Transmission line crimping pipe X-ray detection system and autonomous detection method

Abstract

The invention belongs to the technical field of crimping tube X-ray detection. The invention provides a power transmission line crimping pipe X-ray detection system and an autonomous detection method. A laser radar and a binocular vision sensor are both in communication connection with a control terminal; the robot body comprises a walking rack, an X-ray detection mechanism fixed on the walking rack and a driving wheel located at the bottom of the walking rack, the walking rack, the X-ray detection mechanism and the driving wheel are both of a wave-shaped structure, and the wave-shaped structure at least comprises two concave parts; the driving wheels are driven by a driving motor, the control terminal is in communication connection with the driving motor to control the robot body to move forwards or backwards, and the control terminal is used for conducting navigation control over the X-ray robot according to data of the laser radar and the binocular vision sensor; the detection precision and comprehensiveness of the crimping pipe with the reinforcing wire are improved, and the walking stability of the robot on the power transmission line is improved.

Description

Technical Field

[0001] The invention relates to the technical field of X-ray detection of crimped tubes, and in particular to an X-ray detection system and an autonomous detection method for crimped tubes of power transmission lines. Background Art

[0002] The statements in this section merely provide background art related to the present invention and do not necessarily constitute prior art.

[0003] The transmission line crimping tube reinforcement splicing strip (reinforcement wire) is an auxiliary hardware used to enhance the splicing effect of transmission line conductors. During use, when two conductors need to be spliced, the corresponding connector needs to be installed in the splicing tube (crimping tube). Then, the reinforcement splicing strip is passed through the connector to make it fit tightly with the conductor to be spliced. In this way, the reinforcement splicing strip can provide additional support and fixation to prevent loosening or displacement at the splice.

[0004] On the "three spans" (across highways, railways, and important lines), the connection of the conductors will be reinforced by installing reinforcing wires. Although the proportion is not high, it is a must-check item. The reinforcing wire begins to wrap a distance before the crimping tube, forks upward at the front end of the crimping tube, and finally hangs on the adjustment plate. The angle of the upward fork varies, about 10 degrees.

[0005] In the above environment, traditional robots are unable to complete efficient and comprehensive inspection of crimped tubes. The main reasons are: (1) The robot cannot autonomously navigate and walk on the reinforcement wire, and cannot reach the root position of the crimped tube, and thus cannot achieve full-range inspection of the crimped tube; (2) The reinforcement wire is made of steel, and X-rays cannot penetrate the steel core of the reinforcement wire. The reinforcement wire will block the imaging of the crimped tube, making it impossible to judge the crimping quality of the crimped tube; (3) Because it carries an X-ray transmitter, the overall weight of the X-ray inspection robot is relatively large. The existing X-ray inspection robot uses traditional walking wheels. Once it deviates on the transmission line, it is difficult to correct the position, and there is a risk of the X-ray inspection robot becoming unstable and falling; (4) Most existing X-ray inspection robots rely on remote control by ground personnel and rely on background control terminals to process X-ray inspection images. They cannot accurately realize front-end autonomous navigation and walking control and front-end image self-processing. Summary of the invention

[0006] In order to address the deficiencies in the prior art, the present invention provides an X-ray detection system and an autonomous detection method for pressure-connected pipes on power transmission lines, which solves the problem that the existing X-ray detection robots are unable to fully detect pressure-connected pipes with reinforcing wires, realizes autonomous navigation and autonomous detection processing of the X-ray detection system on power transmission lines, improves the accuracy and comprehensiveness of detection of pressure-connected pipes with reinforcing wires, avoids the influence of the reinforcing wires on the walking of the robot, and improves the stability of the robot walking on the power transmission lines.

[0007] In order to achieve the above object, the present invention adopts the following technical solution: In a first aspect, the present invention provides an X-ray detection system for transmission line crimped tubes.

[0008] A transmission line crimped tube X-ray detection system, comprising: a robot body, a control terminal mounted on the robot body, a laser radar and a binocular vision sensor, wherein the laser radar and the binocular vision sensor are both communicatively connected to the control terminal; The robot body comprises: a walking frame, an X-ray detection mechanism fixed on the walking frame, and a driving wheel located at the bottom of the walking frame, all of which adopt a wave-shaped structure, and the wave-shaped structure comprises at least two concave parts; The control terminal is connected to the drive motor for controlling the robot body to move forward or backward, and the control terminal is used to perform navigation control of the X-ray robot based on data from the laser radar and binocular vision sensor.

[0009] As a further limitation of the first aspect of the present invention, the X-ray detection mechanism comprises: a first receiving plate support beam, a second receiving plate support beam, a transmitter and a transmitter support beam; The transmitter is fixed to the walking frame through the transmitter support beam, the plate surface of the first receiving plate is arranged opposite to the plate surface of the second receiving plate, the first receiving plate support beam is located on one side of the walking frame forward direction, and the second receiving plate support beam is located on the other side of the walking frame forward direction; The first receiving plate is fixed on the first receiving plate supporting beam, and can be horizontally moved to approach or move away from the crimping tube on one side under the drive of the first horizontal driving mechanism located on the walking frame, and the second receiving plate is fixed on the second receiving plate supporting beam, and can be horizontally moved to approach or move away from the crimping tube on the other side under the drive of the second horizontal driving mechanism located on the walking frame.

[0010] As a further limitation of the first aspect of the present invention, the X-ray emission port of the transmitter is located at the front of the walking frame, the X-ray emission port of the transmitter is a rotatable emission port, and the first receiving plate and the second receiving plate extend forward to the front position of the walking frame.

[0011] As a further limitation of the first aspect of the present invention, the plate surface of the first receiving plate and the plate surface of the second receiving plate are arranged relatively parallel, and the plate surface orientation of the first receiving plate and the plate surface orientation of the second receiving plate are both perpendicular to the forward direction of the walking frame.

[0012] As a further limitation of the first aspect of the present invention, the first horizontal driving mechanism includes: a first motor, a first lead screw and a first lead screw slider, the first motor is fixed to the walking frame, the output shaft of the first motor is connected to the first lead screw, the first lead screw can rotate under the drive of the first motor, the first lead screw slider is threadedly connected to the first lead screw, and the first lead screw slider is fixedly connected to the first receiving plate support beam; The second horizontal driving mechanism includes: a second motor, a second lead screw and a second lead screw slider. The second motor is fixed on the walking frame. The output shaft of the second motor is connected to the second lead screw. The second lead screw can rotate under the drive of the second motor. The second lead screw slider is threadedly connected to the second lead screw. The second lead screw slider is fixedly connected to the second receiving plate support beam.

[0013] As a further limitation of the first aspect of the present invention, it also includes a first guide rail, a second guide rail, a third guide rail and a fourth guide rail fixed to the walking frame, the first receiving plate support beam is slidably connected to the first guide rail through a first sliding member, and is slidably connected to the second guide rail through a second sliding member; The second receiving plate support beam is slidably connected to the third guide rail via a third sliding member, and is slidably connected to the fourth guide rail via a fourth sliding member.

[0014] As a further limitation of the first aspect of the present invention, the first lead screw is located between the first guide rail and the second guide rail, and the second lead screw is located between the third guide rail and the fourth guide rail.

[0015] As a further limitation of the first aspect of the present invention, the driving wheels include a first front driving wheel and a second front driving wheel located at the front of the walking frame and a first rear driving wheel and a second rear driving wheel located at the rear of the walking frame; The first front driving wheel, the second front driving wheel, the first rear driving wheel and the second rear driving wheel all adopt a wave-shaped structure, and the wave-shaped structure includes at least two concave parts; The first front driving wheel, the second front driving wheel, the first rear driving wheel and the second rear driving wheel have the same structure, and all include: a supporting wheel hub, an inner baffle plate, an outer baffle plate and an outer corrugated rubber layer; One end of the supporting hub is connected to the inner baffle, the other end of the supporting hub is connected to the outer baffle, and the outer side of the supporting hub is sleeved with an outer layer of corrugated rubber.

[0016] In a second aspect, the present invention provides a method for autonomous X-ray detection of pressure-coupled tubes for power transmission lines.

[0017] A transmission line pressure tube X-ray autonomous detection method includes the following functions: Artificial intelligence autonomous detection system, with image recognition function, based on binocular vision camera, can detect the location, size, outline and distance of the target; After the robot body is hoisted onto the two upper conductors of the transmission line, the control terminal uses the laser radar to scan the surrounding environment of the transmission line, identify the transmission line path, and model the transmission line and the hardware on the transmission line path, and upload it to the artificial intelligence autonomous detection system. After the robot body starts walking, the laser radar keeps scanning and continuously updates the modeling results and the transmission line path; Combined with the modeling results, the robot is controlled to walk along the path of the transmission line, and the binocular vision sensor is started. The hardware and transmission line are judged secondary by the binocular vision sensor and the artificial intelligence image algorithm. When the secondary judgment is correct, the robot continues to move forward. If the difference between the secondary judgments is greater than the set threshold, the robot body stops and allows the laser radar to rescan until it walks to the position to be detected.

[0018] As a further limitation of the second aspect of the present invention, when the robot body moves on the transmission line and the posture is deviated, the concave portion of the wave-shaped structure will clamp the wire, forcing the robot body to return to the transmission line path, and when the robot body further moves out of the wire, the transmission line will be clamped into the concave portion of the adjacent wave-shaped structure; The robot's posture is determined by the binocular vision sensor's recognition of the transmission line path. When the robot's posture deviates from the center of the transmission line, the robot automatically rotates its wheels in the opposite direction to adjust its posture, so that the robot returns to the center of the transmission line and continues to move forward.

[0019] As a further limitation of the second aspect of the present invention, in the artificial intelligence autonomous detection system, the collected X-ray detection image is preprocessed, the preprocessed X-ray detection image is recognized based on the YOLOv8 model of the CSPDarkNet-53 network, a preliminary recognition result is generated, and the original X-ray detection image and the preliminary recognition result are classified into a sample library; In the artificial intelligence autonomous detection system, the configured large model is used to perform deep recognition on the X-ray detection images in the sample library, the preliminary recognition results are compared and verified based on the deep recognition results, and the YOLOv8 model is optimized based on the comparison and verification results.

[0020] As a further limitation of the second aspect of the present invention, preprocessing the X-ray detection image in the artificial intelligence autonomous detection system includes: Count the number of pixels ni at each gray level of the original input image, i=0, 1...L-1, where L is the total number of gray levels; Calculate the histogram of the original image, the probability density of each gray level Pi (ri) = ni / n, n is the total number of pixels in the original image; Compute the cumulative distribution function: , k = 0, 1, 2···L-1; Calculate the final output gray level: gk=INT[(gmax-gmin)sk(rk)+gmin+0.5] / (L-1); k=0, 1, 2...L-1, where INT[] is the integer operator, let gmin=0, gmax=L-1, then the calculation formula is simplified to: gk=INT[(L-1)sk(rk)+0.5] / (L-1); Using the mapping relationship between fk and gk, the grayscale of the original image is modified to obtain an output image whose histogram is approximately uniformly distributed, and fk is the grayscale function of the original image.

[0021] As a further limitation of the second aspect of the present invention, the activation function of the artificial intelligence autonomous detection system Taichung CSPDarkNet-53 network uses LeakyReLU: f(x)=max(ax,x); where a is a hyperparameter less than 1, set to 0.1 or 0.01, when x<0, f(x)=ax; when x>=0, f(x)=x.

[0022] As a further limitation of the second aspect of the present invention, after the robot body walks to the position to be detected, the first motor is started, the first lead screw drives the first lead screw slider to move inward, and the first lead screw slider drives the first receiving plate support beam to move inward. When the first receiving plate is close to the first crimping tube, the first motor stops rotating, the X-ray emission port of the transmitter rotates to a position facing the first receiving plate, the transmitter is started, and an X-ray detection image of the first crimping tube is generated on the first receiving plate; Start the second motor, the second lead screw drives the second lead screw slider to move inward, and the second lead screw slider drives the second receiving plate support beam to move inward. When the second receiving plate is close to the second crimping tube, the second motor stops rotating, and the X-ray emission port of the transmitter rotates to a position facing the second receiving plate. The transmitter starts and generates an X-ray detection image of the second crimping tube on the second receiving plate.

[0023] Compared with the prior art, the present invention has the following beneficial effects: 1. The present invention proposes a safe walking control method for a transmission line crimped tube X-ray inspection robot, develops a crimped tube X-ray inspection robot posture adjustment system and a wavy walking structure, and the walking mechanism can enable the robot to always walk on the plane formed by the conductor and the reinforcement wire, solving the problem that the robot cannot walk on the conductor with the reinforcement wire; when walking on the line, the robot first uses a laser radar to model the conductor, anti-vibration hammer, spacer rod and other hardware and form a motion trajectory, and under the calibration of binocular vision, the relative posture of the robot and each hardware is detected and corrected in real time, realizing the robot's autonomous navigation on the transmission line and improving the robot's adaptability to the transmission line environment.

[0024] 2. The present invention proposes a precise lateral detection method for an X-ray detection robot of a power line crimped tube with a reinforcing wire, and develops an X-ray lateral detection mechanism for crimped tubes with a reinforcing wire. Imaging plates are arranged vertically on both sides of the robot, which can avoid the upper and lower drainage lines and the reinforcing wires wrapped around the drainage lines. Driven by the robot's walking mechanism, the imaging plate can completely penetrate into the root of the crimped tube, realizing full-size detection of the crimped tube with a reinforcing wire. At the same time, strain gauges are arranged around the surface of the imaging plate, and the contact state between the imaging plate and the crimped tube is judged by the change of the strain gauge, thereby improving the detection accuracy of the imaging plate for the crimped tube with a reinforcing wire.

[0025] 3. The present invention proposes an autonomous radiation determination method for X-ray inspection robots for pressure tubes of power transmission lines. The images captured by the robot are first preprocessed by image processing technology based on histogram equalization. The preprocessed images are inferred and predicted using the CSPDarkNet-53 network of YOLOv8. Then, a secondary comparison and judgment of the predicted images in the standard library images is performed to give a preliminary judgment conclusion. At the same time, the images after the secondary judgment are further classified into a sample library, and the sample library is cyclically analyzed using a large model. The present invention solves the problem that the detection process of existing X-ray inspection robots still requires manual participation, improves the safety of X-ray inspection and the efficiency of image recognition, and realizes high-precision cyclic X-ray inspection of pressure tubes.

[0026] Advantages of additional aspects of the present invention will be given in part in the following description, and in part will become obvious from the following description, or will be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] The accompanying drawings in the specification, which constitute a part of the present invention, are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations on the present invention.

[0028] Figure 1 A schematic diagram of a transmission line crimped tube X-ray detection system provided in Example 1 of the present invention; Figure 2 A top view of a transmission line crimp tube X-ray detection system provided in Example 1 of the present invention; Figure 3 Schematic diagram of the X-ray autonomous detection method for the pressure pipe of the transmission line provided in Example 2 of the present invention Figure 1 ; Figure 4 Schematic diagram of the X-ray autonomous detection method for the pressure pipe of the transmission line provided in Example 2 of the present invention Figure 2 ; Figure 5 Schematic diagram of the X-ray autonomous detection method for the pressure pipe of the transmission line provided in Example 2 of the present invention Figure 3 ; Figure 6 Schematic diagram of the X-ray autonomous detection method for the pressure pipe of the transmission line provided in Example 2 of the present invention Figure 4 ; Among them, 1. walking frame; 101. laser radar; 102. binocular vision sensor; 2. transmitter; 3. transmitter support beam; 4. first receiving plate support beam; 5. second receiving plate support beam; 6. first receiving plate; 7. second receiving plate; 8. first front drive wheel; 9. second front drive wheel; 10. first rear drive wheel; 11. second rear drive wheel; 12. lifting frame; 13. control terminal; 14. first motor; 15. first lead screw; 16. first lead screw slider; 17. second motor; 18. second lead screw; 19. second lead screw slider; 20. first guide rail; 21. second guide rail; 22. third guide rail; 23. fourth guide rail. DETAILED DESCRIPTION

[0029] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0030] It should be noted that the following detailed descriptions are exemplary and are intended to provide further explanation of the present invention. Unless otherwise specified, all technical and scientific terms used herein have the same meanings as those commonly understood by those skilled in the art to which the present invention belongs.

[0031] In the absence of conflict, the embodiments of the present invention and the features of the embodiments may be combined with each other.

[0032] Embodiment 1: In this implementation, a transmission line crimped tube X-ray detection system is proposed, comprising: a robot body, a control terminal 13 mounted on the robot body, a laser radar 101 and a binocular vision sensor 102, wherein the laser radar 101 and the binocular vision sensor 102 are both communicatively connected to the control terminal; The robot body comprises: a walking frame 1, an X-ray detection mechanism fixed on the walking frame 1 and a driving wheel located at the bottom of the walking frame, all of which adopt a wave-shaped structure, and the wave-shaped structure comprises at least two concave parts; The driving wheel is driven by a driving motor, and the control terminal 13 is communicated with the driving motor to control the forward or backward movement of the robot body. The control terminal is used to perform navigation control of the X-ray robot based on data from the laser radar 101 and the binocular vision sensor 102.

[0033] The X-ray detection mechanism comprises: a first receiving plate support beam 4, a second receiving plate support beam 5, a transmitter 2 and a transmitter support beam 3; The transmitter 2 is fixed to the walking frame 1 through the transmitter support beam 3, the plate surface of the first receiving plate 6 is arranged opposite to the plate surface of the second receiving plate 7, the first receiving plate support beam 4 is located on one side of the forward direction of the walking frame 1, and the first receiving plate support beam 4 is located on the other side of the forward direction of the walking frame 1; The first receiving plate 6 is fixed on the first receiving plate support beam 4, and can move horizontally under the drive of the first horizontal driving mechanism located on the walking frame 1. The second receiving plate 7 is fixed on the second receiving plate support beam 5, and the second receiving plate 7 can approach or move away from the first receiving plate 6 under the drive of the second horizontal driving mechanism located on the walking frame 1.

[0034] In this implementation, preferably, a hoisting frame 12 and a control terminal 13 are fixed on the walking frame 1, the hoisting frame 12 is used to realize the hoisting of the drone, and a battery module is also arranged in the box where the control terminal 13 is located, and the battery module is used to realize the power supply of various electrical components.

[0035] In the present implementation, preferably, the above-mentioned X-ray detection robot adopts wavy wheels as the walking mechanism. After field testing, the X-ray detection robot can walk on the reinforcement line; two receiving plates (i.e., the first receiving plate 6 and the second receiving plate 7) are distributed on both sides of the X-ray detection robot, and the receiving plates on both sides can move left and right independently (i.e., along a direction perpendicular to the walking direction of the walking mechanism) through the cooperation of the lead screw, the lead screw slider and the guide rail; the X-ray detection mechanism composed of the transmitter and the receiving plate is forwarded on the front side of the X-ray detection robot, and the full range detection of the crimping tubes (i.e., the first crimping tube and the second crimping tube) is realized through the movement of the X-ray detection robot, and the detection of the left and right two crimping tubes (i.e., the first crimping tube and the second crimping tube) is realized through the rotation of the X-ray emission port of the transmitter; in combination with the above-mentioned test, the skeleton of the X-ray detection robot is composed of carbon fiber square tubes connected, and a frame structure is adopted to ensure the stability of the overall structure.

[0036] In this implementation, preferably, the X-ray emission port of the transmitter 2 is located at the front of the walking frame 1.

[0037] In this implementation, preferably, the X-ray emission port of the transmitter 2 is a rotatable emission port.

[0038] In this implementation, preferably, the first receiving plate 6 and the second receiving plate 7 extend forward to the front position of the walking frame 1 .

[0039] As a further limitation of the first aspect of the present invention, the plate surface of the first receiving plate 6 and the plate surface of the second receiving plate 7 are arranged relatively parallel, and the plate surface orientation of the first receiving plate 6 and the plate surface orientation of the second receiving plate 7 are both perpendicular to the forward direction of the walking frame 1.

[0040] In this implementation, preferably, the first horizontal driving mechanism includes: a first motor 14, a first lead screw 15 and a first lead screw slider 16, the first motor 14 is fixed on the walking frame 1, the output shaft of the first motor 14 is connected to the first lead screw 15, the first lead screw 15 can rotate under the drive of the first motor 14, the first lead screw slider 16 is threadedly connected to the first lead screw 15, and the first lead screw slider 16 is fixedly connected to the first receiving plate support beam 4; The second horizontal driving mechanism includes: a second motor 17, a second lead screw 18 and a second lead screw slider 19. The second motor 17 is fixed on the walking frame 1. The output shaft of the second motor 17 is connected to the second lead screw 18. The second lead screw 18 can rotate under the drive of the second motor 17. The second lead screw slider 19 is threadedly connected to the second lead screw 18. The second lead screw slider 19 is fixedly connected to the second receiving plate support beam 5.

[0041] In the present implementation, preferably, it also includes a first guide rail 20, a second guide rail 21, a third guide rail 22 and a fourth guide rail 23 fixed on the walking frame 1, the first receiving plate support beam 4 is slidably connected to the first guide rail 20 through a first sliding member, and is slidably connected to the second guide rail 21 through a second sliding member; the second receiving plate support beam 5 is slidably connected to the third guide rail 22 through a third sliding member, and is slidably connected to the fourth guide rail 23 through a fourth sliding member.

[0042] In this implementation, the first lead screw 15 is located between the first guide rail 20 and the second guide rail 21 (i.e., the first guide rail 20 is located at the front side of the first lead screw 15, and the second guide rail 21 is located at the rear side of the first lead screw 15), and the second lead screw 18 is located between the third guide rail 22 and the fourth guide rail 23 (i.e., the third guide rail 22 is located at the front side of the second lead screw 18, and the fourth guide rail 23 is located at the rear side of the second rod), to prevent the lead screw from getting stuck due to inaccurate positioning, and each guide rail plays a supporting and guiding role. The two ends of the first lead screw 15 and the second lead screw 18 are respectively connected by bearing seats and play a role of mechanical limiters. The left and right moving strokes of the first receiving plate 6 and the second receiving plate 7 are both 0-100mm.

[0043] In this implementation, preferably, the walking mechanism of this implementation is a four-wheel drive structure, and the driving wheels include a first front driving wheel 8 and a second front driving wheel 9 located at the front of the walking frame 1 and a first rear driving wheel 10 and a second rear driving wheel 11 located at the rear of the walking frame 1; The first front drive wheel 8, the second front drive wheel 9, the first rear drive wheel 10 and the second rear drive wheel 11 all adopt a wavy structure, and the wavy structure includes at least two recesses (preferably four recesses can be selected). This structure can adapt to the detection of all upper and lower line crimping tubes, including "three span" and other lines with reinforcement lines; the first front drive wheel 8, the second front drive wheel 9, the first rear drive wheel 10 and the second rear drive wheel 11 have the same structure, and all include: a supporting hub, an inner baffle, an outer baffle and an outer layer of wavy rubber; one end of the supporting hub is connected to the inner baffle, and the other end of the supporting hub is connected to the outer baffle, and the outer side of the supporting hub is provided with an outer layer of wavy rubber.

[0044] In the present implementation, the first front drive wheel 8, the second front drive wheel 9, the first rear drive wheel 10 and the second rear drive wheel 11 are all designed with transmission stoppers on the supporting wheel hubs, and the first front drive wheel 8, the second front drive wheel 9, the first rear drive wheel 10 and the second rear drive wheel 11 are also assembled in a split manner to facilitate later debugging and replacement.

[0045] In this implementation, in order to facilitate posture adjustment, the wavy recessed portions of the first front drive wheel 8, the second front drive wheel 9, the first rear drive wheel 10 and the second rear drive wheel 11 are designed in the middle of the wire, playing a role similar to that of a V-shaped wheel guide. At the same time, when the left and right wires are tilted, the V-shaped outer contour also serves to prevent the robot from sliding sideways.

[0046] In summary, this implementation method proposes a robot body structure for the horizontal double-split upper and lower outlets of the reinforcement line, which adopts an integrated detection method and can complete the detection of the two left and right crimping tubes at one time; a walking mechanism that can walk on the reinforcement line is proposed. After the robot is equipped with this wheel, it can smoothly travel on the horizontal double-split conductor with the reinforcement line, and the shape structure of the corrugated wheel enables the robot to walk on the left and right conductors with a certain inclination angle, thereby improving the safety of the robot; a detection method for the lateral distribution of the imaging plate is proposed, and the motor drives the lead screw so that the imaging plate can be as close to the crimping tube as possible, thereby achieving the best detection effect.

[0047] Embodiment 2: This implementation provides an autonomous X-ray detection method for power transmission line crimping tubes. The robot adopts a fusion method of laser radar and binocular vision sensor. The laser radar is used for environmental modeling and path judgment, and the binocular vision sensor is used to make a secondary judgment on the power transmission line path and hardware identified by the laser radar, so as to realize the autonomous driving of the robot on the wire. Figure 3 As shown, the following process is included: S1.1: After the robot body is hoisted onto the two conductors of the transmission line, the control terminal scans the surrounding environment of the transmission line through the laser radar, identifies the transmission line path, and models the transmission line and the hardware on the transmission line path, and uploads them to the artificial intelligence autonomous detection system. After the robot body starts walking, the laser radar keeps scanning and continuously updates the modeling results and the transmission line path; S1.2: Based on the modeling results, the robot is controlled to walk along the transmission line path, and the binocular vision sensor is started. The hardware and transmission line are judged again through the binocular vision sensor and the calculation of the artificial intelligence image algorithm. When the second judgment is correct, the robot continues to move forward. If the difference between the second judgments is greater than the set threshold, the robot body stops and the laser radar rescans until it walks to the position to be detected.

[0048] In this implementation, the X-ray detection robot adopts wave-shaped structure self-adjustment and binocular vision feedback adjustment, and adopts a dual adjustment method to achieve the posture adjustment of the robot on the wire. Specifically, Figure 4 As shown, including: S2.1: When the robot body moves on the transmission line and its posture deviates, the concave part of the wave-shaped structure will catch the wire, forcing the robot body to return to the transmission line path. When the robot body moves further out of the wire, the transmission line will be caught in the concave part of the adjacent wave-shaped structure. S2.2: The robot body's posture is determined by the binocular vision sensor's recognition result of the transmission line path. When the robot body's posture deviates from the center of the transmission line, the robot body automatically rotates its wheels in the opposite direction to adjust its posture so that the robot body returns to the center of the transmission line and continues to move forward.

[0049] This implementation method innovatively proposes a dual receiving plate movement method to achieve a more comprehensive detection of two crimping tubes, such as Figure 5 As shown, including: S3.1: After the robot body moves to the position to be detected, the first motor 14 is started, the first lead screw 15 drives the first lead screw slider 16 to move inward, and the first lead screw slider 16 drives the first receiving plate support beam 4 to move inward. When the first receiving plate approaches the first crimping tube, the first motor 14 stops rotating, the X-ray emission port of the transmitter rotates to a position facing the first receiving plate, the transmitter is started, and an X-ray detection image of the first crimping tube is generated on the first receiving plate; S3.2: Start the second motor 17, the second lead screw 18 drives the second lead screw slider 19 to move inward, and the second lead screw slider 19 drives the second receiving plate support beam 5 to move inward. When the second receiving plate approaches the second crimping tube, the second motor 17 stops rotating, and the X-ray emission port of the transmitter rotates to a position facing the second receiving plate. The transmitter starts and generates an X-ray detection image of the second crimping tube on the second receiving plate.

[0050] In this implementation, a detection image defect recognition method for X-ray slice image preprocessing and target defect automatic detection is proposed, which realizes unmanned operation of the entire process of X-ray detection, further reduces the manual participation of X-ray detection, improves the safety of X-ray detection, and improves the efficiency of image recognition. Figure 6 As shown: S4.1: In the artificial intelligence autonomous detection system, the collected X-ray detection images are preprocessed, and the YOLOv8 model based on the CSPDarkNet-53 network is used to recognize the preprocessed X-ray detection images, generate preliminary recognition results, and classify the original X-ray detection images and preliminary recognition results into the sample library; S4.2: In the artificial intelligence autonomous detection system, the configured large model is used to perform deep recognition on the X-ray detection images in the sample library, the preliminary recognition results are compared and verified based on the deep recognition results, and the YOLOv8 model is optimized based on the comparison and verification results.

[0051] In this implementation, the images collected by the binocular vision sensor are preprocessed through the image preprocessing technology based on histogram equalization to preprocess the X-ray detection images; the preprocessed images are inferred and predicted using the CSPDarkNet-53 network of YOLOv8, and the predicted images are compared with the images in the standard library to perform secondary judgment. The images after the secondary judgment are further classified into the sample library, and the sample library is analyzed using a large model.

[0052] In this implementation, the X-ray detection image is preprocessed in the artificial intelligence autonomous detection system, including: Count the number of pixels ni at each gray level of the original input image, i=0, 1...L-1, where L is the total number of gray levels; Calculate the histogram of the original image, that is, the probability density of each gray level, Pi (ri) = ni / n, n is the total number of pixels in the original image; Compute the cumulative distribution function: , k = 0, 1, 2···L-1; Calculate the final output gray level: gk=INT[(gmax-gmin)sk(rk)+gmin+0.5] / (L-1); k=0, 1, 2...L-1, where INT[] is the integer operator, let gmin=0, gmax=L-1, then the calculation formula is simplified to: gk=INT[(L-1)sk(rk)+0.5] / (L-1); Using the mapping relationship between fk and gk, the grayscale of the original image is modified to obtain an output image whose histogram is approximately uniformly distributed, and fk is the grayscale function of the original image.

[0053] In this implementation, the activation function of the CSPDarkNet-53 network in the artificial intelligence autonomous detection system uses LeakyReLU: f(x)=max(ax,x); where a is a hyperparameter less than 1, set to 0.1 or 0.01, when x<0, f(x)=ax; when x>=0, f(x)=x.

[0054] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A transmission line crimped tube X-ray detection system, characterized in that: include: A robot body, a control terminal mounted on the robot body, a laser radar and a binocular vision sensor, wherein the laser radar and the binocular vision sensor are both communicatively connected to the control terminal; The robot body comprises: a walking frame, an X-ray detection mechanism fixed on the walking frame, and a driving wheel located at the bottom of the walking frame, all of which adopt a wave-shaped structure, and the wave-shaped structure comprises at least two concave parts; The control terminal is connected to the drive motor for controlling the robot body to move forward or backward, and the control terminal is used to perform navigation control of the X-ray robot based on data from the laser radar and binocular vision sensor.

2. The transmission line crimped tube X-ray detection system according to claim 1, characterized in that: The X-ray detection mechanism comprises: a first receiving plate support beam, a second receiving plate support beam, a transmitter and a transmitter support beam; The transmitter is fixed to the walking frame through the transmitter support beam, the plate surface of the first receiving plate is arranged opposite to the plate surface of the second receiving plate, the first receiving plate support beam is located on one side of the walking frame forward direction, and the second receiving plate support beam is located on the other side of the walking frame forward direction; The first receiving plate is fixed on the first receiving plate supporting beam, and can be horizontally moved to approach or move away from the crimping tube on one side under the drive of the first horizontal driving mechanism located on the walking frame, and the second receiving plate is fixed on the second receiving plate supporting beam, and can be horizontally moved to approach or move away from the crimping tube on the other side under the drive of the second horizontal driving mechanism located on the walking frame.

3. The transmission line crimped tube X-ray detection system according to claim 1, characterized in that: The X-ray emission port of the transmitter is located at the front of the walking frame, and the X-ray emission port of the transmitter is a rotatable emission port. The first receiving plate and the second receiving plate extend forward to the front position of the walking frame.

4. The power transmission line crimped tube X-ray detection system according to claim 2 or 3, characterized in that: The plate surface of the first receiving plate and the plate surface of the second receiving plate are arranged relatively parallel, and the plate surface orientations of the first receiving plate and the second receiving plate are both perpendicular to the forward direction of the walking frame.

5. The power transmission line crimped tube X-ray detection system according to claim 2 or 3, characterized in that: The first horizontal driving mechanism comprises: a first motor, a first lead screw and a first lead screw slider, the first motor is fixed on the walking frame, the output shaft of the first motor is connected to the first lead screw, the first lead screw can rotate under the drive of the first motor, the first lead screw slider is threadedly connected to the first lead screw, and the first lead screw slider is fixedly connected to the first receiving plate support beam; The second horizontal driving mechanism includes: a second motor, a second lead screw and a second lead screw slider. The second motor is fixed on the walking frame. The output shaft of the second motor is connected to the second lead screw. The second lead screw can rotate under the drive of the second motor. The second lead screw slider is threadedly connected to the second lead screw. The second lead screw slider is fixedly connected to the second receiving plate support beam.

6. The transmission line crimped tube X-ray detection system according to claim 5, characterized in that: It also includes a first guide rail, a second guide rail, a third guide rail and a fourth guide rail fixed to the walking frame, the first receiving plate support beam is slidably connected to the first guide rail through a first sliding member, and is slidably connected to the second guide rail through a second sliding member; The second receiving plate support beam is slidably connected to the third guide rail via a third sliding member, and is slidably connected to the fourth guide rail via a fourth sliding member.

7. The transmission line crimped tube X-ray detection system according to claim 6, characterized in that: The first lead screw is located between the first guide rail and the second guide rail, and the second lead screw is located between the third guide rail and the fourth guide rail.

8. The power transmission line crimped tube X-ray detection system according to claim 2 or 3, characterized in that: The driving wheels include a first front driving wheel and a second front driving wheel located at the front of the walking frame and a first rear driving wheel and a second rear driving wheel located at the rear of the walking frame; The first front driving wheel, the second front driving wheel, the first rear driving wheel and the second rear driving wheel all adopt a wave-shaped structure, and the wave-shaped structure includes at least two concave parts; The first front driving wheel, the second front driving wheel, the first rear driving wheel and the second rear driving wheel have the same structure, and all include: a supporting wheel hub, an inner baffle plate, an outer baffle plate and an outer corrugated rubber layer; One end of the supporting hub is connected to the inner baffle, the other end of the supporting hub is connected to the outer baffle, and the outer side of the supporting hub is sleeved with an outer layer of corrugated rubber.

9. A transmission line pressure tube X-ray autonomous detection method, characterized in that: Used in the transmission line crimped tube X-ray detection system according to any one of claims 1 to 8, The process includes: After the robot body is hoisted onto the two conductors on the upper side of the transmission line, the control terminal uses the laser radar to scan the surrounding environment of the transmission line, identify the transmission line path, and model the transmission line and the hardware on the transmission line path, and upload it to the artificial intelligence autonomous detection system. After the robot body starts walking, the laser radar keeps scanning and continuously updates the modeling results and the transmission line path; Combined with the modeling results in the artificial intelligence autonomous detection system, the robot is controlled to walk according to the path of the transmission line, and the binocular vision sensor is started. The hardware and transmission line are judged secondary by the binocular vision sensor. When the secondary judgment is correct, the robot continues to move forward. If the difference between the secondary judgments is greater than the set threshold, the robot body stops and allows the laser radar to rescan until it walks to the position to be detected.

10. The X-ray autonomous detection method for power transmission line crimped tubes according to claim 9, characterized in that: When the robot body moves on the transmission line and its posture deviates, the concave part of the wave-shaped structure will clamp the wire, forcing the robot body to return to the transmission line path. When the robot body moves further out of the wire, the transmission line will be clamped into the concave part of the adjacent wave-shaped structure. The robot's posture is determined by the binocular vision sensor's recognition of the transmission line path. When the robot's posture deviates from the center of the transmission line, the robot automatically rotates its wheels in the opposite direction to adjust its posture, so that the robot returns to the center of the transmission line and continues to move forward.

11. The X-ray autonomous detection method for power transmission line crimped tubes according to claim 9, characterized in that: In the artificial intelligence autonomous detection system, the collected X-ray detection images are preprocessed, and the YOLOv8 model based on the CSPDarkNet-53 network is used to recognize the preprocessed X-ray detection images, generate preliminary recognition results, and classify the original X-ray detection images and preliminary recognition results into the sample library; In the artificial intelligence autonomous detection system, the configured large model is used to perform deep recognition on the X-ray detection images in the sample library, the preliminary recognition results are compared and verified based on the deep recognition results, and the YOLOv8 model is optimized based on the comparison and verification results.

12. The X-ray autonomous detection method for power transmission line crimped tubes according to claim 9, characterized in that: The images collected by the binocular vision sensor are preprocessed by the image preprocessing technology based on histogram equalization in the artificial intelligence autonomous detection system to preprocess the X-ray detection images; In the artificial intelligence autonomous detection system, the CSPDarkNet-53 network of YOLOv8 is used to perform inference and prediction on the preprocessed images, and the predicted images are compared with the images in the standard library to perform secondary judgment. The images after secondary judgment are further classified into the sample library, and the sample library is analyzed using a large model.

13. The X-ray autonomous detection method for power transmission line crimped tubes according to claim 12, characterized in that: Preprocessing of X-ray inspection images in the AI ​​autonomous inspection system includes: Count the number of pixels ni at each gray level of the original input image, i=0, 1...L-1, where L is the total number of gray levels; Calculate the histogram of the original image, that is, the probability density of each gray level, Pi (ri) = ni / n, n is the total number of pixels in the original image; Compute the cumulative distribution function: , k = 0, 1, 2···L-1; Calculate the final output gray level: gk=INT[(gmax-gmin)sk(rk)+gmin+0.5] / (L-1); k=0, 1, 2...L-1, where INT[] is the integer operator, let gmin=0, gmax=L-1, then the calculation formula is simplified to: gk=INT[(L-1)sk(rk)+0.5] / (L-1); Using the mapping relationship between fk and gk, the grayscale of the original image is modified to obtain an output image whose histogram is approximately uniformly distributed, and fk is the grayscale function of the original image.

14. The X-ray autonomous detection method for power transmission line crimped tubes according to claim 13, characterized in that: In the artificial intelligence autonomous detection system, the activation function of the CSPDarkNet-53 network uses LeakyReLU: f(x)=max(ax,x); where a is a hyperparameter less than 1, set to 0.1 or 0.01, when x<0, f(x)=ax; when x>=0, f(x)=x.

15. A transmission line pressure tube X-ray autonomous detection method, characterized in that: By using the transmission line crimped tube X-ray detection system according to claim 5, The process includes: After the robot body moves to the position to be detected, the first motor is started, the first lead screw drives the first lead screw slider to move inward, and the first lead screw slider drives the first receiving plate support beam to move inward. When the first receiving plate is close to the first crimping tube, the first motor stops rotating, the X-ray emission port of the transmitter rotates to a position facing the first receiving plate, the transmitter is started, and an X-ray detection image of the first crimping tube is generated on the first receiving plate; Start the second motor, the second lead screw drives the second lead screw slider to move inward, and the second lead screw slider drives the second receiving plate support beam to move inward. When the second receiving plate is close to the second crimping tube, the second motor stops rotating, and the X-ray emission port of the transmitter rotates to a position facing the second receiving plate. The transmitter starts and generates an X-ray detection image of the second crimping tube on the second receiving plate.

Images