Railway track fastener state detecting and fastening equipment

Abstract

The invention provides railway track fastener state detecting and fastening equipment. Comprising a working vehicle, a power system, a fastener cleaning device, a bolt fastening device, a telescopic pressing strip device, a sensor system, a data collecting and processing device, a control system, a data storage device and the like. The fastener cleaning device is used for cleaning a fastener bolt and an elastic strip. The data collecting and processing device obtains fastener state information through machine vision, and a convolutional neural network model is used for training and predicting the friction coefficient of a fastener bolt and a friction wheel in the bolt fastening device and obtaining the relative position relation between the fastener bolt and the fastener. The telescopic pressing strip device is used for reducing the torque required for fastening a bolt through a telescopic elastic strip; and the bolt fastening device is used for controlling and applying the bolt fastening torque by adjusting the distance between a friction wheel and a rolling wheel. Therefore, the state detection of the railway track fastener and the automatic fastening of the fastener are realized, and the manual and field operation procedures required by on-site fastener state detection and fastener fastening are greatly reduced.

Description

Technical Field

[0001] The present invention relates to the technical fields of railway track fastener detection, fastener fastening, and the like, and in particular to a railway track fastener state detection and fastening device. Background Art

[0002] In railway transportation operations, the fastener system is an important part of the railway rails. It plays the role of maintaining and adjusting the track gauge and track direction, providing longitudinal resistance, and preventing rail creep. Therefore, the fastener system is crucial to the safety of railway operations. However, with the increase in train speed and density, the phenomenon of bolt loosening in the fastener system is increasing, especially in heavy-duty railways. Due to the large vibration and impact of the train, the problem of bolt loosening occurs frequently. If the problem of loose bolts and over-tightening is not solved in time, it will cause the fasteners in adjacent areas to fail, change the geometric parameters of the track, thereby reducing the smoothness of the track and endangering driving safety. At present, in railway maintenance and repair operations, the tightening of railway fasteners is mainly done by manually pushing and operating special tightening wrenches, which has a low degree of automation, slow speed, and low efficiency. Therefore, it is very necessary to provide a highly efficient rail fastener status detection and tightening device. Summary of the invention

[0003] An embodiment of the present invention provides a railway track fastener state detection and fastening device, which is used to solve the problems existing in the prior art.

[0004] In order to achieve the above object, the present invention adopts the following technical scheme.

[0005] A railway track fastener state detection and fastening device, comprising a working vehicle, and a fastener cleaning device, a bolt fastening device, a telescopic layer device, a sensor system, a data acquisition and processing device, a control system and a data storage device arranged on the working vehicle;

[0006] The fastener cleaning device has a rotatable cleaning disk, which is located on the bottom surface of the working vehicle; the bolt fastening device has a friction wheel and a rolling wheel, which are arranged in pairs at intervals from each other and are located on the bottom surface of the working vehicle; the friction wheel is used to drive the bolts of the fastener, and the interval between the friction wheel and the rolling wheel is adjustable;

[0007] The retractable pressure strip device is located on the bottom surface of the working vehicle and has a retractable pressure strip, which is used to press the elastic strip of the fastener;

[0008] The data acquisition and processing device comprises an image acquisition module and a processor; the image acquisition module is used to obtain the appearance information of the fastener; the processor is used to obtain, by calculation according to the appearance information: the state of the fastener, the friction coefficient between the bolt of the fastener and the friction wheel, and the relative position between the fastener and the working vehicle;

[0009] The sensor system is used to monitor: vibration information of fastener cleaning device, bolt tightening device and telescopic strip device, temperature information of data acquisition and processing device, control system and data storage device;

[0010] The control system is respectively connected to the fastener cleaning device, the bolt tightening device, the telescopic pressure strip device, the sensor system, the data acquisition and processing device, the control system and the data storage device; the control system is used to: determine whether the railway track fastener status detection and fastening equipment has a fault based on the vibration information of the fastener cleaning device, the bolt tightening device and the telescopic pressure strip device, and the temperature information of the data acquisition and processing device, the control system and the data storage device; and adjust the spacing between the friction wheel and the rolling wheel based on the friction coefficient between the fastener bolt and the friction wheel calculated by the data acquisition and processing device.

[0011] Preferably, the fastener cleaning device has a fastener cleaning motor, which is drivingly connected to the cleaning disk.

[0012] Preferably, the bolt fastening device comprises:

[0013] Bolt the engine and install it on the work vehicle;

[0014] A drive shaft is detachably connected to the output end of the bolt-tightening engine, and the bolt-tightening engine is drivingly connected to the friction wheel through the drive shaft;

[0015] The shaft system structure includes a positioning rod, the positioning rod is detachably mounted on the bottom of the working vehicle, one end of the positioning rod is provided with a rolling bearing, and the rolling wheel is movably mounted on the one end of the positioning rod through the rolling bearing;

[0016] The transmission rod-sleeve structure has a pair of rings connected to each other through a tensioner, and the rings are respectively in contact with the drive shaft and the positioning rod; the tensioner is connected to the control system circuit and can adjust the spacing between the friction wheel and the rolling wheel by changing its own length.

[0017] Preferably, the bolt fastening device further comprises a friction wheel cleaning device detachably mounted on the outer side of the friction wheel.

[0018] Preferably, the telescopic layering device comprises:

[0019] A pair of first rail grippers, detachably mounted on the working vehicle body, the first rail grippers being used to clamp the rail;

[0020] The telescopic pressure strip is detachably connected to the first rail holding device; the first rail holding device has a window, and the telescopic pressure strip is located in the area of ​​the window;

[0021] A plurality of rollers are mounted on the inner side of the first rail clamp, and when the first rail clamp clamps the rail, the rollers are in contact with the rail surface;

[0022] A pressing drive motor is installed on the first rail holding device of the working vehicle body; an eccentric wheel is provided at the output end of the pressing drive motor, the position of the eccentric wheel corresponds to the telescopic pressure strip, and the pressing drive motor can drive the eccentric wheel to push the telescopic pressure strip out of the window to press the elastic strip of the fastener;

[0023] A pair of return springs are respectively connected to two ends of the telescopic pressure strip.

[0024] Preferably, the data acquisition and processing device calculates the relative position between the fastener and the work vehicle through the SURF algorithm, and also calculates the state of the fastener and the friction coefficient between the bolt of the fastener and the friction wheel through the convolutional neural network algorithm.

[0025] Preferably, the process of obtaining the friction coefficient between the bolt of the fastener and the friction wheel by calculation according to the apparent information comprises:

[0026] S1 preprocesses the first image, including normalization, color space conversion, noise removal and first image enhancement; the first image is an image of the bolt and the friction wheel of the fastener in contact with each other;

[0027] S2 constructs a convolutional network model; the convolutional network model includes an input layer, a convolutional layer, a pooling layer, a fully connected layer and an output layer arranged in sequence along the data flow direction; the input layer is used to set the property parameters of the bolt surface features of the preprocessed first image to the size of the input layer; there are multiple convolutional layers, which are used to extract local features of the bolt surface in the preprocessed first image; the pooling layer is used to reduce the size of the extracted local features of the bolt surface; the fully connected layer is used to map the local features of the bolt surface to the predicted value of the friction coefficient; the output layer is used to output the predicted friction coefficient value;

[0028] S3 uses the ReLU activation function to increase the nonlinear expression ability of the convolutional network model;

[0029] S4 measures the difference between the predicted value and the actual value of the convolutional network model through the mean absolute error loss function, and updates the parameters of the convolutional network model through the SGD optimizer;

[0030] S5 uses the dataset to train, test, and tune the convolutional network model;

[0031] S6 uses the tuned convolutional network model to predict the friction coefficient between the bolt of the fastener and the friction wheel, so that the control system can adjust the distance between the friction wheel and the rolling wheel based on the obtained friction coefficient.

[0032] Preferably, the process of calculating the relative position between the fastener and the work vehicle by using the SURF algorithm includes:

[0033] E1 uses the Hessian matrix to detect key points in the second image, and obtains the positions of potential key points in the second image by performing extreme value detection on the determinant of the Hessian matrix; the second image is an image of the fastener used to calculate the relative position between the fastener and the work vehicle;

[0034] E2 eliminates unstable or mismatched feature points by calculating the distance and angle difference between feature points in the second image;

[0035] E3 obtains a descriptor by calculation based on the feature points in the second image after executing step E2;

[0036] E3 normalizes the descriptor;

[0037] E4 obtains the Euclidean distance of the descriptor by calculation based on the normalized descriptor;

[0038] E5 screens and obtains feature points that match the random sampling consensus algorithm based on the ratio of the calculated Euclidean distance;

[0039] E6 estimates the homography matrix based on the feature points obtained by screening through a random sampling consistency algorithm;

[0040] E7 uses the image acquisition module to calibrate parameters;

[0041] Based on the calibrated parameters and the corresponding geometric correspondence, E8 calculates the three-dimensional coordinates of the same feature points in the second images of two adjacent frames, and calculates the relative position between the fastener and the work vehicle by comparing the differences in the three-dimensional coordinates of the feature points.

[0042] Preferably, the sensor system includes a temperature sensor, a vibration sensor and a displacement sensor; the temperature sensor is used to obtain the temperature information of the data acquisition and processing device, the control system and the data storage device in real time; the vibration sensor is used to obtain the vibration information of the fastener cleaning device, the bolt tightening device and the telescopic pressure strip device in real time; the displacement sensor is used to obtain the position information, driving distance information, and working stroke information of the fastener cleaning device, the bolt tightening device and the telescopic pressure strip device of the work vehicle.

[0043] Preferably, it also includes a driving device and a power device installed on the work vehicle, as well as running wheels; the running wheels are located on the bottom surface of the work vehicle and cooperate with the track; the driving device is connected to the running wheels and is used to drive the work vehicle to move on the track through the running wheels; the power device is used to provide power to the driving device, the fastener cleaning device, the bolt tightening device, the telescopic pressure strip device, the sensor system, the data acquisition and processing device, the control system and the data storage device.

[0044] It can be seen from the technical solutions provided by the above embodiments of the present invention that the present invention provides a railway track fastener state detection and tightening device. It includes a work vehicle, a power system, a fastener cleaning device, a bolt tightening device, a telescopic layer device, a sensor system, a data acquisition and processing device, a control system, a data storage device, etc. Among them, the fastener cleaning device is used to clean the fastener bolts and spring bars. The data acquisition and processing device obtains the fastener state information through machine vision, and uses the convolutional neural network model training to predict the friction coefficient between the fastener bolt and the friction wheel in the bolt tightening device and obtain the relative position relationship with the fastener. The telescopic layer device reduces the torque required for tightening the bolt by retracting the spring bar, and the bolt tightening device controls the application of the bolt tightening torque by adjusting the distance between the friction wheel and the rolling wheel. Thereby, the detection of the railway track fastener state and the automatic tightening of the fastener are realized. The invention greatly reduces the labor and on-site operation procedures required for on-site fastener state detection and fastener tightening, realizes the integration of railway fastener state detection and fastener bolt tightening operations, greatly improves the fastener detection and maintenance efficiency, and provides certain theoretical and technical support for the intelligent detection and automatic maintenance of railway fasteners.

[0045] Additional aspects and advantages of the present invention will be given in part in the following description, which will become obvious from the following description, or may be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0047] Figure 1 A schematic diagram of the upper structure of a railway track fastener state detection and fastening device provided by the present invention;

[0048] Figure 2 A schematic diagram of the bottom structure of a railway track fastener status detection and fastening device provided by the present invention;

[0049] Figure 3 A schematic diagram of a running wheel structure of a railway track fastener state detection and fastening device provided by the present invention;

[0050] Figure 4 A three-dimensional schematic diagram of a retractable pressure strip device of a railway track fastener state detection and fastening device provided by the present invention;

[0051] Figure 5 A schematic diagram of an initial state of a telescopic pressure strip of a telescopic pressure strip device of a railway track fastener state detection and fastening device provided by the present invention;

[0052] Figure 6 A schematic diagram of a retractable pressure strip pressing state of a retractable pressure strip device of a railway track fastener state detection and fastening device provided by the present invention;

[0053] Figure 7 A schematic diagram of the partial structure of a telescopic pressure strip device of a railway track fastener state detection and fastening device provided by the present invention, which is used to show the connection relationship between the pressing drive motor, the eccentric wheel, the spring and the telescopic pressure strip;

[0054] Figure 8 A schematic diagram of a local structure of a telescopic pressure strip device of a railway track fastener state detection and fastening device provided by the present invention from another perspective;

[0055] Fig. 9 A schematic diagram of a bolt fastening device of a railway track fastener state detection and fastening device provided by the present invention;

[0056] Fig.10 A schematic diagram of a fastener cleaning device for a railway track fastener state detection and fastening device provided by the present invention;

[0057] Fig.11 A schematic diagram of the structure of a visual sensor for railway track fastener status detection and fastening equipment provided by the present invention.

[0058] In the figure:

[0059] 1. Driving device, 2. Seat, 3. Industrial control computer, 4. Positioning system, 5. Wind deflector, 6. Traveling wheel structure, 7. Fastener cleaning device, 8. Data acquisition and processing device, 9. Lighting and alarm device, 10. Bolt fastening device, 11. Telescopic pressure strip device, 121. Traveling wheel, 122. Second track holding device, 13. First track holding device, 131. Roller, 14. Telescopic pressure strip, 15. Bolt fastening engine, 16. Friction wheel cleaning device, 17. Friction wheel, 18. Rolling wheel, 191. Ring, 192. Tensioner, 20. Driving shaft, 21. Cleaning disk, 22. Fastener cleaning engine, 23. Shock absorber, 24. Visual sensor, 25. Power device, 26. Positioning rod, 27. Data storage device, 28. Press drive motor, 29. Eccentric wheel, 30. Return spring. DETAILED DESCRIPTION

[0060] The embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and cannot be interpreted as limiting the present invention.

[0061] It will be understood by those skilled in the art that, unless expressly stated, the singular forms "one", "said", and "the" used herein may also include plural forms. It should be further understood that the term "comprising" used in the specification of the present invention refers to the presence of the features, integers, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or groups thereof. It should be understood that when we refer to an element as being "connected" or "coupled" to another element, it may be directly connected or coupled to the other element, or there may be intermediate elements. In addition, the "connection" or "coupling" used herein may include wireless connection or coupling. The term "and / or" used herein includes any unit and all combinations of one or more associated listed items.

[0062] It will be understood by those skilled in the art that, unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as those generally understood by those skilled in the art in the art to which the present invention belongs. It should also be understood that terms such as those defined in common dictionaries should be understood to have meanings consistent with the meanings in the context of the prior art, and will not be interpreted with idealized or overly formal meanings unless defined as herein.

[0063] To facilitate understanding of the embodiments of the present invention, several specific embodiments will be further explained below with reference to the accompanying drawings, and each embodiment does not constitute a limitation on the embodiments of the present invention.

[0064] See also Figures 1 to 11 The present invention provides a railway track fastener status detection and fastening device, including a working vehicle body, and a fastener cleaning device 7, a bolt fastening device 10, a telescopic pressure strip device 11, a sensor system, a data acquisition and processing device 8, a control system and a data storage device arranged on the working vehicle body.

[0065] The fastener cleaning device 7 has a rotatable cleaning disk 21, which is located on the bottom surface of the working vehicle body. The bolt fastening device has a friction wheel 17 and a rolling wheel 18, which are arranged in pairs and are located on the bottom surface of the working vehicle. Each pair of friction wheels 17 and rolling wheels 18 can rotate relative to each other. The telescopic pressure strip device 11 is located on the bottom surface of the working vehicle body and has a telescopic pressure strip 14 that can be extended and retracted. The telescopic pressure strip 14 is used to press the elastic strip of the fastener.

[0066] The data acquisition and processing device 8 has an image acquisition module and a processor. The image acquisition module is used to obtain the appearance information of the fastener. The processor is used to obtain, based on the appearance information, the following by calculation: the state of the fastener, the friction coefficient between the bolt of the fastener and the friction wheel, and the relative position between the fastener and the working vehicle.

[0067] The sensor system is used to monitor: vibration information of the fastener cleaning device, the bolt tightening device and the telescopic pressure strip device, and temperature information of the data acquisition and processing device, the control system and the data storage device.

[0068] The control system is respectively connected to the fastener cleaning device, the bolt fastening device, the telescopic pressure strip device, the sensor system, the data acquisition and processing device 8, the control system and the data storage device. The control system is used to: determine whether the railway track fastener state detection and fastening equipment fails according to the vibration information of the fastener cleaning device 7, the bolt fastening device 10 and the telescopic pressure strip device 11, the temperature information of the data acquisition and processing device 8, the control system and the data storage device; and adjust the interval between the friction wheel and the rolling wheel according to the friction coefficient between the bolt of the fastener and the friction wheel calculated by the data acquisition and processing device.

[0069] In the preferred embodiment provided by the present invention, the specific configuration of each device and system is as follows:

[0070] The drive device 1 is arranged on the working vehicle body. It drives the working vehicle to move along the railway track by using the torque generated by the motor or other power source through a specific mechanical transmission structure and running parts in contact with the rails (such as rail-holding wheels, etc.). Its power transmission and control methods are designed to enable the working vehicle to run smoothly and flexibly on the track, so that it can smoothly reach the fasteners at different positions on the track to prepare for subsequent inspection and tightening operations. In actual operation, the speed of the drive device 1 can be adjusted as needed to adapt to different operating scenarios and requirements.

[0071] The power system 25 is installed at a suitable position of the working vehicle, and is mainly composed of a power source (such as an engine, a generator set, etc.) and corresponding power distribution components (such as a distribution box, a hydraulic pump station, etc.). The energy generated by the power source is transmitted through a circuit or hydraulic pipeline to provide sufficient and stable power for each working device.

[0072] like Figure 2 As shown, the fastener cleaning device 7 can be installed at the rear of the working vehicle's wheel structure 6, which can facilitate the operation of the spring clip fastener bolts. It is equipped with a specially designed cleaning disc 21, and the cleaning disc 21 can be made of suitable materials. Fig.10 As shown, the cleaning disc 21 is driven by a fastener cleaning engine 22 installed on the working vehicle body. When the working vehicle reaches the fastener position that needs to be processed, the cleaning disc 21 removes rust, impurities, etc. on the bolt surface by contacting the bolt surface. During the cleaning process, the speed, pressure, cleaning time and other parameters of the cleaning disc can be adjusted to ensure the cleaning effect of bolts in different states, so that the bolts can be restored to a good state, creating favorable conditions for subsequent tightening operations.

[0073] The bolt tightening device 10 is arranged in front of the rear wheel at the bottom of the work vehicle, and this position can ensure that it can accurately operate the bolts of the spring clip fasteners. It adopts electric or hydraulic power drive, and realizes the control of the tightening torque through the transmission rod-sleeve structure. When the work vehicle drives to the fastener bolt that needs to be tightened, the bolt information collected in advance by the data acquisition and processing device 8 is analyzed by the industrial control computer 3, and the position of the friction wheel 17 and the rolling wheel 18 is adjusted through the shaft system device to apply the required pressure to the bolt, and apply the appropriate torque to the bolt, so that the connection between the fastener and the track is more stable.

[0074] In some preferred embodiments, Fig. 9 As shown, the bolt fastening device includes:

[0075] The engine 15 is fastened with bolts and installed on the work vehicle.

[0076] The drive shaft 20 is detachably connected to the output end of the bolt-tightening engine 15 , and the bolt-tightening engine is drivingly connected to the friction wheel 17 via the drive shaft 20 .

[0077] The shaft system structure includes a positioning rod 26, which is detachably mounted on the bottom of the work vehicle. One end of the positioning rod 26 has a rolling bearing, and the rolling wheel 18 is movably mounted on one end of the positioning rod 26 through the rolling bearing.

[0078] The transmission rod-sleeve structure has a pair of collars 191 interconnected by a tensioner 192, and the collars 191 are in contact with the drive shaft 20 and the positioning rod 26 respectively. The tensioner 192 is connected to the control system circuit, and can adjust the spacing distance between the friction wheel 17 and the rolling wheel 18 by changing its own length. In the embodiment provided by the present invention, the tensioner 192 is arranged based on the prior art (commercially available products), and a rod-type structure is preferably used, which will not be described here. The tensioner 192 adjusts the spacing distance between the friction wheel 17 and the rolling wheel 18 in a small range, which belongs to fine-tuning. The initial spacing distance between the friction wheel 17 and the rolling wheel 18 is pre-installed and set according to the model of the rail fastener bolt.

[0079] In some improvements, the outer side of the friction wheel 17 (the side away from the rolling wheel 1818) is also provided with a friction wheel 17 cleaning device 16, such as Fig. 9 As shown, a cover type structure can be adopted, in which a cleaning body, such as a sponge or wiping cotton yarn, etc., can be used to wipe the friction wheel 1717 when it rotates to remove dirt.

[0080] The retractable pressure strip device 11 is installed in the middle of the bolt fastening device on one side. The device avoids hitting other parts of the spring strip by retracting the pressure strip 14 at a suitable position. When the working vehicle moves forward under the action of the driving device, the retractable pressure strip device 11 is pressed to the appropriate position of the spring strip by mutual friction contact during operation, thereby reducing the torque required for tightening the bolt.

[0081] For example, in some preferred embodiments, Figure 4 , 7 As shown in 8, the telescopic layer 14 device includes:

[0082] A pair of first track clamps 13 are detachably mounted on the working vehicle body. The first track clamps 13 provide stable support for the telescopic pressure strip 14 device by clamping the track.

[0083] The telescopic pressure strip 14 is located in the first rail holding device 13 and is detachably connected to the first rail holding device 13. Figure 4 , 5 As shown in Figure 6, the first rail clamp 13 has a window, and the telescopic pressure strip 14 is located in the area of ​​the window and can be extended or retracted from the window.

[0084] A plurality of rollers 131 are mounted on the inner side of the first rail clamp 13. When the first rail clamp 13 clamps the rail, the rollers 131 are in contact with the surface of the rail.

[0085] The pressing drive motor 28 is installed in the first track clamp 13 or at the bottom of the working vehicle body; the output end of the pressing drive motor 28 has an eccentric wheel 29, the position of the eccentric wheel 29 corresponds to the telescopic pressure strip 14, and the pressing drive motor 28 can drive the eccentric wheel 29 so that the convex part of the eccentric wheel 29 approaches and pushes the telescopic pressure strip 14 to extend out from the window, thereby pressing the elastic strip of the fastener.

[0086] A pair of return springs 30 are respectively connected to the two ends of the telescopic pressure strip 14 (the other end of the spring is connected to the inner wall of the first rail clamp 13). When the pressing drive motor 28 drives the eccentric wheel 29 to push the telescopic pressure strip 14 out, the return spring 30 is stretched and lengthened. When the pressing drive motor 28 drives the eccentric wheel 29 to make its convex part away from the telescopic pressure strip 14, the return spring 30 can pull the telescopic pressure strip 14 back into the first rail clamp 13 through its tension (elastic restoring force).

[0087] like Figure 8 and 9 As shown, when the eccentric wheel 29 is at the current position, the telescopic pressure strip 14 is not extended, and at this time, there is some elastic force in the return spring 30 to retract. When the driving motor drives the eccentric wheel 29 to rotate 180°, the telescopic pressure strip 14 is in an extended state, and the extension amount is the largest at this time. When the eccentric wheel 29 continues to rotate, the telescopic pressure strip 14 will retract in real time under the influence of the tension of the return spring 30.

[0088] The pressing drive motor 28 is controlled in real time by an electrical signal output by the control system. Since the pressing drive motor 28 is mainly used to overcome the resistance of the return spring 30, the volume and power of the pressing drive motor 28 can be small, but the control accuracy is high.

[0089] The sensor system is installed at various necessary parts of the work vehicle, including but not limited to the vicinity of the drive device 1, the fastener cleaning device 7, the bolt fastening device 10, etc. There are various types of sensors, such as temperature sensors, vibration sensors, displacement sensors, etc. The temperature sensor can monitor the temperature changes of key components of the equipment in real time to prevent overheating and damage; the vibration sensor can detect the vibration of the equipment during operation, analyze whether there is abnormal vibration, and determine whether the mechanical parts are worn or loose; the displacement sensor can monitor the driving distance of the work vehicle, the working stroke of each device, etc. The data collected by these sensors are transmitted to the control system and the industrial control computer 3 through signal lines, so as to identify and judge whether the working status of each part of the work vehicle is normal, realize real-time monitoring of the equipment operation, and promptly discover possible faults.

[0090] The data acquisition and processing device 8 is arranged at a suitable position on the working vehicle, and includes an image acquisition module (such as Fig.11 The image acquisition module can capture the apparent state of the fastener, obtain the apparent information of the fastener bolt, and train the model through the convolutional neural network algorithm to accurately obtain the friction coefficient between the fastener bolt and the friction wheel; the distance to the track or other fixed reference points is calculated through the SURF algorithm to determine the relative position information between the spring clip fastener and the working vehicle. In addition, the visual sensor 24 can also identify the damage, missing, misalignment of the fastener and the surface condition of the rail. These collected data are transmitted to the industrial control computer 3 through the data transmission line, providing an important basis for analyzing the condition of the fastener bolt and the accurate operation of the subsequent equipment. The industrial control computer 3 is set on the working vehicle, and it is electrically connected to the sensor system, the data acquisition system 8, and the control system through data lines. The terminal is equipped with a large-capacity storage device (such as a hard disk, a solid-state hard disk, etc.) and a high-performance processor, and has a built-in convolutional neural network algorithm and a SURF algorithm. During the operation process, it receives and saves the equipment operation data collected by the sensor system, the fastener information and position data obtained by the data acquisition and processing device, and the operation parameters and instruction records of the control system. At the same time, the processing software in the device uses the above-mentioned algorithm to analyze and process these data. These processed data provide a reliable basis for subsequent analysis and decision-making, which helps to further optimize the operation process and equipment performance.

[0091] In a preferred embodiment provided by the present invention, the process of predicting the friction coefficient by a convolutional neural network algorithm includes:

[0092] (1) Image preprocessing: The main purpose of image preprocessing is to eliminate irrelevant information in the first image, restore useful real information, enhance the detectability of relevant information and simplify the data to the maximum extent, thereby improving the reliability of feature extraction, image segmentation, matching and recognition.

[0093] The first image is an image of the bolt of the fastener and the friction wheel in contact with each other.

[0094] 1) Image normalization: The linear normalization method is used to linearly scale the pixel values ​​of the first image to a specific range (such as 0 to 1 or -1 to 1) to eliminate the differences between different first images due to different pixel value ranges. According to the maximum and minimum values ​​of the image data, the pixel values ​​are mapped to the target range through a linear transformation formula.

[0095] 2) Color space conversion: The conversion from RGB to grayscale images is adopted, and the grayscale image is generated by calculating the weighted sum of the three RGB channels, so as to simplify the first image data, reduce the computational complexity, and retain the texture and shape information of the image.

[0096] 3) Image denoising: linear filtering methods, including mean filtering and Gaussian filtering, are used to remove noise by calculating the weighted average of the pixels surrounding each pixel in the image.

[0097] 4) Image enhancement:

[0098] ① Use sharpening methods: including Laplace operator and high-pass filtering. The sharpening effect is achieved by enhancing the high-frequency components in the image. The edges and details of the image are clearer, and the resolution of the image is improved.

[0099] ② Use contrast enhancement methods: including using histogram stretching and histogram equalization. By adjusting the brightness distribution of the image to enhance the contrast, the brightness difference between different objects in the image is more obvious, and the clarity of the image is improved.

[0100] 5) Image cropping and scaling: to meet the needs of subsequent processing or analysis.

[0101] (2) Construction of convolutional neural network model:

[0102] 1) Choose a deep learning framework based on PyTorch for construction and training;

[0103] 2) Design network structure:

[0104] ① Input layer: According to the image preprocessed in the previous step, the properties of the preprocessed bolt surface image are

[0105] The number is set to the size of the input layer;

[0106] ② Convolutional layer: Add multiple convolutional layers to extract local features of the bolt surface in the image. Each convolutional layer can

[0107] Set different convolution kernel sizes, numbers, and step sizes to ensure accurate operation of the model;

[0108] ③Pooling layer: Add a pooling layer after the convolutional layer to reduce the size of the feature map and reduce the amount of calculation. The pooling method can be maximum pooling and average pooling.

[0109] ④Fully connected layer: Flatten the feature map output by the pooling layer and input it into the fully connected layer. The fully connected layer is responsible for mapping the extracted features to the final friction coefficient prediction value.

[0110] ⑤Output layer: Set the number of neurons in the output layer according to the requirements of the prediction task. In this task, the output layer usually has only one neuron, which is used to output the predicted friction coefficient value.

[0111] 3) Select activation function: Use ReLU activation function after convolutional layer and fully connected layer to increase the nonlinear expression ability of the model. Use linear activation function after the output layer to output the predicted friction coefficient value.

[0112] 4) Design loss function and optimizer: The mean absolute error (MAE) loss function is selected to measure the difference between the model prediction value and the actual value. And the SGD optimizer is selected to update the model parameters.

[0113] (3) Model training and evaluation

[0114] 1) Divide the data set: Divide the data set in the built database into training set, validation set and test set. The training set is used to train the model, the validation set is used to adjust the model's hyperparameters and evaluate the model's performance, and the test set is used to finally evaluate the model's generalization ability.

[0115] 2) Training model: Use the training set data to train the model and use the validation set data to monitor the performance of the model.

[0116] During the training process, the model's hyperparameters are adjusted based on the performance on the validation set.

[0117] 3) Evaluate the model: Use the test set data to evaluate the performance of the model. Calculate the prediction error of the model on the test set to evaluate the accuracy of the model.

[0118] 4) Model tuning: Tune the model based on the evaluation results, and repeat the process of training, evaluation, and tuning until satisfactory model performance is obtained.

[0119] (4) Model deployment and application:

[0120] Deploy the trained model to the data collection and processing system, and make reasonable adjustments based on the actual operating conditions.

[0121] By using the predicted friction coefficient, the distance between the friction wheel and the rolling wheel is reasonably adjusted to ensure that sufficient torque is output to the nut.

[0122] In some other preferred embodiments, the process of calculating the relative distance by the SURF algorithm includes:

[0123] (1) Feature point extraction:

[0124] a. Construct the Hessian matrix: Use the Hessian matrix to detect the key points in the second image. By performing extreme value detection on the determinant of the Hessian matrix, the potential key point positions in the second image can be determined.

[0125] The second image is an image of the fastener used to calculate the relative position between the fastener and the working vehicle.

[0126] b. Feature point screening: By calculating the distance and angle difference between feature points, those unstable or mismatched feature points can be eliminated.

[0127] (2) Feature point description:

[0128] a. Calculate the descriptor

[0129] b. Normalization: To enhance the robustness of features, the SURF algorithm normalizes the descriptors to ensure that each feature vector has unit length.

[0130] (3) Feature point matching:

[0131] a. Calculate Euclidean distance

[0132] b. Matching screening: Matching screening based on distance ratio and outlier elimination based on RANSAC (random sampling consensus) algorithm. These methods can further eliminate wrongly matched feature point pairs.

[0133] (4) Calculate the actual distance

[0134] a. Calculate the geometric transformation relationship: Using the matched feature point pairs, the RANSAC algorithm can be used to estimate the homography matrix, which describes the perspective transformation relationship from one second image to another adjacent frame second image.

[0135] b, Use camera calibration parameters.

[0136] c. Calculate 3D coordinates and distance: Using calibration parameters and geometric transformation relationships, the 3D coordinates of corresponding points in the two second images can be calculated. By comparing the differences in these 3D coordinates, the actual object distance can be calculated.

[0137] The lighting and alarm device 9 is arranged on the working vehicle, and the lighting part thereof adopts high-brightness, multi-angle lighting lamps, which are distributed on the working vehicle body, especially near the position of the spring clip fastener. These lamps can generate light of sufficient intensity, effectively improve the visibility of the spring clip fastener, and provide good light conditions for detection and fastening operations whether in insufficient light during the day or at night. The alarm function is connected to the sensor system and the control system. When the sensor detects that the equipment has a fault (such as the temperature of a device is too high, the vibration of a key component is abnormal, etc.), the alarm function will be activated immediately, and the staff will be prompted to handle it by means of sound and light alarm (such as emitting an alarm sound and flashing a warning light).

[0138] The control system is located on the working vehicle body. The control system is electrically connected to the bolt tightening device 10, the drive device 1, the data acquisition and processing device 8, and the sensor system through the control circuit and control algorithm. It receives various monitoring data from the sensor system, determines the operating status of the equipment according to the preset program and algorithm, and sends control instructions to the drive device to adjust the driving speed and direction of the working vehicle; issues instructions such as tightening torque and operation time to the bolt tightening device; and controls the acquisition frequency and parameters of the data acquisition and processing device. By coordinating the work of each device in this way, an automated detection and tightening process is realized, and the work efficiency and accuracy are improved.

[0139] The positioning system 4 is set up next to the data storage and processing terminal on the working vehicle. It uses advanced positioning technology (such as GPS, Beidou and other satellite positioning systems combined with auxiliary positioning methods such as track mileage markers). The satellite positioning system can provide a wide range of position coordinate information, while auxiliary positioning methods such as track mileage markers can more accurately determine the specific position of the equipment on the railway track. This position information is transmitted to the control system and industrial control computer 3 through the data transmission line, providing accurate position information for the equipment, facilitating accurate positioning and recording of the working position, thereby ensuring that the equipment can accurately detect and tighten the target fasteners.

[0140] As shown in the figure, a driving device and a power device, as well as running wheels, are also installed on the work vehicle. The running wheels are located on the bottom surface of the work vehicle and cooperate with the track. As shown in the figure, the running wheels are arranged in pairs. Four pairs of running wheels are arranged at the bottom of a work vehicle. A second track clamp is also provided in the interval area of ​​each pair of running wheels. The second track clamp also adopts a double-arm structure similar to the above-mentioned first track clamp, which can provide stable support for the running wheels. The driving device is connected to the running wheel drive and is used to drive the work vehicle to move on the track through the running wheel. The power device is used to provide power to the driving device, the fastener cleaning device, the bolt tightening device, the telescopic pressure strip device, the sensor system, the data acquisition and processing device, the control system and the data storage device.

[0141] In summary, the present invention provides a railway track fastener state detection and tightening device. It includes a work vehicle, a power system, a fastener cleaning device, a bolt tightening device, a telescopic layer device, a sensor system, a data acquisition and processing device, a control system, a data storage device, etc. Among them, the fastener cleaning device is used to clean the fastener bolts and spring bars. The data acquisition and processing device obtains the fastener state information through machine vision, and uses the convolutional neural network model training to predict the friction coefficient between the fastener bolt and the friction wheel in the bolt tightening device and obtain the relative position relationship with the fastener. The telescopic layer device reduces the torque required for tightening the bolt by retracting the spring bar, and the bolt tightening device controls the application of the bolt tightening torque by adjusting the distance between the friction wheel and the rolling wheel. Thereby, the detection of the railway track fastener state and the automatic tightening of the fastener are realized. The invention greatly reduces the labor and on-site operation procedures required for on-site fastener state detection and fastener tightening, realizes the integration of railway fastener state detection and fastener bolt tightening operations, greatly improves the fastener detection and maintenance efficiency, and provides certain theoretical and technical support for the intelligent detection and automatic maintenance of railway fasteners.

[0142] Those skilled in the art can understand that the accompanying drawings are only schematic diagrams of an embodiment, and the modules or processes in the accompanying drawings are not necessarily required to implement the present invention.

[0143] It can be known from the description of the above implementation methods that those skilled in the art can clearly understand that the present invention can be implemented by means of software plus a necessary general hardware platform. Based on such an understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which can be stored in a storage medium such as ROM / RAM, a magnetic disk, an optical disk, etc., and includes a number of instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in the various embodiments of the present invention or certain parts of the embodiments.

[0144] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the device or system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment. The device and system embodiments described above are merely schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0145] The above is only a preferred specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed by the present invention should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A railway track fastener status detection and fastening device, characterized in that: It includes a work vehicle, and a fastener cleaning device, a bolt tightening device, a telescopic layer device, a sensor system, a data acquisition and processing device, a control system and a data storage device arranged on the work vehicle; The fastener cleaning device has a rotatable cleaning disk, and the cleaning disk is located on the bottom surface of the working vehicle; The bolt fastening device comprises a friction wheel and a rolling wheel, wherein the friction wheel and the rolling wheel are arranged in pairs at intervals from each other and are located on the bottom surface of the working vehicle; The friction wheel is used to drive the bolt of the fastener, and the interval between the friction wheel and the rolling wheel is adjustable; The retractable pressure strip device is located on the bottom surface of the working vehicle and has a retractable pressure strip, and the pressure strip is used to press the elastic strip of the fastener; The data acquisition and processing device comprises an image acquisition module and a processor; the image acquisition module is used to obtain the appearance information of the fastener; the processor is used to obtain, by calculation according to the appearance information: the state of the fastener, the friction coefficient between the bolt of the fastener and the friction wheel, and the relative position between the fastener and the working vehicle; The sensor system is used to monitor: vibration information of the fastener cleaning device, the bolt tightening device and the telescopic layer device, and temperature information of the data acquisition and processing device, the control system and the data storage device; The control system is respectively connected in circuit to the fastener cleaning device, bolt tightening device, telescopic pressure strip device, sensor system, data acquisition and processing device, control system and data storage device; the control system is used to: determine whether the railway track fastener status detection and tightening equipment has a fault according to the vibration information of the fastener cleaning device, bolt tightening device and telescopic pressure strip device, and the temperature information of the data acquisition and processing device, control system and data storage device; and adjust the spacing distance between the friction wheel and the rolling wheel according to the friction coefficient between the bolt of the fastener and the friction wheel calculated by the data acquisition and processing device.

2. The device according to claim 1, characterized in that The fastener cleaning device comprises a fastener cleaning engine, and the fastener cleaning engine is drivingly connected to the cleaning disk.

3. The device according to claim 1, characterized in that The bolt fastening device comprises: Bolt-fastening the engine and installing it on the work vehicle; A drive shaft, detachably connected to the output end of the bolt-tightening engine, wherein the bolt-tightening engine is drivingly connected to the friction wheel via the drive shaft; The shaft system structure comprises a positioning rod, the positioning rod is detachably mounted on the bottom of the working vehicle, one end of the positioning rod has a rolling bearing, and the rolling wheel is movably mounted on the one end of the positioning rod through the rolling bearing; The transmission rod-sleeve structure has a pair of rings connected to each other through a stretcher, and the rings are respectively in contact with the drive shaft and the positioning rod; the stretcher is connected to the control system circuit and can adjust the spacing between the friction wheel and the rolling wheel by changing its own length.

4. The device according to claim 3, characterized in that The bolt fastening device also includes a friction wheel cleaning device, which is detachably mounted on the outer side of the friction wheel.

5. The device according to claim 1, characterized in that The telescopic layering device comprises: A pair of first rail grippers, detachably mounted on the working vehicle body, wherein the first rail grippers are used to clamp the rail; a telescopic pressure strip detachably connected to the first rail holding device; the first rail holding device has a window, and the telescopic pressure strip is located in the area of ​​the window; A plurality of rollers are mounted on the inner side of the first rail holding device, and when the first rail holding device clamps the rail, the rollers are in contact with the surface of the rail; A pressing drive motor is installed on the first rail clamp of the working vehicle body; an eccentric wheel is provided at the output end of the pressing drive motor, the position of the eccentric wheel corresponds to the telescopic pressure strip, and the pressing drive motor can push the telescopic pressure strip to extend from the window and press the elastic strip of the buckle by driving the eccentric wheel; A pair of return springs are respectively connected to two ends of the telescopic pressure strip.

6. The device according to claim 1, characterized in that The data acquisition and processing device calculates the relative position between the fastener and the work vehicle through the SURF algorithm, and also calculates the state of the fastener and the friction coefficient between the bolt of the fastener and the friction wheel through the convolutional neural network algorithm.

7. The device according to claim 6, characterized in that The process of the processor obtaining the friction coefficient between the bolt of the fastener and the friction wheel by calculation according to the surface information includes: S1 preprocesses the first image, including normalization, color space conversion, noise removal and first image enhancement; the first image is an image of the bolt and the friction wheel of the fastener in contact with each other; S2 constructs a convolutional network model; the convolutional network model includes an input layer, a convolutional layer, a pooling layer, a fully connected layer and an output layer arranged in sequence along the data flow direction; the input layer is used to set the property parameters of the bolt surface features of the preprocessed first image to the size of the input layer; the convolutional layers are multiple and are used to extract local features of the bolt surface in the preprocessed first image; the pooling layer is used to reduce the size of the extracted local features of the bolt surface; the fully connected layer is used to map the local features of the bolt surface to the predicted value of the friction coefficient; the output layer is used to output the predicted friction coefficient value; S3 uses the ReLU activation function to increase the nonlinear expression ability of the convolutional network model; S4 measures the difference between the predicted value and the actual value of the convolutional network model through the mean absolute error loss function, and updates the parameters of the convolutional network model through the SGD optimizer; S5 uses the dataset to train, test, and tune the convolutional network model; S6 uses the tuned convolutional network model to predict the friction coefficient between the bolt of the fastener and the friction wheel, so that the control system can adjust the spacing distance between the friction wheel and the rolling wheel based on the obtained friction coefficient.

8. The device according to claim 6, characterized in that The process of calculating the relative position between the fastener and the work vehicle by using the SURF algorithm includes: E1 uses the Hessian matrix to detect key points in the second image, and obtains the positions of potential key points in the second image by performing extreme value detection on the determinant of the Hessian matrix; the second image is an image of the fastener used to calculate the relative position between the fastener and the work vehicle; E2 eliminates unstable or mismatched feature points by calculating the distance and angle difference between feature points in the second image; E3 obtains a descriptor by calculation based on the feature points in the second image after executing step E2; E3 normalizes the descriptor; E4: obtaining the Euclidean distance of the descriptor by calculation based on the descriptor after normalization; E5 screens and obtains feature points that match the random sampling consensus algorithm based on the ratio of the calculated Euclidean distance; E6 estimates the homography matrix based on the feature points obtained by screening through a random sampling consistency algorithm; E7 uses the image acquisition module to calibrate parameters; E8 obtains the three-dimensional coordinates of the same feature points in the second images of two adjacent frames based on the calibrated parameters and the corresponding geometric correspondence, and calculates the relative position between the fastener and the work vehicle by comparing the differences in the three-dimensional coordinates of the feature points.

9. The device according to claim 1, characterized in that The sensor system includes a temperature sensor, a vibration sensor and a displacement sensor; the temperature sensor is used to obtain the temperature information of the data acquisition and processing device, the control system and the data storage device in real time; the vibration sensor is used to obtain the vibration information of the fastener cleaning device, the bolt tightening device and the telescopic pressure strip device in real time; the displacement sensor is used to obtain the position information, driving distance information of the work vehicle, and the working stroke information of the fastener cleaning device, the bolt tightening device and the telescopic pressure strip device.

10. The device according to claim 1, characterized in that It also includes a driving device and a power device installed on the work vehicle, as well as a running wheel; the running wheel is located on the bottom surface of the work vehicle and cooperates with the track; the driving device is connected to the running wheel and is used to drive the work vehicle to move on the track through the running wheel; the power device is used to provide power to the driving device, the fastener cleaning device, the bolt tightening device, the telescopic pressure strip device, the sensor system, the data acquisition and processing device, the control system and the data storage device.

Images