Power transmission line early warning system

Abstract

The invention discloses a power transmission line early warning system, and belongs to the technical field of power transmission line monitoring, the early warning system is installed on an angle steel tower through a reinforcing support, and the early warning system comprises two millimeter-scale radars, an AI recognizer and a holder monitoring device; a millimeter-level radar is used for detecting surrounding objects in real time, and the width difference between the power transmission line and the early warning equipment is measured and marked as A when the millimeter-level radar is installed in the early warning equipment. According to the invention, through the provided pitch angle automatic calculation and real-time AI identification technology, the automatic pitch angle adjustment avoids the identification error of an object at the edge of a picture, the AI detection precision is improved, the real-time identification technology enables the system to trigger early warning at the moment of the occurrence of a threat, the timeliness demand of the early warning of a power transmission line is satisfied, and the early warning efficiency is improved. The pre-warning system is installed on the angle steel tower in an inclined strut reinforcing mode, the position of the pre-warning system can be horizontally adjusted, and the installation firmness degree and adaptability of the pre-warning system are improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of power transmission line monitoring, and particularly relates to a power transmission line early warning system. Background Art

[0002] Currently, the monitoring system of power transmission lines usually relies on fixed monitoring devices to obtain images and detects potential dangers through video analysis. However, the traditional system has the following limitations in terms of the real-time performance of processing the position of monitored objects, pitch angle calculation, and AI recognition. Manual pitch angle calculation: The traditional pitch angle calculation method usually adjusts the corresponding angle based on the distance of the object. However, this method has a large deviation in the position of the monitored object, especially when the object is at the edge of the monitoring screen, the calculation error will increase significantly, resulting in a decrease in the subsequent AI recognition accuracy. AI recognition delay: The traditional AI recognition method usually records a video first and then performs overall recognition, with a long processing time, which cannot meet the real-time early warning requirements. Moreover, the early warning system generally relies on a long pole installed on an angle steel tower, with poor reliability and limitations. We have designed a power transmission line early warning system with automatic pitch angle adjustment to improve the position of dangerous objects to be recognized in the screen; real-time artificial intelligence recognition to improve the early warning speed. Moreover, this early warning system is installed on the angle steel tower by means of diagonal bracing reinforcement, and can also horizontally adjust the position of the early warning system, improving its installation firmness and increasing the adaptability of installation. Summary of the Invention

[0003] The purpose of the present invention is to provide a power transmission line early warning system, which has the advantages of automatic pitch angle adjustment to improve the position of dangerous objects to be recognized in the screen; real-time artificial intelligence recognition to improve the early warning speed, and this early warning system is installed on the angle steel tower by means of diagonal bracing reinforcement, and can also horizontally adjust the position of the early warning system, improving its installation firmness and increasing the adaptability of installation, so as to solve the above-mentioned problems in the background art.

[0004] The technical solution of the present invention to solve the above technical problems is as follows: A power transmission line early warning system, which is installed on the angle steel tower through a reinforcement bracket. The early warning system includes two millimeter-level radars, an AI recognizer, and a pan-tilt monitoring device;

[0005] The millimeter-level radars are used to detect surrounding objects in real time. During installation, the width difference between the power transmission line and the early warning device is measured, marked as A. If the warning area is set within 50 meters close to the power transmission line, then for the object x information reported by the radar, when x - A <= 50, it meets the condition for video surveillance AI recognition;

[0006] When an object is detected to enter the preset safe distance, report the xy information of the object. The system automatically rotates the camera to aim at the detected object and starts recording video according to the reported information; when installing, calculate the distance on flat ground and the distance on a slope. Within the flat ground distance, based on the reported xy and the camera height, assuming the height is H and the distance between the object and the tower pole is Y, the pitch angle w to be adjusted is:

[0007] W = 90 - arctan(H / Y)

[0008] If the object appears on a sloped surface, calculate the logic of the pitch angle;

[0009] Once the pitch angle is adjusted in place, start recording the video. While recording and saving the video, send each frame, which is an h264 video frame, to the AI recognition module;

[0010] The video is transmitted frame by frame to the shared memory module. The shared memory module is divided into multiple blocks (such as a circular queue), and each block can accommodate one frame of data. The camera recording module, as the producer, puts the video frames into the idle shared memory blocks, and the AI module, as the consumer, reads the video frames from the shared memory blocks to ensure seamless data flow. YOLOv8 is a lightweight and real-time deep learning model for object detection and classification, supporting input of images with multiple resolutions, and outputting bounding boxes, classes, and confidence levels. The model file is loaded in the pre-trained weight format, based on the PyTorch or ONNXRuntime framework. When initializing the inference engine, set the computing device (such as GPU / TPU or CPU), read the video frames from the shared memory, and decode them into the original image format (such as BGR or RGB), normalize the image, adjust the size (usually 640x640 or 416x416), and perform channel swapping to adapt to the input requirements of YOLOv8. Then input the processed image into the YOLOv8 model to obtain the detection results;

[0011] The results output by YOLOv8 include:

[0012] Bounding box: The position of the target in the image;

[0013] Class: The class of the detected target object (such as person, car, drone, etc.);

[0014] Confidence: The confidence level of the model for the detection result;

[0015] The AI recognition software in the system can automatically identify the object category and match it with the preset monitoring objects; if the identified object belongs to the category that the system needs to monitor, the system will transmit the data to the monitoring server through the 4G network. After receiving the relevant alarm, the server will immediately push the alarm information to the management personnel through the WeChat official account so that they can take corresponding measures quickly;

[0016] The monitoring device includes a solar power supply system, an embedded intelligent recognition system, a 4G communication module, and a varifocal camera. The solar power supply system provides power support for the embedded intelligent recognition system and the spherical camera to ensure the continuous operation of the device in remote areas. The 4G communication module is responsible for transmitting the detection data and recognition results to the monitoring server in a timely manner, while the camera is connected to the embedded intelligent recognition system through the RS485 bus to provide high-quality video data to support AI analysis;

[0017] In addition, the embedded intelligent recognition system also integrates an audio module, including an audio power amplifier and a high - pitched horn, which can issue an audio alarm when necessary to enhance the early warning ability of the system in a multi - level manner;

[0018] The design of the entire system is centered around modularization to ensure the efficient collaborative work of each part. Especially in the AI part, the automated identification and learning functions are emphasized, reducing the dependence on manual intervention and achieving true intelligent monitoring and self - optimization;

[0019] The reinforcement bracket includes a longitudinal angle steel arranged on the angle steel tower and a transverse angle steel welded to one side of the longitudinal angle steel. A vertical pipe is arranged between the two millimeter-level radars. Both of the two millimeter-level radars are installed on the surface of the vertical pipe through angle-adjustable clamps. The pan-tilt monitoring device is installed on the top of the AI identifier. An installation angle plate is installed at the bottom of the AI identifier. A round rod is arranged between the bottom of the installation angle plate and the top of the vertical pipe. Both the bottom of the installation angle plate and the top are installed on the surface of the round rod through angle-adjustable clamps. A fixed pipe is slidably connected to the surface of the round rod. One end of the fixed pipe is welded with an installation angle code. Both ends of the installation angle code are installed with Z-shaped parts through bolts. The Z-shaped parts are sleeved on one side of the longitudinal angle steel. The longitudinal angle steel abuts against the side where the Z-shaped parts tightly abut. Four fastening bolts threadedly connected to the Z-shaped parts are arranged through the Z-shaped parts. The end of the fastening bolt abuts against the longitudinal angle steel. An inclined strut reinforcement component is arranged between the surface of the fixed pipe and the transverse angle steel. The inclined strut reinforcement component includes a sector-shaped clamp and an arc-shaped clamp sleeved on the surface of the fixed pipe. Both ends of the sector-shaped clamp and both ends of the arc-shaped clamp are fixedly installed through bolts. Adjustable inclined strut rods are installed on the surfaces of the bolts at both ends of the sector-shaped clamp and the arc-shaped clamp. Two strip holes are respectively opened on both sides of the top of the transverse angle steel. A screw rod one is slidably connected to the inner cavity of the strip hole. The bottoms of the two screw rods one are fixedly connected with a connecting sleeve. The inner cavity of the connecting sleeve is movably installed with the adjustable inclined strut rod through a bolt. Washers are sleeved on the surfaces of the two screw rods one. A washer nut one threadedly connected to the surface of the screw rod one is arranged on the top of the washer. A tying component is arranged between the two adjustable inclined strut rods. A horizontal adjustment component is arranged between the reinforcing rib and the fixed pipe.

[0020] Preferably, the adjustable inclined strut rod includes a rod one and a rod two. Connecting heads are respectively arranged at the opposite ends of the rod one and the rod two. The connecting head at the top is installed in the inner cavity of the connecting sleeve. The connecting head at the bottom is installed on the surface of the bolt between the sector-shaped clamp and the arc-shaped clamp.

[0021] Preferably, threaded rods are respectively threadedly connected to the inner cavities on the opposite sides of the rod one and the rod two. An auxiliary rotating block is welded between the two threaded rods.

[0022] Preferably, the tying component includes a perforation opened on the surface of the rod one. A screw rod two is inserted into the inner cavity of the perforation. Washer nuts two are respectively threadedly connected to both ends of the surface of the screw rod two. One side of the washer nut two abuts against the surface of the perforation. Inner threaded rods are rotatably connected to the surfaces of the washer nuts two through rotating shafts. A bidirectional threaded rod is threadedly connected to the inner cavities of the two inner threaded rods.

[0023] Preferably, an auxiliary handle is fixedly installed in the middle of the surface of the bidirectional threaded rod.

[0024] Preferably, the horizontal adjustment assembly includes a steel plate fixedly connected to the inner cavity of the end of the fixed pipe. A long threaded column is rotatably connected to the middle of the right side of the steel plate, and the round rod is threadedly connected to the surface of the long threaded column.

[0025] Preferably, a first bevel gear is fixedly installed at the left end of the surface of the long threaded column. A rotatable rotating member is penetrated through the surface of the fixed pipe, and a second bevel gear is fixedly installed at the rear end of the rotating member. The second bevel gear meshes with the first bevel gear.

[0026] Preferably, second bevel gears are fixedly connected to both the top and bottom of the right side of the steel plate, and the round rod is slidably connected to the surfaces of the two guide rods.

[0027] Preferably, a plurality of reinforcing ribs are welded between the surface of the fixed pipe and the mounting angle bracket.

[0028] The beneficial effects of the present invention are as follows:

[0029] 1. Through the proposed automatic calculation of the pitch angle and real-time AI recognition technology, the automatic pitch angle adjustment of the present invention avoids the recognition error when the object is at the edge of the picture, improves the accuracy of AI detection, and the real-time recognition technology enables the system to trigger an alarm instantly when a threat appears, meeting the timeliness requirement of transmission line warning, effectively reducing the occurrence of potential damage. The warning system is installed on the angle steel tower by means of diagonal bracing reinforcement, and the position of the warning system can also be horizontally adjusted, improving its installation firmness and adaptability;

[0030] 2. Through the cooperation of the Z-shaped part and the fastening bolt, the present invention can ensure that the mounting angle bracket is installed on the longitudinal angle steel without damaging the structure on the longitudinal angle steel, guaranteeing the integrity of the longitudinal angle steel structure;

[0031] 3. Through the arrangement of the guide rods, when the round rod slides in the inner cavity of the fixed pipe, it will slide on the surfaces of the two guide rods, effectively avoiding the rotation of the round rod in the inner cavity of the fixed pipe and ensuring the stability of the round rod sliding;

[0032] 4. Through the cooperation of the auxiliary rotating block and the threaded rod, the present invention can make rod one move towards and away from rod two, providing convenience for installing the connecting head at the top of rod two into the inner cavity of the connecting sleeve. After installation, it can be tightened to improve the connection strength;

[0033] 5. Through the arrangement of the tying component, the present invention can tie and reinforce the two adjustable diagonal bracing rods, increasing the stability of the connection between the two adjustable diagonal bracing rods and the horizontal angle steel, and further increasing the stability of the overall structure. Description of the Drawings

[0034] Advantages of the above and / or other aspects of the present invention will become clearer and easier to understand through the following detailed description in conjunction with the accompanying drawings, which are merely schematic and do not limit the present invention, wherein:

[0035] Figure 1 Schematic three-dimensional view of an embodiment of the present invention;

[0036] Figure 2 Schematic three-dimensional view of the connection between a round rod and an early warning system in an embodiment of the present invention;

[0037] Figure 3 Schematic exploded view of a longitudinal angle steel and a mounting angle code in an embodiment of the present invention;

[0038] Figure 4 Schematic cross-sectional view of a fixed pipe and a round rod in an embodiment of the present invention;

[0039] Figure 5 Schematic three-dimensional view of a diagonal bracing and reinforcement component in an embodiment of the present invention;

[0040] Figure 6 Schematic exploded view of an adjustable diagonal strut in an embodiment of the present invention;

[0041] Figure 7 For an embodiment of the present invention Figure 5 Partial enlarged view of point A in;

[0042] Figure 8 Schematic diagram of the main process system in an embodiment of the present invention;

[0043] Figure 9 Schematic diagram of the pitching angle of a pan-tilt in an embodiment of the present invention;

[0044] Figure 10 Schematic diagram of the pan-tilt pitching angle calculation system in an embodiment of the present invention.

[0045] In the drawings, the list of components represented by each reference numeral is as follows:

[0046] 1. Round rod, 2. Reinforcing rib, 3. Fixed pipe, 4. Installation angle bracket, 5. Vertical angle steel, 6. Horizontal angle steel, 7. Diagonal bracing and reinforcement assembly, 71. Sector-shaped hoop, 72. Arc-shaped hoop, 73. Adjustable diagonal bracing rod, 731. Rod one, 732. Rod two, 733. Auxiliary rotating block, 734. Threaded rod, 735. Connector, 74. Strip-shaped hole, 75. Screw rod one, 76. Connecting sleeve, 77. Base plate, 78. Gasket nut one, 79. Tie component, 791. Perforation, 792. Screw rod two, 793. Gasket nut two, 794. Inner threaded rod, 795. Bi-directional threaded rod, 796. Auxiliary handle, 8. Horizontal adjustment component, 81. Steel plate, 82. Long threaded column, 83. Bevel gear one, 84. Rotating part, 85. Bevel gear two, 86. Guide rod, 9. Z-shaped part, 10. Fastening bolt, 11. Vertical pipe, 12. Millimeter-level radar, 13. AI identifier, 14. Pan-tilt monitoring device, 15. Installation angle plate. Detailed implementation manners

[0047] In the following, embodiments of the power transmission line warning system of the present invention will be described with reference to the accompanying drawings.

[0048] The embodiments described herein are specific specific implementation manners of the present invention for explaining the concept of the present invention, and are all explanatory and exemplary, and should not be construed as limiting the implementation manners of the present invention and the scope of the present invention. Except for the embodiments described herein, those skilled in the art can also adopt other obvious technical solutions based on the content disclosed in the claims and the specification of the present application, and these technical solutions include technical solutions that make any obvious substitutions and modifications to the embodiments described herein.

[0049] The drawings in this specification are schematic diagrams to assist in explaining the concept of the present invention, and schematically show the shapes of various parts and their mutual relationships. Please note that in order to clearly show the structures of the various components of the embodiments of the present invention, the drawings are not drawn in the same proportion. The same reference numerals are used to represent the same parts.

[0050] Embodiment 1: Figures 1-10 A power transmission line warning system showing an embodiment of the present invention is installed on an angle steel tower through a reinforcement bracket. The warning system includes two millimeter-level radars 12, an AI identifier 13, and a pan-tilt monitoring device 14;

[0051] The millimeter-level radar 12 is used to detect surrounding objects in real time. During installation, the width difference between the power transmission line and the warning device is measured, marked as A here. If the warning area is set within 50 meters of the power transmission line, then for the object x information reported by the radar, when x - A <= 50, it meets the condition for video monitoring and AI identification;

[0052] When an object is detected to enter the preset safe distance, the xy information of the object is reported, and the system automatically rotates the camera to aim at the detected object and starts recording a video according to the reported information; when installing, calculate the distance on flat ground and the distance on a slope. Within the flat ground distance, based on the reported xy and the camera height, assuming the height is H, and the distance between the object and the tower pole is Y, the pitch angle w to be adjusted is:

[0053] W = 90 - arctan(H / Y)

[0054] If the object appears on a sloped surface, the logic for calculating the pitch angle;

[0055] Once the pitch angle is adjusted in place, the video starts recording. While recording and saving the video, each frame of the h264 video frame is sent to the AI recognition module;

[0056] The video is transmitted frame by frame to the shared memory module. The shared memory module is divided into multiple blocks (such as a circular queue), and each block can hold one frame of data. The camera recording module, as the producer, puts the video frames into the idle shared memory blocks, and the AI module, as the consumer, reads the video frames from the shared memory blocks to ensure seamless data flow. YOLOv8 is a lightweight and real-time deep learning model for object detection and classification, supporting input of images with multiple resolutions, and outputting bounding boxes, classes, and confidence levels. The model file is loaded in the pre-trained weight format, based on the PyTorch or ONNXRuntime framework. When initializing the inference engine, set the computing device (such as GPU / TPU or CPU), read the video frames from the shared memory, and decode them into the original image format (such as BGR or RGB), normalize the image, resize it (usually to 640x640 or 416x416), and perform channel swapping to adapt to the input requirements of YOLOv8. Then input the processed image into the YOLOv8 model to obtain the detection results;

[0057] The results output by YOLOv8 include:

[0058] Bounding boxes: The position of the target in the image;

[0059] Classes: The classes of the detected target objects (such as people, cars, drones, etc.);

[0060] Confidence: The confidence level of the model for the detection results;

[0061] The AI recognition software in the system can automatically identify the object category and match it with the preset monitoring objects. If the recognized object belongs to the category that the system needs to monitor, the system will transmit the data to the monitoring server through the 4G network. After receiving the relevant alarm, the server will immediately push the alarm information to the management personnel through the WeChat official account so that they can take corresponding measures quickly.

[0062] The monitoring device includes a solar power supply system, an embedded intelligent recognition system, a 4G communication module, and a varifocal camera. The solar power supply system provides power support for the embedded intelligent recognition system and the spherical camera to ensure the continuous operation of the device in remote areas. The 4G communication module is responsible for transmitting the detection data and recognition results to the monitoring server in a timely manner, while the camera is connected to the embedded intelligent recognition system through the RS485 bus to provide high-quality video data to support AI analysis.

[0063] In addition, the embedded intelligent recognition system also integrates an audio module, which includes an audio power amplifier and a high-pitched horn, and can issue an audio alarm when necessary to enhance the early warning ability of the system in a multi-level manner.

[0064] Refer to Figure 8 The design of the entire system is centered around modularization to ensure the efficient collaborative work of each part. Especially in the AI part, the emphasis is on the automated identification and learning functions, reducing the dependence on manual intervention and achieving true intelligent monitoring and self-optimization.

[0065] Refer to Figure 9 and Figure 10 Description:

[0066] Pan-tilt height: AC

[0067] Segment 1: Distance CD, angle 0;

[0068] Segment 2: Distance DE, angle GDI;

[0069] Target: Distance AG;

[0070] The angle GAJ is the pan-tilt angle to be calculated, and GI is the height of the target.

[0071] Calculation steps:

[0072] Calculation: Distance AD = square root (AC x AC + CD x CD);

[0073] Calculation: Angle ADC = arctan(AC / CD);

[0074] Calculation: Angle ADG, Angle ADG = 180 - Angle ADC - Angle GDI;

[0075] Calculation: Angle AGD = anti sin(sin(angle ADG) / AG * AD);

[0076] Calculation: Angle DAG = 180 - angle AGD – angle ADG;

[0077] Calculation: Since angle DAJ = angle ADC, angle GAJ = angle DAJ - angle DAG;

[0078] The reinforcement bracket includes a longitudinal angle steel 5 arranged on the angle steel tower and a transverse angle steel 6 welded to one side of the longitudinal angle steel 5. A vertical pipe 11 is arranged between two millimeter-level radars 12. Both millimeter-level radars 12 are installed on the surface of the vertical pipe 11 through angle-adjustable clamps. A pan-tilt monitoring device 14 is installed on the top of an AI identifier 13. An installation angle plate 15 is installed at the bottom of the AI identifier 13. A round rod 1 is arranged between the bottom of the installation angle plate 15 and the top of the vertical pipe 11. Both the bottom of the installation angle plate 15 and the top of 16 are installed on the surface of the round rod 1 through angle-adjustable clamps. A fixed pipe 3 is slidably connected to the surface of the round rod 1. One end of the fixed pipe 3 is welded with an installation angle code 4. A plurality of reinforcing ribs 2 are welded between the surface of the fixed pipe 3 and the installation angle code 4. Both ends of the installation angle code 4 are installed with Z-shaped members 9 through bolts. The Z-shaped member 9 is sleeved on one side of the longitudinal angle steel 5. The longitudinal angle steel 5 abuts against the Z-shaped member 9 which tightly abuts against one side of the longitudinal angle steel 5. Four fastening bolts 10 threadedly connected to the Z-shaped member 9 are penetrated through the Z-shaped member 9. The end of the fastening bolt 10 abuts against the longitudinal angle steel 5. By the combined use of the Z-shaped member 9 and the fastening bolt 10, the installation angle code 4 can be installed on the longitudinal angle steel 5 without damaging the structure of the longitudinal angle steel 5, ensuring the integrity of the structure of the longitudinal angle steel 5. An inclined strut reinforcement assembly 7 is arranged between the surface of the fixed pipe 3 and the transverse angle steel 6. The inclined strut reinforcement assembly 7 includes a sector-shaped clamp 71 sleeved on the surface of the fixed pipe 3 and an arc-shaped clamp 72. Both ends of the sector-shaped clamp 71 and both ends of the arc-shaped clamp 72 are fixedly installed through bolts. Adjustable inclined strut rods 73 are installed on the surfaces of the bolts at both ends of the sector-shaped clamp 71 and the arc-shaped clamp 72. Two strip holes 74 are opened on both sides of the top of the transverse angle steel 6. A screw rod one 75 is slidably connected to the inner cavity of the strip hole 74. The bottoms of the two screw rods one 75 are fixedly connected with a connecting sleeve 76. The inner cavity of the connecting sleeve 76 is movably installed with the adjustable inclined strut rod 73 through bolts. Washers 77 are sleeved on the surfaces of the two screw rods one 75. A gasket nut one 78 threadedly connected to the surface of the screw rod one 75 is arranged on the top of the washer 77. A tie-in assembly 79 is arranged between the two adjustable inclined strut rods 73. A horizontal adjustment assembly 8 is arranged between the reinforcing rib 2 and the fixed pipe 3.

[0079] Embodiment 2: It is basically the same as Embodiment 1. Further, the adjustable diagonal brace 73 includes a rod one 731 and a rod two 732. Connecting heads 735 are provided at the opposite ends of the rod one 731 and the rod two 732. The connecting head 735 at the top is installed in the inner cavity of the connecting sleeve 76, and the connecting head 735 at the bottom is installed on the surface of the bolt between the sector-shaped hoop 71 and the arc-shaped hoop 72. Threaded rods 734 are threadedly connected to the inner cavities on the opposite sides of the rod one 731 and the rod two 732. An auxiliary rotating block 733 is welded between the two threaded rods 734. By the combined use of the auxiliary rotating block 733 and the threaded rods 734, the rod one 731 and the rod two 732 can move towards and away from each other, providing convenience for installing the connecting head 735 at the top of the rod two 732 in the inner cavity of the connecting sleeve 76. After installation, it can be operated to tighten, improving the connection strength.

[0080] Embodiment 3: It is basically the same as Embodiment 1. Further, the tie component 79 includes a perforation 791 opened on the surface of the rod one 731. A screw rod two 792 is inserted into the inner cavity of the perforation 791. Gasket nuts two 793 are threadedly connected to both ends of the surface of the screw rod two 792. One side of the gasket nut two 793 abuts against the surface of the perforation 791. Inner threaded rods 794 are rotatably connected to the surface of the gasket nut two 793 through a rotating shaft. A bidirectional threaded rod 795 is threadedly connected to the inner cavities of the two inner threaded rods 794. An auxiliary handle 796 is fixedly installed in the middle of the surface of the bidirectional threaded rod 795. Through the setting of the tie component 79, the two adjustable diagonal braces 73 can be tied and reinforced, increasing the stability of the connection between the two adjustable diagonal braces 73 and the horizontal angle steel 6, and further increasing the stability of the overall structure.

[0081] Embodiment 4: It is basically the same as Embodiment 1. Further, the horizontal adjustment component 8 includes a steel plate 81 fixedly connected to the inner cavity of the end of the fixed pipe 3. A long threaded column 82 is rotatably connected to the middle of the right side of the steel plate 81. The round rod 1 is threadedly connected to the surface of the long threaded column 82. A bevel gear one 83 is fixedly installed at the left end of the surface of the long threaded column 82. A rotatable rotating member 84 is penetrated through the surface of the fixed pipe 3. A bevel gear two 85 is fixedly installed at the rear end of the rotating member 84. The bevel gear two 85 meshes with the bevel gear one 83. Bevel gears two 85 are fixedly connected to the top and bottom of the right side of the steel plate 81. The round rod 1 is slidably connected to the surfaces of the two guide rods 86. Through the setting of the guide rods 86, when the round rod 1 slides in the inner cavity of the fixed pipe 3, it will slide on the surfaces of the two guide rods 86, effectively preventing the round rod 1 from rotating in the inner cavity of the fixed pipe 3 and ensuring the stability of the sliding of the round rod 1.

[0082] Reinforcement Bracket Installation Principle: Move the mounting angle bracket 4 on the connecting fixed pipe 3 close to the longitudinal angle steel 5. Then slide the Z-shaped part 9 until the Z-shaped part 9 abuts against the longitudinal angle steel 5. Tighten the bolt between the Z-shaped part 9 and the mounting angle bracket 4. Then turn the fastening bolt 10 until the fastening bolt 10 firmly abuts against the fixed pipe 3, realizing the installation of the mounting angle bracket 4 and the longitudinal angle steel 5. Then at the connection position, two millimeter-level radars 12 are installed on the surface of the vertical pipe 11 through angle-adjustable clamps. The bottom of the AI identifier 13 is installed with a mounting angle plate 15. Both the bottom of the mounting angle plate 15 and the top of the vertical pipe 11 are installed on the surface of the round rod 1 through angle-adjustable clamps. Then put the sector clamp 71 and the arc clamp 72 on the end of the surface of the fixed pipe 3, fasten them with two bolts and install the adjustable diagonal brace 73 on the surface of the bolts. Install the connecting sleeve 76 inside the transverse angle steel 6 with the backing plate 77 and the gasket nut 1 78. Rotate the other end of the adjustable diagonal brace 73 into the inner cavity of the connecting sleeve 76 and install it with bolts. Through the operation of the auxiliary rotating block 733 and the threaded rod 734, the installation convenience can be improved. Then insert the screw rod 2 792 into the inner cavity of the through hole 791 and tighten the gasket nut 2 793. Operate the bidirectional threaded rod 795 to rotate, so that the two inner threaded rods 794 move towards each other, and the two adjustable diagonal braces 73 are tied by the screw rod 2 792 to improve the stability.

[0083] In summary, for this power transmission line warning system, through the proposed automatic pitch angle calculation and real-time AI recognition technology, the automatic pitch angle adjustment avoids the recognition error when the object is at the edge of the picture, improves the accuracy of AI detection, and the real-time recognition technology enables the system to trigger an alarm instantly when a threat appears, meeting the timeliness requirement of power transmission line warning and effectively reducing the occurrence of potential damages. This warning system is installed on the angle steel tower by means of diagonal brace reinforcement, and the position of this warning system can also be adjusted horizontally, improving its installation firmness and adaptability.

[0084] The disclosed technical features are not limited to the combinations with other disclosed features. Those skilled in the art can also make other combinations among the technical features according to the purpose of the invention, subject to the purpose of realizing the invention.

Claims

1. A transmission line early warning system, characterized in that: The early warning system is installed through a reinforced bracket and an angle steel tower, and the early warning system includes two millimeter-level radars (12), an AI identifier (13) and a pan-tilt monitoring device (14); The millimeter-level radar (12) is used to detect surrounding objects in real time. When installed, the width difference between the power transmission line and the early warning device is measured, which is marked as A. If the warning zone is set within 50 meters of the power transmission line, then the object x information reported by the radar, when xA<=50, meets the conditions for video surveillance AI recognition; When an object is detected to enter the preset safety distance, the xy information of the object is reported. The system automatically rotates the camera to aim at the detected object and starts recording video based on the reported information. During installation, calculate the distance on flat ground and the distance on slope. Within the distance on flat ground, based on the reported xy and camera height, assuming the height is H and the distance between the object and the tower is Y, the pitch angle w that needs to be adjusted is: W = 90-arctan (H / Y) Logic for calculating the pitch angle if the object appears on a sloped surface; When the pitch angle is adjusted to the right position, the video recording starts. While recording and saving the video, each h264 video frame is sent to the AI ​​recognition module; The video frames are transmitted to the shared memory module frame by frame. The shared memory module is divided into multiple blocks (such as a circular queue), each of which can hold one frame of data. The camera recording module acts as a producer and puts the video frames into the free shared memory blocks. The AI ​​module acts as a consumer and reads the video frames from the shared memory blocks to ensure seamless data flow transmission. YOLOv8 is a lightweight, real-time deep learning model for target detection and classification. It supports input of images of multiple resolutions and outputs bounding boxes, classes, and confidence. The model file is loaded in the pre-trained weight format based on the PyTorch or ONNXRuntime framework. When the inference engine is initialized, the computing device (such as GPU / TPU or CPU) is set, the video frames are read from the shared memory and decoded into the original image format (such as BGR or RGB), the images are normalized, resized (usually 640x640 or 416x416), and the channels are swapped to adapt to the input requirements of YOLOv8, and the processed images are input into the YOLOv8 model to obtain the detection results. The results output by YOLOv8 include: Bounding box: the location of the object in the image; Category: the category of the detected target object (such as people, cars, drones, etc.); Confidence: The confidence level of the model in the detection results; The AI ​​recognition software in the system can automatically identify the category of objects and match them with the preset monitoring objects. If the identified object belongs to the category that the system needs to monitor, the system will transmit the data to the monitoring server through the 4G network. After receiving the relevant alarm, the server will immediately push the alarm information to the management personnel through the WeChat public account so that they can take quick response measures. The monitoring equipment includes a solar power supply system, an embedded intelligent recognition system, a 4G communication module and a zoom camera. The solar power supply system provides power support for the embedded intelligent recognition system and the dome camera to ensure the continuous operation of the equipment in remote areas. The 4G communication module is responsible for transmitting the detection data and recognition results to the monitoring server in a timely manner, while the camera is connected to the embedded intelligent recognition system through the RS485 bus to provide high-quality video data to support AI analysis. In addition, the embedded intelligent recognition system also integrates an audio module, including an audio power amplifier and a tweeter, which can issue an audio alarm when necessary, enhancing the system's early warning capabilities in a multi-level manner; The design of the entire system is centered on modularization to ensure efficient collaboration of all parts, especially in the AI ​​part, which emphasizes automated identification and learning functions, reduces reliance on human intervention, and achieves truly intelligent monitoring and self-optimization. The reinforcement bracket comprises a longitudinal angle steel (5) arranged on an angle steel tower and a transverse angle steel (6) welded to one side of the longitudinal angle steel (5); a vertical pipe (11) is arranged between the two millimeter-level radars (12); the two millimeter-level radars (12) are mounted on the surface of the vertical pipe (11) through an angle-adjustable clamp; the pan-tilt monitoring device (14) is mounted on the top of the AI ​​identifier (13); a mounting angle plate (15) is mounted on the bottom of the AI ​​identifier (13); a round rod (1) is arranged between the bottom of the mounting angle plate (15) and the top of the vertical pipe (11); the mounting angle plate (15) is provided with a plurality of round rods (1) disposed on the bottom of the mounting angle plate (15) and the top of the vertical pipe (11); The bottom and the top of (16) are both mounted on the surface of the round rod (1) through an angle-adjustable clamp. The surface of the round rod (1) is slidably connected with a fixing tube (3). One end of the fixing tube (3) is welded with a mounting angle code (4). Both ends of the mounting angle code (4) are mounted with Z-shaped parts (9) through bolts. The Z-shaped part (9) is sleeved on one side of the longitudinal angle steel (5). The longitudinal angle steel (5) is pressed against the Z-shaped part (9) and is tightly pressed against one side of the longitudinal angle steel (5). Four fastening bolts (10) threadedly connected to the Z-shaped part (9) are provided through the Z-shaped part (9). The fastening bolts (10) are The end portion is against the longitudinal angle steel (5), and a diagonal brace reinforcement assembly (7) is arranged between the surface of the fixed tube (3) and the transverse angle steel (6). The diagonal brace reinforcement assembly (7) comprises a fan-shaped clamp (71) and an arc-shaped clamp (72) sleeved on the surface of the fixed tube (3). Both ends of the fan-shaped clamp (71) and both ends of the arc-shaped clamp (72) are fixedly installed by bolts. The surfaces of the bolts at both ends of the fan-shaped clamp (71) and the arc-shaped clamp (72) are installed with adjustable diagonal brace rods (73). Two strip holes (74) are provided on both sides of the top of the transverse angle steel (6). The strip holes (74) The inner cavity is slidably connected to a screw rod (75), the bottoms of the two screw rods (75) are fixedly connected to a connecting sleeve (76), the inner cavity of the connecting sleeve (76) is movably installed with the adjustable diagonal support rod (73) through bolts, the surfaces of the two screw rods (75) are sleeved with a pad (77), the top of the pad (77) is provided with a gasket nut (78) threadedly connected to the surface of the screw rod (75), a tie assembly (79) is provided between the two adjustable diagonal support rods (73), and a horizontal adjustment assembly (8) is provided between the reinforcing rib (2) and the fixed tube (3).

2. A transmission line early warning system according to claim 1, characterized in that: The adjustable diagonal brace rod (73) comprises a rod 1 (731) and a rod 2 (732), and the ends of the rod 1 (731) and the rod 2 (732) opposite to each other are both provided with a connecting head (735), the connecting head (735) at the top is installed in the inner cavity of the connecting sleeve (76), and the connecting head (735) at the bottom is installed on the surface of the bolt between the fan-shaped clamp (71) and the arc-shaped clamp (72).

3. A transmission line early warning system according to claim 2, characterized in that: The inner cavities on the opposite sides of the rod one (731) and the rod two (732) are both threadedly connected with threaded rods (734), and an auxiliary rotating block (733) is welded between the two threaded rods (734).

4. A transmission line early warning system according to claim 3, characterized in that: The tie assembly (79) includes a through hole (791) opened on the surface of rod one (731), the inner cavity of the through hole (791) is inserted with a screw rod two (792), both ends of the surface of the screw rod two (792) are threadedly connected with a gasket nut two (793), one side of the gasket nut two (793) is against the surface of the through hole (791), the surface of the gasket nut two (793) is rotatably connected with an internal threaded rod (794) through a rotating shaft, and the inner cavities of the two internal threaded rods (794) are threadedly connected with a bidirectional threaded rod (795).

5. A transmission line early warning system according to claim 4, characterized in that: An auxiliary handle (796) is fixedly mounted in the middle of the surface of the bidirectional threaded rod (795).

6. A transmission line early warning system according to claim 5, characterized in that: The horizontal adjustment assembly (8) comprises a steel plate (81) fixedly connected to the inner cavity of the end of the fixed tube (3); a long threaded column (82) is rotatably connected to the middle of the right side of the steel plate (81); and the round rod (1) is threadedly connected to the surface of the long threaded column (82).

7. A transmission line early warning system according to claim 6, characterized in that: A bevel gear 1 (83) is fixedly mounted on the left end of the surface of the long threaded column (82), a rotatable rotating member (84) is penetrated through the surface of the fixed tube (3), a bevel gear 2 (85) is fixedly mounted on the rear end of the rotating member (84), and the bevel gear 2 (85) is meshed with the bevel gear 1 (83).

8. A transmission line early warning system according to claim 7, characterized in that: The top and bottom of the right side of the steel plate (81) are fixedly connected with a bevel gear 2 (85), and the round rod (1) is slidably connected to the surfaces of the two guide rods (86).

9. A transmission line early warning system according to claim 8, characterized in that: A plurality of reinforcing ribs (2) are welded between the surface of the fixing tube (3) and the mounting angle code (4).

Images