Multi-source data fusion processing method based on airborne terminal of unmanned aerial vehicle

Abstract

The invention relates to the technical field of multi-source data fusion processing based on an unmanned aerial vehicle airborne terminal, and discloses a multi-source data fusion processing method based on an unmanned aerial vehicle airborne terminal, and the method comprises the following steps: employing an unmanned aerial vehicle to carry out the inspection of a distribution network line channel and a distribution network tower; and an AI intelligent analysis platform and an electric power inspection comprehensive management platform are adopted to carry out unmanned aerial vehicle inspection auxiliary control. According to the multi-source data fusion processing method based on the airborne terminal of the unmanned aerial vehicle, the route is automatically generated through point cloud calculation by one key, the information of the inspection route of the unmanned aerial vehicle is calculated, the inspection route synchronization of the unmanned aerial vehicle can be quickly executed, and the AI image recognition platform is matched with thermal infrared inspection to perform data information integration; the method has the advantages of expanding single defect hidden danger table data, providing enough data for construction of a hidden danger management table report, improving the accuracy of the data and the like, and solves the problems that a single identification airborne terminal can only identify a certain type of defects and cannot carry out comprehensive defect integrated management.

Description

Technical Field

[0001] The present invention relates to the technical field of multi-source data fusion processing based on an unmanned aerial vehicle airborne terminal, and in particular to a multi-source data fusion processing method based on an unmanned aerial vehicle airborne terminal. Background Art

[0002] The distribution network refers to a power grid that receives electric energy from the transmission network or regional power plant, and distributes it locally or step by step according to the voltage to various users through distribution facilities. It is a network composed of overhead lines, cables, poles, distribution transformers, disconnectors, reactive power compensators and some ancillary facilities, and plays an important role in distributing electric energy in the power grid. Distribution network lines are generally exposed outdoors, and they will inevitably be affected by the natural environment and social environment during operation. Extreme weather such as lightning and storms may destroy or damage distribution network equipment, or cause accidents such as electric shock. In the process of urban and rural integration, there are problems such as unreasonable greening plant construction and illegal building construction; artificial theft, random wiring, and deliberate destruction of distribution network lines. These problems will affect the normal operation of distribution equipment and even lead to safety accidents and legal risks.

[0003] Due to the high reliability requirements of distribution network equipment, there are few power outage test items for equipment. Daily inspections are the main means of equipment status control. Operation and maintenance personnel have an urgent need for effective and efficient inspection methods. During the distribution network inspection process, the daily inspection work will waste a lot of manpower, and the inspection process is inevitably wrong. The inspection environment is harsh, and sometimes it causes harm to the personal safety of the inspectors. Therefore, drones have begun to be used for distribution network line inspections.

[0004] During drone inspections, automatic identification and intelligent diagnosis of equipment faults are particularly important. At present, domestic research in the field of power image diagnosis mainly focuses on the detection and identification of common line defect types or specific problems, such as broken strands of transmission lines, screw detachment, damaged insulators, damaged anti-vibration hammers, damaged wire clamps, and positioning of spacers. If other situations occur, identification errors will occur, which leads to low accuracy in distribution network defect identification. A single identification airborne terminal can only identify a certain type of defect and cannot perform comprehensive integrated defect management.

[0005] Therefore, the present invention provides a multi-source data fusion processing method based on an unmanned aerial vehicle airborne terminal to solve the above-mentioned problems. Summary of the invention

[0006] 1. Technical issues to be resolved In view of the shortcomings of the prior art, the present invention provides a multi-source data fusion processing method based on an unmanned aerial vehicle airborne terminal, which uses point cloud solution to automatically generate a route with the LiPowerline software with one click; uses an AI image recognition platform to cooperate with thermal infrared inspections to integrate data information, expands the data of a single defect hidden danger table, provides sufficient data for the construction of a hidden danger management form report, and improves the accuracy of the data, etc., which solves the problem that a single identification airborne terminal can only identify a certain type of defect and cannot perform comprehensive integrated defect management.

[0007] (II) Technical solution In order to realize the above-mentioned one-key automatic generation of routes by LiPowerline software through point cloud solution, calculate the information of drone inspection routes, reduce the amount of data transmission, and quickly execute drone inspection route synchronization; use AI image recognition platform to cooperate with thermal infrared inspection to integrate data information, expand the data of single defect hidden danger table, provide sufficient data for the construction of hidden danger management table report, improve data accuracy, etc., solve the problem that a single identification airborne terminal can only identify a certain type of defect and cannot perform comprehensive defect integrated management. The present invention provides the following technical solutions: The multi-source data fusion processing method based on the UAV airborne terminal includes the following steps: Step 1: Use drones to inspect distribution network line channels and distribution network towers; Step 2: Use AI intelligent analysis platform and power inspection integrated management platform to integrate multi-source data to assist in inspection control of drones; Step 3: First, use point cloud solution to conduct detailed inspection route planning and analysis of dangerous points when the drone is conducting inspections; Through point cloud solution, LiPowerline software can automatically generate routes with one click, calculate the information of drone inspection routes, reduce the amount of data transmission, and quickly perform drone inspection route synchronization; Step 4: Then use the drone to perform AI image recognition for network distribution; Step 5: Finally, conduct thermal infrared inspection by drone; Step 6: Then integrate the data obtained in steps 4 and 5 to complete the generation of the hidden danger management form report; The AI ​​image recognition platform is used in conjunction with thermal infrared inspections to integrate data information, expand the data of a single defect hidden danger table, provide sufficient data for the construction of hidden danger management form reports, and improve data accuracy. This solves the problem that a single identification airborne terminal can only identify a certain type of defect and cannot perform comprehensive defect integrated management.

[0008] Preferably, the AI ​​intelligent analysis platform provides auxiliary work such as autonomous calibration of drone inspection defects, defect identification, and model training.

[0009] Preferably, the electric power inspection integrated management platform performs integrated management of inspection equipment, automatically generates inspection routes, manages inspection tasks, and monitors the operating status of drones in real time.

[0010] Preferably, the UAV itself is equipped with high-precision geometric calibration and time synchronization technology, GNSS / IMU combined navigation algorithm and robot continuous synchronous positioning and navigation SLAM algorithm, which fuses multi-source data to obtain the precise positioning of the UAV.

[0011] Preferably, the point cloud solution in step three includes adaptive scenario-based point cloud solution technology, point cloud processing algorithm and big data and three-dimensional feature enhancement algorithm, and point cloud data is acquired through laser scanning.

[0012] Preferably, the AI ​​image recognition in step 4 includes insulator self-explosion recognition, hardware corrosion recognition, bolt missing recognition and foreign body monitoring recognition.

[0013] Preferably, the drone also includes cloud-based route loading, breakpoint resume flight, and tower marking.

[0014] (III) Beneficial effects Compared with the prior art, the present invention provides a multi-source data fusion processing method based on an unmanned aerial vehicle airborne terminal, which has the following beneficial effects: 1. This multi-source data fusion processing method based on the UAV airborne terminal uses point cloud solution to automatically generate routes with the LiPowerline software, calculate the information of the UAV inspection route, reduce the amount of data transmission, and can quickly execute the synchronization of UAV inspection routes.

[0015] 2. The multi-source data fusion processing method based on the UAV airborne terminal integrates data information by using the AI ​​image recognition platform in conjunction with thermal infrared inspections, expands the data of a single defect hidden danger table, provides sufficient data for the construction of hidden danger management form reports, and improves data accuracy. It solves the problem that a single identification airborne terminal can only identify a certain type of defect and cannot perform comprehensive defect integrated management. DETAILED DESCRIPTION

[0016] The following will be combined with the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0017] The multi-source data fusion processing method based on the UAV airborne terminal includes the following steps: Step 1: Use drones to inspect distribution network line channels and distribution network towers; Step 2: Use AI intelligent analysis platform and power inspection integrated management platform to integrate multi-source data to assist in inspection control of drones; Step 3: First, use point cloud solution to conduct detailed inspection route planning and analysis of dangerous points when the drone is conducting inspections; Step 4: Then use the drone to perform AI image recognition for network distribution; Step 5: Finally, conduct thermal infrared inspection by drone; Step six: Then integrate the data obtained from steps four and five to complete the generation of the hidden danger management form report. Use the AI ​​image recognition platform in conjunction with thermal infrared inspections to integrate data information, expand the data of a single defect hidden danger form, and provide sufficient data for the construction of the hidden danger management form report. Improve the accuracy of the data and other advantages, and solve the problem that a single identification airborne terminal can only identify a certain type of defect and cannot perform comprehensive defect integrated management.

[0018] The AI ​​intelligent analysis platform provides auxiliary work for autonomous calibration of drone inspection defects, defect identification, and model training; the comprehensive power inspection management platform conducts comprehensive management of inspection equipment, automatically generates inspection routes, manages inspection tasks, and monitors the operating status of drones in real time.

[0019] The drone itself is equipped with high-precision geometric calibration and time synchronization technology, GNSS / IMU combined navigation algorithm and robot continuous synchronous positioning and navigation SLAM algorithm, which integrates multi-source data to obtain the precise positioning of the drone; Point cloud solution includes adaptive scenario-based point cloud solution technology, point cloud processing algorithm and big data and 3D feature enhancement algorithm. Through point cloud solution, LiPowerline software can automatically generate routes with one click, calculate the information of drone inspection routes, reduce the amount of data transmission, and quickly execute drone inspection route synchronization; Point cloud solution can mainly use LiGeoreference software and LiFuser-BP software; Point cloud processing can mainly use LiDar360 software and LiCloud software; AI image recognition includes insulator self-explosion recognition, hardware corrosion recognition, bolt missing recognition and foreign body monitoring recognition; The drone also includes cloud-based route loading, breakpoint resumption, and tower marking.

[0020] First, data is collected through drones, and point cloud solutions are used to analyze dangerous points in the distribution network and plan the routes of drones. Use laser scanning to obtain point cloud data, use point cloud solution to complete detailed inspection route planning, and facilitate automatic inspection by drones; Analyze the dangerous points of the distribution network to obtain the dangerous point report output form; The generated route data is transmitted to the drone hangar through the data transmission platform, and the inspection drone is arranged to conduct inspection. The inspection drone is equipped with AI image recognition terminal and thermal infrared inspection terminal; The data obtained by the AI ​​image recognition terminal and the thermal infrared inspection terminal are collected through the comprehensive power inspection management platform, and the data is classified and defect identification and defect detail reports are made, such as: whether the insulator is intact; whether the hardware is rusted; whether the pole root is normal; whether the wire is intact; whether there are foreign objects.

[0021] The beneficial effects of the present invention are as follows: the multi-source data fusion processing method based on the UAV airborne terminal can automatically generate the route with one click of LiPowerline software through point cloud solution, calculate the information of the UAV inspection route, reduce the data transmission volume, and can quickly execute the synchronization of the UAV inspection route. By utilizing the AI ​​image recognition platform in conjunction with thermal infrared inspection to integrate data information, the data of a single defect hidden danger table is expanded to provide sufficient data for the construction of the hidden danger management table report, improve the accuracy of the data, etc., which solves the problem that a single identification airborne terminal can only identify a certain type of defect and cannot perform comprehensive integrated management of defects.

[0022] Working principle: First, data is collected through drones, and point cloud solutions are used to analyze dangerous points in the distribution network and plan the routes of drones. Use laser scanning to obtain point cloud data, use point cloud solution to complete detailed inspection route planning, and facilitate automatic inspection by drones; Analyze the dangerous points of the distribution network to obtain the dangerous point report output form; The generated route data is transmitted to the drone hangar through the data transmission platform, and the inspection drone is arranged to conduct inspection. The inspection drone is equipped with AI image recognition terminal and thermal infrared inspection terminal; The data obtained by the AI ​​image recognition terminal and the thermal infrared inspection terminal are collected through the comprehensive power inspection management platform, and the data is classified and defect identification and defect detail reports are made, such as: whether the insulator is intact; whether the hardware is rusted; whether the pole root is normal; whether the wire is intact; whether there are foreign objects.

[0023] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A multi-source data fusion processing method based on an unmanned aerial vehicle airborne terminal is characterized in that: The following steps are involved: Step 1: Use drones to inspect distribution network line channels and distribution network towers; Step 2: Use AI intelligent analysis platform and power inspection integrated management platform to integrate multi-source data to assist in inspection control of drones; Step 3: First, use point cloud solution to conduct detailed inspection route planning and analysis of dangerous points when the drone is conducting inspections; Step 4: Then use the drone to perform AI image recognition for network distribution; Step 5: Finally, conduct thermal infrared inspection by drone; Step 6: Then integrate the data obtained in steps 4 and 5 to complete the generation of the hidden danger management form report.

2. The multi-source data fusion processing method based on the UAV airborne terminal according to claim 1 is characterized in that: The AI ​​intelligent analysis platform provides auxiliary work for autonomous calibration of drone inspection defects, defect identification, and model training.

3. The multi-source data fusion processing method based on the UAV airborne terminal according to claim 1 is characterized in that: The electric power inspection integrated management platform comprehensively manages inspection equipment, automatically generates inspection routes, manages inspection tasks, and monitors the operating status of drones in real time.

4. The multi-source data fusion processing method based on the UAV airborne terminal according to claim 1 is characterized in that: The UAV itself is equipped with high-precision geometric calibration and time synchronization technology, GNSS / IMU combined navigation algorithm and robot continuous synchronous positioning and navigation SLAM algorithm, which fuses multi-source data to obtain the precise positioning of the UAV.

5. The multi-source data fusion processing method based on the UAV airborne terminal according to claim 1 is characterized in that: The point cloud solution in step three includes adaptive scenario-based point cloud solution technology, point cloud processing algorithm, and big data and three-dimensional feature enhancement algorithm, and point cloud data is acquired through laser scanning.

6. The multi-source data fusion processing method based on the UAV airborne terminal according to claim 1 is characterized in that: The AI ​​image recognition in step 4 includes insulator self-explosion recognition, hardware corrosion recognition, bolt missing recognition and foreign body monitoring recognition.

7. The multi-source data fusion processing method based on the UAV airborne terminal according to claim 1 is characterized in that: The drone also includes cloud-based route loading, breakpoint resumption, and tower marking.