supplementary lighting system will adjust based on the image contrast; if the contrast is low, the drive current will be increased.

[0095] The next stage is the visual measurement triggering phase. Encoder 42 monitors the chain displacement in real time. When a chain link moves to near the center of the field of view, the industrial camera acquires an image, and the exposure time is dynamically adjusted according to the chain speed. In the chain link positioning and enhancement phase, the first frame uses the YOLOv7-tiny model to locate the chain link and determine the ROI region. Subsequent frames use optical flow to track displacement and compensate for swaying. Then, a multi-scale algorithm is used to enhance the image and improve texture details. Next, contour detection and center positioning are performed, i.e., the center of the inner hole of the chain link is determined by ellipse fitting. Finally, sub-pixel compensation is performed based on relevant factors to improve positioning accuracy and solve the problem of large positioning accuracy error of the chain link center in high dust and oil environments.

[0096] Once the system enters the health diagnosis and early warning phase, it periodically calibrates the baseline length and calculates the cumulative elongation rate of multiple chain links. Based on the elongation rate, it then executes different levels of early warnings, including audible and visual alarms, speed reduction, emergency shutdown, and notifications. For extreme operating conditions, the system has corresponding handling measures: switching algorithms when dust concentration is too high, maintaining operation via supercapacitors during power fluctuations, and storing data locally when communication is interrupted.

[0097] After completing one processing step, the system immediately returns to the environmental perception step, forming a closed-loop detection flow. Key linkage mechanisms exist throughout the process, such as forcibly switching the supplementary lighting mode when dust concentration changes abruptly, shortening the baseline calibration cycle when a single link in the chain has a large deviation, and reducing supplementary lighting power and activating a simplified visual algorithm when the temperature is too high. Simultaneously, the system ensures real-time performance through strict timing control and hardware synchronization, prioritizes tasks in resource management, and switches to preset modes or triggers self-checks in case of faults.

[0098] In summary, the visual measurement device for chain link wear elongation provided by the present invention has at least the following beneficial effects:

[0099] (1) Adaptive and stable mechanical structure design: Through the combination of support frame fixing plate 11, shock absorber 14 and universal joint 15, the monitoring equipment can be stably installed in the high vibration environment of scraper conveyor, effectively offsetting machine vibration interference and solving the problem of image blurring caused by vibration in traditional fixed installation methods.

[0100] (2) The ROI region adaptive extraction mechanism based on YOLOv7-tiny model first frame localization and optical flow tracking can compensate for chain swing in real time, maintain stable tracking in high-speed motion scenarios, and respond quickly, solving the problem that traditional manual annotation or static box selection cannot adapt to dynamic working conditions.

[0101] (3) An anti-interference contour detection method using an improved Canny operator, morphological reconstruction, and constrained RANSAC ellipse fitting, combined with sub-pixel compensation technology, effectively overcomes the interference of high dust and oil pollution environments, and the chain link center positioning error is effectively controlled, providing accurate data support for the calculation of cumulative elongation.

[0102] (4) The innovative 10-link cumulative elongation rate early warning model introduces the attenuation weight algorithm and dynamic benchmark update mechanism to comprehensively reflect the overall wear trend of the chain and reduce the number of missed reports; the three-level early warning mechanism combined with multi-frame verification logic reduces false reports and effectively avoids misoperation caused by accidental deviation of a single link.

[0103] (5) The cross-system linkage mechanism of environment, supplementary lighting and vision can dynamically switch the supplementary lighting mode according to the real-time monitored illuminance, dust and temperature parameters, such as polarized LED / near-infrared pulse / laser scanning. The closed-loop adjustment of image contrast ensures stable imaging quality under extreme working conditions such as low illuminance and high dust, thus expanding the system's adaptability.

[0104] The present invention also provides a computing device, including a memory and a processor, wherein the memory stores executable code, and when the processor executes the executable code, it runs the method as described in the above embodiments.

[0105] The present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed in a computer, causes the computer to perform the method as described in any of the above embodiments.

[0106] The modules or units in the device of this invention can be merged, divided, and deleted according to actual needs. The above-disclosed embodiments are merely preferred embodiments of the present invention and should not be construed as limiting the scope of the invention. Those skilled in the art will understand that implementing all or part of the processes of the above embodiments and making equivalent changes according to the claims of this invention still fall within the scope of the invention.

Claims

1. A visual measurement device for chain link wear elongation, characterized in that, include: The system includes a support component, an environmental sensing system, an adaptive lighting system, a visual measurement system, a health diagnostic system, and an industrial control computer. The support component is mounted on the coal retaining plate at the tail of the scraper conveyor. The environmental sensing system, adaptive lighting system, and visual measurement system are mounted on the support component and are all electrically connected to the industrial control computer. The health diagnostic system is integrated into the industrial control computer. The environmental sensing system is used to collect environmental parameters of the coal mine where the scraper conveyor is located in real time, and to upload the collected environmental parameters to the industrial control computer; wherein, the environmental parameters include ambient light intensity, ambient temperature and dust concentration. The industrial control computer is used to determine the environmental state based on the uploaded environmental parameters, and to switch the supplementary lighting mode and adjust the supplementary lighting power of the adaptive supplementary lighting system according to the environmental state, so as to adjust the current environmental state in the coal mine. The vision measurement system is used to acquire visual images of the chain links and upload the acquired image data to an industrial control computer, which then processes the image data to obtain chain link positioning data. The health diagnostic system is used to analyze and judge the wear status of the chain links based on the chain link positioning data and the cumulative elongation rate analysis principle of multiple chain links, and to execute early warning operations based on the analysis and judgment results.

2. The visual measurement device for chain link wear elongation according to claim 1, characterized in that, The support assembly includes: a support frame fixing plate, a camera vertical support frame, a camera horizontal support frame, a shock absorber, a universal joint, and a camera mounting bracket; the support frame fixing plate has a U-shaped structure and is installed on the tail coal retaining plate of the scraper conveyor; one end of the camera horizontal support frame is fixedly installed on the outside of the support frame fixing plate, and the camera vertical support frame is installed on the camera horizontal support frame in a vertically intersecting manner and is adjustable in the horizontal direction; the upper end of the shock absorber is installed on the camera vertical support frame and is adjustable in the vertical direction; the upper end of the universal joint is installed and connected to the lower end of the shock absorber, and the lower end is installed and connected to the upper end of the camera mounting bracket; the environmental perception system and the adaptive lighting system are installed on the camera horizontal support frame, and the visual measurement system is installed on the camera mounting bracket.

3. The visual measurement device for chain link wear elongation according to claim 1, characterized in that, The environmental sensing system includes: a light sensor, a dust sensor, and a temperature sensor; The light sensor is used to collect the ambient light intensity at the current location of the coal mine. The dust sensor is used to detect the concentration of particulate matter in the air; The temperature sensor is used to detect the ambient temperature at the current location in the coal mine.

4. The visual measurement device for chain link wear elongation according to claim 3, characterized in that, The industrial control computer is equipped with a filtering module, a data normalization module, and an environmental comprehensive score calculation module. The filtering module is used to perform noise reduction and smoothing processing on the environmental parameters uploaded by the environmental perception system using Kalman filtering to eliminate instantaneous interference and obtain filtered data. The data normalization module is used to normalize the ambient light intensity, ambient temperature and dust concentration in the filtered data to obtain normalized data. The environmental comprehensive score calculation module is used to perform weighted fusion calculation on the normalized values ​​of light intensity, dust concentration and temperature in the normalized data to obtain the environmental comprehensive score, and then send the comprehensive score to the adaptive supplementary lighting system.

5. The visual measurement device for chain link wear elongation according to claim 4, characterized in that, The environmental comprehensive score calculation module calculates the environmental comprehensive score based on the following formula: ； in, For comprehensive environmental scoring, , , These are the weighted values ​​for ambient light intensity, dust concentration, and ambient temperature, respectively. This is the normalized value of light intensity. This is the normalized value of dust concentration. This is the normalized temperature value.

6. The visual measurement device for chain link wear elongation according to claim 4, characterized in that, The adaptive supplemental lighting system includes: a polarized LED module, a near-infrared pulse module, a laser scanning module, and a drive controller; The drive controller is used to receive the comprehensive environmental score and analyze and judge the comprehensive environmental score; if the comprehensive environmental score is not less than a first preset score threshold, the polarized LED module is activated for conventional supplemental lighting; if the comprehensive environmental score is less than the first preset score threshold but not less than a second preset score threshold, the near-infrared pulse module is activated for supplemental lighting; if the comprehensive environmental score is less than the second preset score threshold, the laser scanning module is activated for supplemental lighting; wherein, the first preset score threshold is greater than the second preset score threshold.

7. The visual measurement device for chain link wear elongation according to claim 6, characterized in that, The drive controller is also used to increase the supplementary light power by a first preset threshold when the comprehensive environmental score decreases by a third preset threshold, based on the upper limit of the comprehensive environmental score for each supplementary light mode.

8. The visual measurement device for chain link wear elongation according to claim 1, characterized in that, The vision measurement system includes: an explosion-proof industrial camera and an encoder electrically connected to it; The explosion-proof industrial camera is equipped with a lens with an anti-fouling coating to reduce the interference of dust and oil on imaging. The encoder is used to capture chain displacement signals in real time, and when the chain link moves to the vicinity of the center of the field of view of the explosion-proof industrial camera, it triggers the explosion-proof industrial camera to acquire images, and the explosion-proof industrial camera uploads the acquired images to the industrial control computer. The industrial control computer is also used to quickly locate chain links and determine the ROI region based on the YOLOv7-tiny model, and to locate chain links through an edge detection algorithm.

9. The visual measurement device for chain link wear elongation according to claim 1, characterized in that, The health diagnostic system includes: an edge computing module, an early warning actuator, and a communication module; The edge computing module is used to periodically calibrate the reference length based on the chain link positioning data, and analyze the wear trend by calculating the cumulative elongation rate of multiple chain links; and to trigger the early warning actuator to perform corresponding early warning operations based on the cumulative elongation rate analysis results. The communication module is used to synchronously transmit the early warning information to the back-end system.

10. The visual measurement device for chain link wear elongation according to claim 9, characterized in that, The process by which the edge computing module calculates the cumulative elongation of multiple chain links includes: Determine the initial reference length; wherein, after each measurement of a detection unit is completed, the average pitch of the chain links of the previous detection unit is used as the reference length of the next detection unit; one detection unit contains N consecutive chain links; Based on the chain link positioning data, determine the chain link spacing of each chain link in the current detection unit; Calculate the elongation rate of a single link based on the link spacing of each link in the current detection unit; Determine the attenuation weight of each link in the chain; Based on the elongation rate of each link and the attenuation weight of each link, the weighted average of the elongation rate of the current detection unit is calculated to obtain the cumulative elongation rate.

Images