Distributed AI Intelligent Remote Inspection Method and System Based on Edge Computing

Abstract

The present invention relates to the field of intelligent remote inspection technology, a distributed AI intelligent remote inspection method and system based on edge computing, including: obtaining an inspection area set, sorting the inspection area set by area, obtaining an ordered inspection area set, extracting an ordered inspection area from the ordered inspection area set in turn, and performing the following operations on the extracted ordered inspection areas: obtaining the electromagnetic interference intensity of the ordered inspection area, setting the optimal inspection path, confirming the state of the lighting module, performing line inspection, obtaining a line image set, identifying insulation damage points on the correction image, obtaining a damaged area set, obtaining a voltage intensity set of the damaged area set, repairing the damaged area, obtaining a repair area, constructing a repair database based on the repair area group set, and completing distributed AI intelligent remote inspection based on edge computing based on the repair database. The present invention can improve the efficiency and reliability of remote inspection.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent remote inspection, and particularly to a distributed AI intelligent remote inspection method, system, electronic device and computer-readable storage medium based on edge computing. Background Art

[0002] Edge computing is a technology that processes, stores, and analyzes data as close as possible to the data source. Distributed AI is a technology that distributes the computing tasks of an AI model to multiple computing nodes. Intelligent remote inspection is a method of using artificial intelligence technology to automatically inspect and monitor remote devices or systems.

[0003] Traditional remote inspection methods have problems such as insufficient real-time performance, high bandwidth pressure, high maintenance costs, and lack of intelligence, making it difficult to meet the requirements of complex scenarios. Therefore, how to improve the efficiency and reliability of remote inspection is a technical problem that urgently needs to be solved. Summary of the Invention

[0004] The present invention provides a distributed AI intelligent remote inspection method and a computer-readable storage medium based on edge computing, and its main purpose is to improve the intelligence level of titanium rod high-temperature testing and reduce the excessive consumption of time and human resources.

[0005] To achieve the above object, a distributed AI intelligent remote inspection method based on edge computing provided by the present invention includes:

[0006] Obtain a set of inspection areas, and use a preset inspection order to sort the set of inspection areas to obtain an ordered set of inspection areas. Among them, an ordered inspection area includes a substation, and the substation includes: multiple transmission lines;

[0007] Successively extract an ordered inspection area from the ordered set of inspection areas, and perform the following operations on each of the extracted ordered inspection areas:

[0008] Obtain the electromagnetic interference intensity of the ordered inspection area, and set the optimal inspection path according to the electromagnetic interference intensity;

[0009] Collect the ambient brightness, confirm the status of the lighting module according to the ambient brightness to obtain the lighting module status. Among them, the lighting module includes: a decrement counter, an all-zero detector, and a period counter;

[0010] Receive an intelligent remote inspection instruction, and perform line inspection on the multiple transmission lines in the ordered inspection area according to the intelligent remote inspection instruction, the lighting module status, and the optimal inspection path to obtain a set of line images. Among them, one line image includes one or more transmission lines;

[0011] Perform distortion correction on each line image in the line image set to obtain a corrected image set. Sequentially extract a corrected image from the corrected image set, and perform the following operations on each of the extracted corrected images:

[0012] Identify the insulation skin break points in the corrected image to obtain a set of damaged areas, and obtain a set of voltage intensities of the damaged area set, where the damaged area set includes one or more damaged areas;

[0013] If it is confirmed that there is a voltage intensity in the voltage intensity set that is greater than the preset voltage intensity threshold, send a pre-constructed alarm signal to the pre-constructed maintenance unit. When the maintenance unit receives the alarm signal, use the maintenance unit to repair the damaged area to obtain a repaired area, and summarize the repaired areas to obtain a set of repaired areas;

[0014] Summarize the set of repaired area groups corresponding to the orderly inspection areas to obtain a set of repaired area groups;

[0015] Construct a maintenance database based on the set of repaired area groups, and complete distributed AI intelligent remote inspection based on edge computing based on the maintenance database.

[0016] Optionally, the obtaining of the electromagnetic interference intensity of the orderly inspection area includes:

[0017] Obtain the regional environmental parameters and regional current intensity of the orderly inspection area, and obtain the regional air density value based on the regional environmental parameters, where the regional environmental parameters include: regional temperature and regional air pressure;

[0018] Judge whether the regional air density value is within the preset theoretical density interval;

[0019] If it is confirmed that the regional air density value is within the theoretical density interval, use the regional current intensity to correct the regional air density value to obtain a corrected density air value;

[0020] Calculate the electromagnetic interference intensity of multiple transmission lines in the orderly inspection area using the corrected density air value, where the calculation formula of the electromagnetic interference intensity is as follows:

[0021] ,

[0022] ,

[0023] Among them, represents the corrected density air value, represents the preset standard temperature, represents the regional temperature, represents the regional air pressure, represents the preset standard air pressure, represents the regional current intensity, represents the electromagnetic interference intensity, represents the preset roughness coefficient of the insulating skin, represents the preset radius of the transmission line.

[0024] Optionally, setting the optimal inspection path according to the electromagnetic interference intensity includes:

[0025] Obtaining the electromagnetic intensity attenuation value according to the electromagnetic interference intensity, and determining the inspection field of view length and inspection field of view width of the pre-built unmanned aerial vehicle (UAV). Among them, the UAV includes: a camera, and the inspection field of view length and inspection field of view width are expressed as:

[0026] ,

[0027] ,

[0028] ,

[0029] Among them, represents the inspection field of view length, represents the inspection field of view width, represents the proportional adjustment coefficient, represents the length of the camera, represents the width of the camera, represents the preset flight altitude, represents the relative position height difference of the camera, represents the electromagnetic intensity attenuation value, represents the electromagnetic interference demarcation point, represents the line measurement distance;

[0030] Calculating the optimal inspection path of the ordered inspection area by using the pre-built optimal path formula, inspection field of view length and inspection field of view width.

[0031] Optionally, the optimal path formula is as follows:

[0032] ,

[0033] Among them, represents the optimal inspection path, represents the total inspection length of the UAV, represents the line spacing of the transmission line, represents the number of transmission lines.

[0034] Optionally, confirming the status of the lighting module according to the ambient brightness to obtain the lighting module status includes:

[0035] Obtaining the ambient brightness value according to the ambient brightness;

[0036] If the environmental brightness value is within a preset standard inspection brightness range, obtain an enabling signal, turn on the lighting module based on the enabling signal, and after confirming that the turned-on lighting module is the preset turned-on lighting module, confirm the all-zero detector as the initial state indicator;

[0037] If the environmental brightness value is not within a preset standard inspection brightness range, obtain a disabling signal, turn off the lighting module based on the disabling signal, and after confirming that the turned-off lighting module is the preset turned-off lighting module, confirm the all-zero detector as the initial state indicator;

[0038] Set the initial value of the decrement counter based on the initial state indicator and the cycle counter to obtain the initial value J of the decrement counter, and perform a decrement operation on the initial value J of the decrement counter using a pre-constructed divided-frequency clock signal to obtain a decremented output value;

[0039] If the decremented output value is not zero, use the decremented output value as the initial value J of the decrement counter and return to the step of performing a decrement operation on the initial value J of the decrement counter using the pre-constructed divided-frequency clock signal until the decremented output value is zero, then confirm the initial state indicator as the target all-zero detector, and confirm the status of the lighting module based on the target all-zero detector.

[0040] Optionally, performing distortion correction on each line image in the line image set to obtain a corrected image set includes:

[0041] Extract a line image from the line image set in sequence, and perform the following operations on the extracted line image:

[0042] Construct a rectangular coordinate system based on the line image, obtain the coordinates of each pixel point in the line image according to the rectangular coordinate system to obtain a pixel point coordinate set;

[0043] Extract a pixel point coordinate from the pixel point coordinate set in sequence, and perform the following operations on the extracted pixel point coordinate:

[0044] Perform distortion correction on the pixel point coordinate to obtain a corrected pixel point coordinate, where the corrected pixel point coordinate is expressed as:

[0045] ,

[0046] where, represents the corrected pixel point coordinate, represents the pixel point coordinate, represents the coordinate deviation value, represents the radial distortion parameter, represents the tangential distortion parameter, represents the abscissa of the corrected pixel point coordinate, represents the ordinate of the corrected pixel point coordinate, represents the abscissa of the pixel point, represents the ordinate of the pixel point;

[0047] Summarize and correct the pixel point coordinates to obtain a set of corrected pixel point coordinates, and obtain a corrected image based on the set of corrected pixel point coordinates;

[0048] Summarize the corrected images to obtain a set of corrected images.

[0049] Optionally, identifying the damaged insulation points of the corrected image to obtain a set of damaged areas includes:

[0050] Identify the target transmission line from the corrected image, where there are multiple target transmission lines in the corrected image;

[0051] Perform the following operations on all target transmission lines:

[0052] Use a preset division range to divide the target transmission line to obtain multiple line segments;

[0053] Extract one line segment from the multiple line segments in sequence, and perform the following operations on the extracted line segment:

[0054] Obtain the standard gray level interval, filter the line segment to obtain a filtered line segment, and perform graying on the filtered line segment to obtain a gray line segment, where the gray line segment includes: multiple gray pixels, and the gray pixel includes: a gray pixel value, where the gray pixel value is the gray value of the gray pixel;

[0055] Perform the following operations on each gray pixel among the multiple gray pixels:

[0056] Judge whether the gray pixel value corresponding to the gray pixel is within the standard gray level interval;

[0057] If the gray pixel value corresponding to the gray pixel is not within the standard gray level interval, then mark the gray pixel as an abnormal pixel;

[0058] Summarize the abnormal pixels to obtain multiple abnormal pixels, confirm the number of abnormal pixels among the multiple abnormal pixels to obtain the number of abnormal pixels;

[0059] Compare the number of abnormal pixels with the preset standard number of pixels;

[0060] If the number of abnormal pixels is greater than the preset standard number of pixels, then obtain the unit damaged area composed of the multiple abnormal pixels;

[0061] If the number of abnormal pixels is not greater than the preset standard number of pixels, then return to the step of extracting one line segment from the multiple line segments in sequence until all the multiple line segments are extracted;

[0062] Summarize the damaged areas of the unit to obtain the unit damaged area group corresponding to the target transmission line, and summarize the unit damaged area group to obtain the damaged area set corresponding to the corrected image.

[0063] Optionally, the obtaining the set of voltage intensities of the damaged area set includes:

[0064] Extract the damaged areas from the damaged area set in sequence, and perform the following operations on each of the extracted damaged areas:

[0065] Denoise the damaged area to obtain a denoised area image;

[0066] Extract edge points from the denoised area image to obtain an edge point set, calculate the leakage voltage component of each edge point in the edge point set, and sum the leakage voltage components of each edge point in the edge point set to obtain the total leakage voltage component;

[0067] Calculate the voltage intensity of the damaged area according to the total leakage voltage component, where the calculation of the voltage intensity is as follows:

[0068] ,

[0069] Where, represents the voltage intensity, represents pi, represents the capacitance parameter of the transmission line, represents the th edge point in the edge point set represents the leakage voltage component,

[0070] Summarize the voltage intensities to obtain the set of voltage intensities.

[0071] Optionally, the denoising the damaged area to obtain a denoised area image includes:

[0072] Perform a convolution operation on each damaged image pixel point in the damaged area image using a pre-constructed Gaussian kernel formula to obtain a set of denoised pixel points, where the Gaussian kernel formula is as follows:

[0073] ,

[0074] Where, represents the denoised pixel point, represents the standard deviation of the Gaussian function, represents the exponential function with base e, represents the abscissa of the denoised pixel point, represents the ordinate of the denoised pixel point, represents the center position of the Gaussian kernel;

[0075] Confirm the denoised area image according to the set of denoised pixel points.

[0076] To achieve the above object, the present invention further provides a distributed AI intelligent remote inspection system based on edge computing, including:

[0077] An inspection area management module, configured to obtain a set of inspection areas, sort the set of inspection areas using a preset inspection order to obtain an ordered set of inspection areas, where an ordered inspection area includes a substation, and the substation includes: multiple transmission lines;

[0078] An inspection path planning module, configured to sequentially extract an ordered inspection area from the ordered set of inspection areas, and perform the following operations on each of the extracted ordered inspection areas: obtain the electromagnetic interference intensity of the ordered inspection area, set the optimal inspection path according to the electromagnetic interference intensity, collect the ambient brightness, confirm the status of the lighting module according to the ambient brightness to obtain the lighting module status, where the lighting module includes: a decrement counter, an all-zero detector, and a cycle counter, receive an intelligent remote inspection instruction, and perform line inspection on the multiple transmission lines in the ordered inspection area according to the intelligent remote inspection instruction, the lighting module status, and the optimal inspection path to obtain a set of line images, where one line image includes one or more transmission lines;

[0079] A line image processing module, configured to perform distortion correction on each line image in the set of line images to obtain a set of corrected images, sequentially extract a corrected image from the set of corrected images, and perform the following operations on each of the extracted corrected images: identify the insulation skin break points of the corrected image to obtain a set of damaged areas, and obtain the voltage intensity set of the set of damaged areas, where the set of damaged areas includes one or more damaged areas;

[0080] A line maintenance management module, configured to, if it is confirmed that there is a voltage intensity in the voltage intensity set that is greater than a preset voltage intensity threshold, send a pre-constructed alarm signal to a pre-constructed maintenance unit. When the maintenance unit receives the alarm signal, use the maintenance unit to repair the damaged area to obtain a repaired area, summarize the repaired areas to obtain a set of repaired area groups, summarize the set of repaired area groups corresponding to the ordered inspection areas to obtain a set of repaired area groups, construct a maintenance database based on the set of repaired area groups, and complete the distributed AI intelligent remote inspection based on edge computing based on the maintenance database.

[0081] To solve the above problems, the present invention further provides an electronic device, and the electronic device includes:

[0082] A memory, storing at least one instruction;

[0083] A processor that executes the instructions stored in the memory to implement the above-mentioned distributed AI intelligent remote inspection method based on edge computing.

[0084] To solve the above problems, the present invention also provides a computer-readable storage medium, in which at least one instruction is stored, and the at least one instruction is executed by a processor in an electronic device to implement the above-mentioned distributed AI intelligent remote inspection method based on edge computing.

[0085] To solve the problems described in the background art, the present invention obtains a set of inspection areas, sorts the inspection area set using a preset inspection order to obtain an ordered inspection area set. Among them, an ordered inspection area includes a substation, and the substation includes: a plurality of transmission lines. By using the preset inspection order, the present invention can optimize the inspection path, reduce the inspection time and improve the inspection efficiency. Sequentially extract an ordered inspection area from the ordered inspection area set, and perform the following operations on each of the extracted ordered inspection areas: The present invention processes the inspection areas one by one to achieve modular management, which is convenient for system maintenance and expansion. Obtain the electromagnetic interference intensity of the ordered inspection area, and set the optimal inspection path according to the electromagnetic interference intensity. By detecting the electromagnetic interference intensity and setting according to it, the present invention can avoid areas with severe interference and select a path with better signals to ensure the stability and accuracy of data transmission. Collect the ambient brightness, confirm the state of the lighting module according to the ambient brightness to obtain the lighting module state. Among them, the lighting module includes: a decrement counter, an all-zero detector and a cycle counter. The present invention automatically adjusts the lighting module according to the ambient brightness to ensure that images can be clearly captured even in low-light conditions and improve the image quality. Receive an intelligent remote inspection instruction, and perform line inspection on the plurality of transmission lines in the ordered inspection area according to the intelligent remote inspection instruction, the lighting module state and the optimal inspection path to obtain a set of line images. Among them, one line image includes one or more transmission lines. The present invention combines the inspection instruction, the lighting state and the optimal path to efficiently collect the image data of the transmission lines, ensuring that each transmission line can be inspected and improving the integrity of the data. Perform distortion correction on each line image in the set of line images to obtain a set of corrected images. Sequentially extract a corrected image from the set of corrected images, and perform the following operations on each of the extracted corrected images: Through distortion correction, the present invention can correct the geometric distortion of the image, improve the clarity and accuracy of the image, and the corrected image is more convenient for subsequent analysis and processing. Identify the damaged points of the insulation skin on the corrected image to obtain a set of damaged areas, and obtain the voltage intensity set of the set of damaged areas. Among them, the set of damaged areas includes one or more damaged areas. Through the identification of the damaged points of the insulation skin, the present invention can highlight abnormal areas such as damaged insulation skin, which is convenient for identification. If it is confirmed that there is a voltage intensity in the voltage intensity set that is greater than the preset voltage intensity threshold, send a pre-constructed alarm signal to a pre-constructed maintenance unit. When the maintenance unit receives the alarm signal, use the maintenance unit to repair the damaged area to obtain a repaired area, and summarize the repaired areas to obtain a set of repaired area groups. When the present invention detects a high-voltage area, it immediately issues an alarm to remind relevant personnel to take emergency measures. Summarize the set of repaired area groups corresponding to the ordered inspection areas to obtain a set of repaired area groups. The present invention optimizes the allocation and utilization of maintenance resources by centrally processing maintenance tasks. Build a maintenance database based on the set of repaired area groups, and complete the distributed AI intelligent remote inspection based on edge computing based on the maintenance database.The maintenance database can record historical maintenance data, provide reference for future inspections and maintenance, and achieve data-driven decision-making. Therefore, the present invention can improve the efficiency and reliability of remote inspections. Description of the Drawings

[0086] Figure 1 It is a schematic flowchart of a distributed AI intelligent remote inspection method based on edge computing provided by an embodiment of the present invention;

[0087] Figure 2 It is a functional module diagram of a distributed AI intelligent remote inspection system based on edge computing provided by an embodiment of the present invention;

[0088] Figure 3 It is a schematic structural diagram of an electronic device for implementing the distributed AI intelligent remote inspection method based on edge computing provided by an embodiment of the present invention.

[0089] Description of the Reference Numerals:

[0090] 1. Electronic device; 10. Processor; 11. Memory; 12. Bus.

[0091] The implementation, functional features and advantages of the object of the present invention will be further described in conjunction with the embodiments with reference to the drawings. Detailed Embodiments

[0092] It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0093] An embodiment of the present application provides a distributed AI intelligent remote inspection method based on edge computing. The execution subject of the distributed AI intelligent remote inspection method based on edge computing includes, but is not limited to, at least one of electronic devices such as a server, a terminal, etc. that can be configured to execute the method provided by the embodiment of the present application. In other words, the distributed AI intelligent remote inspection method based on edge computing can be executed by software or hardware installed on a terminal device or a server device, and the software can be a blockchain platform. The server includes, but is not limited to: a single server, a server cluster, a cloud server, or a cloud server cluster, etc.

[0094] Refer to Figure 1 As shown, it is a schematic flowchart of a distributed AI intelligent remote inspection method based on edge computing provided by an embodiment of the present invention. In this embodiment, the distributed AI intelligent remote inspection method based on edge computing includes:

[0095] S1. Obtain a set of inspection areas, and use a preset inspection order to sort the set of inspection areas to obtain an ordered set of inspection areas.

[0096] Specifically, an ordered inspection area includes a substation, and the substation includes: multiple transmission lines.

[0097] It should be explained that the inspection area set refers to the set of all areas that need to be inspected. The inspection order refers to a preset inspection order. Optionally, the inspection areas in the inspection area set are sorted in descending order according to the probability of failure of the substations corresponding to the inspection areas in history to obtain the inspection order. Area sorting refers to the operation of sorting each inspection area in the inspection area set. The ordered inspection area set refers to the set of inspection areas arranged in the preset inspection order.

[0098] S2. Sequentially extract an ordered inspection area from the ordered inspection area set, and perform the following operations on each of the extracted ordered inspection areas: Obtain the electromagnetic interference intensity of the ordered inspection area, and set the optimal inspection path according to the electromagnetic interference intensity.

[0099] Specifically, the obtaining of the electromagnetic interference intensity of the ordered inspection area includes:

[0100] Obtain the area environment parameters and area current intensity of the ordered inspection area, and obtain the area air density value based on the area environment parameters. Among them, the area environment parameters include: area temperature and area air pressure;

[0101] Judge whether the area air density value is within the preset theoretical density interval;

[0102] If it is confirmed that the area air density value is within the theoretical density interval, then correct the area air density value using the area current intensity to obtain the corrected density air value;

[0103] Calculate the electromagnetic interference intensity of multiple transmission lines in the ordered inspection area using the corrected density air value. The calculation formula of the electromagnetic interference intensity is as follows:

[0104] ,

[0105] ,

[0106] Among them, represents the corrected density air value, represents the preset standard temperature, represents the area temperature, represents the area air pressure, represents the preset standard air pressure, represents the area current intensity, represents the electromagnetic interference intensity, represents the preset insulation skin roughness coefficient, represents the preset transmission line radius.

[0107] It is understandable that the theoretical density range is a preset standard range. If the regional air density value is not within the theoretical density range, the calculation result of the electromagnetic interference intensity will be inaccurate. If the regional air density value is within the theoretical density range, the calculation result of the electromagnetic interference intensity will be more accurate. Using the regional current intensity to correct the regional air density value can further improve the accuracy of the calculation result of the electromagnetic interference intensity.

[0108] It should be explained that obtaining the regional environmental parameters of the ordered inspection area means detecting the regional environmental parameters of the ordered inspection area by using a temperature sensor, a pressure sensor and a current intensity sensor. The regional temperature refers to the environmental temperature within the ordered inspection area. The regional air pressure and the regional current intensity respectively refer to the environmental air pressure within the ordered inspection area and the current intensity in multiple transmission lines. In the step of obtaining the regional air density value based on the regional environmental parameters, the following formula is used to obtain the regional air density value:

[0109] ,

[0110] Among them, represents the regional air density value, represents the gas constant.

[0111] It should be explained that the corrected density air value refers to the regional air density value within the theoretical density range, which is used to more accurately reflect the actual environmental conditions. The standard temperature refers to the temperature under standard atmospheric conditions. The standard air pressure represents the air pressure under standard atmospheric conditions. The roughness coefficient of the insulating skin refers to the roughness coefficient of the surface of the insulating skin of the transmission line, which is determined in advance according to the processing technology. The radius of the transmission line is the radius determined in advance according to the design of the transmission line.

[0112] Specifically, setting the optimal inspection path according to the electromagnetic interference intensity includes:

[0113] Obtaining the electromagnetic intensity attenuation value according to the electromagnetic interference intensity, and determining the inspection vision length and inspection vision width of the pre-built unmanned aerial vehicle based on the electromagnetic intensity attenuation value. Among them, the unmanned aerial vehicle includes: a camera, and the inspection vision length and inspection vision width are expressed as:

[0114] ,

[0115] ,

[0116] ,

[0117] Among them, represents the inspection vision length, represents the inspection vision width, represents the proportional adjustment coefficient, Represents the length of the camera, Represents the width of the camera, Represents the preset flight altitude, Represents the relative position height difference of the camera, Represents the electromagnetic intensity attenuation value, Represents the electromagnetic interference demarcation point, Represents the line measurement distance;

[0118] Calculate the optimal inspection path of the ordered inspection area by using the pre - constructed optimal path formula, inspection vision length and inspection vision width.

[0119] It should be explained that the inspection vision length refers to the length along the power transmission line that the camera can cover during the inspection process of the UAV. The inspection vision width refers to the width perpendicular to the power transmission line that the camera can cover during the inspection process of the UAV. The proportional adjustment coefficient is a coefficient set by humans for correcting the length and width of the camera vision. The flight altitude refers to the height of the UAV relative to the ground preset in advance during the inspection process. The relative position height difference refers to the height of the camera installation position relative to the bottom of the UAV. The electromagnetic interference demarcation point refers to a boundary value, which is the safety boundary point of electromagnetic interference that the UAV can withstand during flight.

[0120] Importantly, the step of obtaining the electromagnetic intensity attenuation value according to the electromagnetic interference intensity is: calculate the electromagnetic intensity attenuation value by using the following formula:

[0121] ,

[0122] Wherein, Represents the natural constant, Represents the preset attenuation coefficient.

[0123] It can be understood that the attenuation coefficient is a coefficient used to describe the attenuation degree of electromagnetic interference intensity with distance. The larger the attenuation coefficient, the faster the attenuation speed of electromagnetic interference intensity with the increase of distance. The smaller the attenuation coefficient, the slower the attenuation speed of electromagnetic interference intensity with the increase of distance.

[0124] Specifically, the optimal path formula is as follows:

[0125] ,

[0126] Wherein, Represents the optimal inspection path, Represents the total inspection length of the UAV, Represents the line spacing of the power transmission line, Represents the number of power transmission lines.

[0127] It should be noted that the optimal inspection path refers to the most efficient UAV flight path calculated through the optimal path formula based on factors such as electromagnetic interference intensity, UAV inspection vision, and transmission line layout, aiming to improve inspection efficiency and safety.

[0128] S3. Collect the ambient brightness, confirm the status of the lighting module according to the ambient brightness, and obtain the status of the lighting module.

[0129] Specifically, the lighting module includes: a decrement counter, an all-zero detector, and a cycle counter.

[0130] In detail, the step of confirming the status of the lighting module according to the ambient brightness and obtaining the status of the lighting module includes:

[0131] Obtain the ambient brightness value according to the ambient brightness;

[0132] If the ambient brightness value is within the preset standard inspection brightness range, obtain the turn-on signal, turn on the lighting module based on the turn-on signal, and after confirming that the turned-on lighting module is the preset turned-on lighting module, confirm the all-zero detector as the initial status indicator;

[0133] If the ambient brightness value is not within the preset standard inspection brightness range, obtain the turn-off signal, turn off the lighting module based on the turn-off signal, and after confirming that the turned-off lighting module is the preset turned-off lighting module, confirm the all-zero detector as the initial status indicator;

[0134] Set the initial value of the decrement counter based on the initial status indicator and the cycle counter to obtain the initial value J of the decrement counter, and perform a decrement operation on the initial value J of the decrement counter using the pre-constructed divided-frequency clock signal to obtain the decremented output value;

[0135] If the decremented output value is not zero, use the decremented output value as the initial value J of the decrement counter, and return to the step of performing the decrement operation on the initial value J of the decrement counter using the pre-constructed divided-frequency clock signal until the decremented output value is zero, then confirm the initial status indicator as the target all-zero detector, and confirm the status of the lighting module based on the target all-zero detector.

[0136] It should be noted that the collected ambient brightness refers to the ambient brightness collected by a photoresistor. The ambient brightness value refers to the specific value obtained by measuring the ambient brightness using a photoresistor. The standard inspection brightness range refers to a preset range of brightness values. When the ambient brightness value is within the standard inspection brightness range, it indicates that the brightness in the environment is appropriate and no additional lighting is required. The all-zero detector is a logic device that indicates that the lighting module should be turned on when the ambient brightness is within the standard inspection brightness range, and indicates that the lighting module should be turned off when the ambient brightness is not within the standard inspection brightness range. The initial state indicator is a device that confirms the initial state of the lighting module based on the signal of the all-zero detector. The decrement operation refers to the operation of performing a subtraction of 1 on the initial value N of the decrement counter using the decrement operator. Turning on the lighting module means the lighting module with the light in the on state. Turning off the lighting module means the lighting module with the light in the off state.

[0137] Importantly, the period counter is a device used to measure the time from the change in the state of the lighting module to the end of a preset time period. The decrement counter is a counter that performs a subtraction of 1 each time. The divided-frequency clock signal is a clock signal used to control the decrement frequency of the decrement counter. The decremented output value refers to the value output by the decrement counter after each decrement operation. The target all-zero detector is a device that, when the decremented output value of the decrement counter is zero, confirms that the initial state indicator is the target all-zero detector and is used to confirm the final state of the lighting module. The turn-on signal is a signal used to indicate that the system should turn on the lighting module. The turn-off signal is a signal used to indicate that the system should turn off the lighting module.

[0138] S4. Receive the intelligent remote inspection instruction, and perform line inspections on multiple transmission lines in the ordered inspection area according to the intelligent remote inspection instruction, the lighting module state, and the optimal inspection path, to obtain a set of line images.

[0139] Specifically, one line image includes one or more transmission lines.

[0140] It should be noted that the intelligent remote inspection instruction is an instruction issued by a human to guide the drone to inspect a specific area. The lighting module state refers to the on / off state of the lighting device during the inspection process. The transmission line refers to the line in the power transmission network for long-distance transmission of electric energy. For example, high-voltage transmission lines and low-voltage distribution lines, etc. The set of line images refers to the set of images of multiple transmission lines collected by a camera during the inspection process. Line inspection refers to the operation of taking pictures of multiple transmission lines along the optimal inspection path at a preset inspection frequency.

[0141] S5. Perform distortion correction on each line image in the set of line images to obtain a set of corrected images.

[0142] Specifically, performing distortion correction on each line image in the line image set to obtain a corrected image set, including:

[0143] Sequentially extracting a line image from the line image set, and performing the following operations on the extracted line image:

[0144] Constructing a plane rectangular coordinate system based on the line image, obtaining the coordinates of each pixel point in the line image according to the plane rectangular coordinate system, and obtaining a pixel point coordinate set;

[0145] Sequentially extracting a pixel point coordinate from the pixel point coordinate set, and performing the following operations on the extracted pixel point coordinate:

[0146] Performing distortion correction on the pixel point coordinate to obtain a corrected pixel point coordinate, where the corrected pixel point coordinate is expressed as:

[0147] ,

[0148] where, represents the corrected pixel point coordinate, represents the pixel point coordinate, represents the coordinate deviation value, represents the radial distortion parameter, represents the tangential distortion parameter, represents the abscissa of the corrected pixel point coordinate, represents the ordinate of the corrected pixel point coordinate, represents the abscissa of the pixel point, represents the ordinate of the pixel point;

[0149] Summarizing the corrected pixel point coordinates to obtain a corrected pixel point coordinate set, and obtaining a corrected image based on the corrected pixel point coordinate set;

[0150] Summarizing the corrected images to obtain a corrected image set.

[0151] It should be explained that the step of constructing a plane rectangular coordinate system according to the line image is: constructing a plane rectangular coordinate system with the lower left corner of the line image as the origin. The pixel point coordinate set refers to the set composed of the coordinates of all pixel points in the line image on the plane rectangular coordinate system. Distortion correction refers to the operation of correcting the pixel point coordinates using the radial distortion parameter and the tangential distortion parameter. The corrected pixel point coordinate refers to the coordinate obtained after the distortion correction operation. Due to the influence of the camera lens, the original image may have distortion, resulting in the straight lines in the line image looking curved or the shape being distorted, so pixel point distortion correction is required.

[0152] It should be noted that the calibrated pixel coordinate set refers to the set composed of all calibrated pixel coordinates. The radial distortion parameter refers to the parameter used to correct the phenomenon that the pixel points at the edge of the line image shift inward or outward. The tangential distortion parameter refers to the parameter used to correct the pixel position shift caused by the tangential distortion of the lens in the line image. The calibrated image refers to the image composed of the calibrated pixel coordinate set. The calibrated image set refers to the set composed of all calibrated images after distortion correction.

[0153] S6. Extract a calibrated image from the calibrated image set in sequence, and perform the following operations on each of the extracted calibrated images: Identify the damaged points of the insulating skin on the calibrated image to obtain a damaged area set, and obtain the voltage intensity set of the damaged area set, where the damaged area set includes one or more damaged areas.

[0154] Specifically, the identifying the damaged points of the insulating skin on the calibrated image to obtain a damaged area set includes:

[0155] Identify the target transmission line from the calibrated image, where the calibrated image includes multiple target transmission lines;

[0156] Perform the following operations on each of the target transmission lines:

[0157] Divide the target transmission line using a preset division range to obtain multiple line segments;

[0158] Extract a line segment from the multiple line segments in sequence, and perform the following operations on each of the extracted line segments:

[0159] Obtain the standard gray level interval, filter the line segment to obtain a filtered line segment, and perform graying on the filtered line segment to obtain a gray line segment, where the gray line segment includes: multiple gray pixels, and the gray pixel includes: a gray pixel value, where the gray pixel value is the gray value of the gray pixel;

[0160] Perform the following operations on each of the multiple gray pixels:

[0161] Judge whether the gray pixel value corresponding to the gray pixel is within the standard gray level interval;

[0162] If the gray pixel value corresponding to the gray pixel is not within the standard gray level interval, record the gray pixel as an abnormal pixel;

[0163] Summarize the abnormal pixels to obtain multiple abnormal pixels, confirm the number of abnormal pixels among the multiple abnormal pixels to obtain the number of abnormal pixels;

[0164] Compare the number of abnormal pixels with the preset standard number of pixels;

[0165] If the number of abnormal pixel points is greater than a preset standard number of pixel points, obtain a unit damaged area composed of multiple abnormal pixel points;

[0166] If the number of abnormal pixel points is not greater than the preset standard number of pixel points, return the step of sequentially extracting one line segment from multiple line segments until all the multiple line segments are extracted;

[0167] Summarize the unit damaged areas to obtain a set of unit damaged areas corresponding to the target transmission line, and summarize the set of unit damaged areas to obtain a damaged area set corresponding to the corrected image.

[0168] It should be explained that the target transmission line refers to an image of a transmission line to be detected extracted from the corrected image. The method for identifying the target transmission line in the embodiments of the present invention is a prior art, such as an edge detection algorithm, which will not be elaborated here.

[0169] Importantly, the division range refers to a preset range for dividing the target transmission line into multiple small segments. For example, the division range is 3 cm. A line segment refers to an image of a section of wire obtained by dividing the target transmission line using the division range. The filtering refers to the operation of filtering the line segment using a Gaussian filter. The filtered line segment refers to the image corresponding to the filtered line segment. The technology of performing grayscale conversion on the filtered line segment in the embodiments of the present invention is a prior art and will not be elaborated here. The grayscale line segment refers to the image corresponding to the filtered line segment after grayscale conversion. The method for obtaining the standard grayscale interval is as follows: The detection personnel take pictures of the target transmission line without any damage under different weather conditions from the historical monitoring time period to obtain a set of captured images, and perform the same image processing as the method for obtaining the corresponding image of the grayscale line segment in the embodiments of the present invention on the set of captured images, so as to obtain a continuous interval composed of the grayscale values of the transmission line without damage. The continuous interval is the standard grayscale interval.

[0170] It should be explained that the number of abnormal pixel points refers to the number of pixel points whose grayscale values are not within the standard grayscale interval. The standard number of pixel points refers to a preset number used to determine whether the abnormal pixel points are noise points in the line segment image. The grayscale pixel points refer to the pixel points in the grayscale line segment. The unit damaged area refers to the image corresponding to the area of the maximum circumscribed rectangle composed of multiple abnormal pixel points. The set of unit damaged areas refers to the set composed of all unit damaged areas. The damaged area set refers to the set composed of all sets of unit damaged areas in the corrected image.

[0171] Exemplarily, there are three power transmission lines in the calibrated image. There is one unit damaged area in the first line segment of the first power transmission line, two unit damaged areas in the second line segment of the second power transmission line, and no unit damaged area in the third power transmission line. Then, the one unit damaged area of the first power transmission line is the unit damaged area group corresponding to the first power transmission line, and the two unit damaged areas of the second power transmission line are the unit damaged area group corresponding to the second power transmission line. Summarize the unit damaged area group of the first power transmission line and the unit damaged area group of the second power transmission line to obtain the damaged area set.

[0172] Specifically, the obtaining of the voltage intensity set of the damaged area set includes:

[0173] Sequentially extract damaged areas from the damaged area set, and perform the following operations on each of the extracted damaged areas:

[0174] Denoise the damaged area to obtain a denoised area image;

[0175] Extract edge points from the denoised area image to obtain an edge point set, calculate the leakage voltage component of each edge point in the edge point set, and sum up the leakage voltage components of each edge point in the edge point set to obtain the total leakage voltage component;

[0176] Calculate the voltage intensity of the damaged area according to the total leakage voltage component. Among them, the calculation of the voltage intensity is as follows:

[0177] ,

[0178] Among them, represents the voltage intensity, represents pi, represents the capacitance parameter of the power transmission line, represents the th edge point in the edge point set, represents the number of edge points in the edge point set;

[0179] Summarize the voltage intensities to obtain the voltage intensity set.

[0180] It should be explained that the denoised area image refers to the image obtained by denoising the damaged area. The extraction of edge points from the denoised area image refers to using an edge detection algorithm to extract the edge points of the power transmission line from the denoised area image. The edge detection algorithm described in the embodiments of the present invention is a prior art and will not be elaborated here.

[0181] It should be noted that the edge point set refers to the set composed of all the extracted edge points. The steps for calculating the leakage voltage component of each edge point in the edge point set are as follows: extract edge points from the denoised area image, find the corresponding position in the actual transmission line according to the edge point, and detect the leakage voltage component at the corresponding position. The total leakage voltage component refers to the sum of the leakage voltage components of all the edge points in the edge point set. The capacitance parameter refers to the parameter of the capacitance characteristic of the transmission line.

[0182] Specifically, the denoising of the damaged area to obtain the denoised area image includes:

[0183] Performing a convolution operation on each damaged image pixel point in the damaged area image using a pre-constructed Gaussian kernel formula to obtain a set of denoised pixel points, where the Gaussian kernel formula is as follows:

[0184] ,

[0185] where, represents the denoised pixel point, represents the standard deviation of the Gaussian function, represents the exponential function with base e, represents the abscissa of the denoised pixel point, represents the ordinate of the denoised pixel point, represents the center position of the Gaussian kernel;

[0186] Confirm the denoised area image according to the set of denoised pixel points.

[0187] It should be noted that the convolution operation refers to the operation of using the Gaussian kernel formula as a sliding window to cover each damaged image pixel point in the image, and for each damaged image pixel point, calculating the sum of the products of the pixel values in the neighborhood of the damaged image pixel point and the Gaussian kernel. The standard deviation of the Gaussian function refers to a preset value. The larger the standard deviation of the Gaussian function, the stronger the denoising effect; the smaller the standard deviation of the Gaussian function, the weaker the denoising effect. The center position of the Gaussian kernel refers to the center point of the Gaussian kernel matrix, usually located at the geometric center of the matrix.

[0188] S7. If it is confirmed that there is a voltage intensity greater than the preset voltage intensity threshold in the voltage intensity concentration, send a pre-constructed alarm signal to the pre-constructed maintenance unit. When the maintenance unit receives the alarm signal, use the maintenance unit to repair the damaged area to obtain a repaired area, and summarize the repaired areas to obtain a group of repaired areas.

[0189] It should be explained that the voltage intensity threshold refers to a preset voltage value. The use of the maintenance unit to repair the damaged area to obtain the repaired area means replacing the transmission line or repairing the insulation skin of the damaged area according to the alarm signal. The repaired area refers to the damaged area after being repaired. The repaired area group refers to the set of all repaired areas. The maintenance unit is a unit that can repair the damaged area.

[0190] S8. Summarize the repaired area groups corresponding to the orderly inspection areas to obtain a set of repaired area groups.

[0191] It should be explained that the set of repaired area groups refers to the set formed by summarizing the repaired area groups corresponding to all orderly inspection areas together.

[0192] S9. Build a maintenance database based on the set of repaired area groups, and complete the distributed AI intelligent remote inspection based on the edge computing based on the maintenance database.

[0193] It should be explained that the steps of building a maintenance database based on the set of repaired areas are as follows: Obtain maintenance data from the repaired areas, where the maintenance data includes: maintenance records, historical fault data, maintenance time, and maintenance measures. Store the repaired area groups corresponding to different orderly inspection areas in different tables to obtain a set of maintenance tables, and use the set of maintenance tables as the maintenance database. The main function of building a maintenance database based on the set of repaired area groups is to provide a centralized platform to store and manage maintenance data, facilitate access and maintenance, and at the same time, based on historical data and analysis results, assist in formulating maintenance strategies and predictive maintenance plans.

[0194] To solve the problems described in the background art, the present invention obtains a set of inspection areas, sorts the set of inspection areas using a preset inspection order to obtain an ordered set of inspection areas. Among them, an ordered inspection area includes a substation, and the substation includes: a plurality of transmission lines. The present invention can optimize the inspection path, reduce the inspection time and improve the inspection efficiency through the preset inspection order. Sequentially extract an ordered inspection area from the ordered set of inspection areas, and perform the following operations on each of the extracted ordered inspection areas: The present invention processes the inspection areas one by one to achieve modular management, which is convenient for system maintenance and expansion. Obtain the electromagnetic interference intensity of the ordered inspection area, and set the optimal inspection path according to the electromagnetic interference intensity. The present invention can avoid areas with severe interference and select a path with better signals by detecting the electromagnetic interference intensity and setting according to it, ensuring the stability and accuracy of data transmission. Collect the ambient brightness, confirm the status of the lighting module according to the ambient brightness to obtain the lighting module status. Among them, the lighting module includes: a decrement counter, an all-zero detector, and a period counter. The present invention automatically adjusts the lighting module according to the ambient brightness to ensure that images can be clearly captured even in low-light conditions and improve the image quality. Receive an intelligent remote inspection instruction, and perform line inspection on the plurality of transmission lines in the ordered inspection area according to the intelligent remote inspection instruction, the lighting module status, and the optimal inspection path to obtain a set of line images. Among them, one line image includes one or more transmission lines. The present invention combines the inspection instruction, the lighting status, and the optimal path to efficiently collect the image data of the transmission lines, ensuring that each transmission line can be inspected and improving the integrity of the data. Perform distortion correction on each line image in the set of line images to obtain a set of corrected images. Sequentially extract a corrected image from the set of corrected images, and perform the following operations on each of the extracted corrected images: The present invention can correct the geometric distortion of the image through distortion correction, improve the clarity and accuracy of the image, and the corrected image is more convenient for subsequent analysis and processing. Identify the insulation skin break points of the corrected image to obtain a set of damaged areas, and obtain the voltage intensity set of the set of damaged areas. Among them, the set of damaged areas includes one or more damaged areas. The present invention can highlight abnormal areas such as insulation skin breaks through the identification of insulation skin break points, which is convenient for identification. If it is confirmed that there is a voltage intensity in the voltage intensity set that is greater than a preset voltage intensity threshold, send a pre-constructed alarm signal to a pre-constructed maintenance unit. When the maintenance unit receives the alarm signal, use the maintenance unit to repair the damaged area to obtain a repaired area, and summarize the repaired areas to obtain a set of repaired area groups. The present invention immediately issues an alarm when detecting a high-voltage area, reminding relevant personnel to take emergency measures. Summarize the set of repaired area groups corresponding to the ordered inspection areas to obtain a set of repaired area groups. The present invention optimizes the allocation and utilization of maintenance resources by centrally processing maintenance tasks. Build a maintenance database based on the set of repaired area groups, and complete distributed AI intelligent remote inspection based on edge computing based on the maintenance database.The maintenance database can record historical maintenance data, provide reference for future inspections and maintenance, and achieve data-driven decision-making. Therefore, the present invention can improve the efficiency and reliability of remote inspections.

[0195] As Figure 2 shown, it is a functional module diagram of a distributed AI intelligent remote inspection system based on edge computing provided by an embodiment of the present invention.

[0196] The distributed AI intelligent remote inspection system 100 based on edge computing according to the present invention can be installed in an electronic device. According to the functions achieved, the distributed AI intelligent remote inspection system 100 based on edge computing can include an inspection area management module 101, an inspection path planning module 102, a line image processing module 103, and a line maintenance management module 104. The modules in the present invention can also be referred to as units, which refer to a series of computer program segments that can be executed by a processor of an electronic device and can complete fixed functions, and are stored in the memory of the electronic device;

[0197] The inspection area management module 101 is used to obtain an inspection area set, sort the inspection area set using a preset inspection order to obtain an ordered inspection area set. Among them, an ordered inspection area includes a substation, and the substation includes: a plurality of transmission lines;

[0198] The inspection path planning module 102 is used to sequentially extract an ordered inspection area from the ordered inspection area set, and perform the following operations on each of the extracted ordered inspection areas: obtain the electromagnetic interference intensity of the ordered inspection area, set the optimal inspection path according to the electromagnetic interference intensity, collect the ambient brightness, confirm the status of the lighting module according to the ambient brightness to obtain the lighting module status. Among them, the lighting module includes: a decrement counter, an all-zero detector, and a cycle counter. Receive an intelligent remote inspection instruction, and perform line inspections on the plurality of transmission lines in the ordered inspection area according to the intelligent remote inspection instruction, the lighting module status, and the optimal inspection path to obtain a line image set. Among them, one line image includes one or more transmission lines;

[0199] The line image processing module 103 is used to perform distortion correction on each line image in the line image set to obtain a corrected image set, sequentially extract a corrected image from the corrected image set, and perform the following operations on each of the extracted corrected images: identify the insulation skin break points on the corrected image to obtain a break area set, and obtain the voltage intensity set of the break area set. Among them, the break area set includes one or more break areas;

[0200] The circuit maintenance management module 104 is configured to send a pre-constructed alarm signal to a pre-constructed maintenance unit if it is confirmed that there is a voltage intensity greater than a preset voltage intensity threshold concentrated in the voltage intensity. When the maintenance unit receives the alarm signal, the damaged area is repaired by the maintenance unit to obtain a repaired area, and the repaired areas are summarized to obtain a set of repaired areas. The set of repaired areas corresponding to the ordered inspection areas is summarized to obtain a set of repaired area groups. A maintenance database is constructed based on the set of repaired area groups, and distributed AI intelligent remote inspection based on edge computing is completed based on the maintenance database.

[0201] Specifically, each module in the distributed AI intelligent remote inspection system 100 based on edge computing in the embodiment of the present invention adopts the same technical means as the Figure 1 distributed AI intelligent remote inspection method based on edge computing described above, and can produce the same technical effects, which will not be elaborated here.

[0202] As Figure 3 shown, it is a schematic structural diagram of an electronic device for implementing the distributed AI intelligent remote inspection method based on edge computing provided by an embodiment of the present invention.

[0203] The electronic device 1 may include a processor 10, a memory 11, and a bus 12, and may also include a computer program stored in the memory 11 and executable on the processor 10, such as a distributed AI intelligent remote inspection method program based on edge computing.

[0204] Among them, the memory 11 includes at least one type of readable storage medium, and the readable storage medium includes flash memory, mobile hard disk, multimedia card, card-type memory (such as SD or DX memory, etc.), magnetic memory, magnetic disk, optical disk, etc. The memory 11 may be an internal storage unit of the electronic device 1 in some embodiments, such as the mobile hard disk of the electronic device 1. The memory 11 may also be an external storage device of the electronic device 1 in other embodiments, such as a plug-in mobile hard disk, a Smart Media Card (SMC), a Secure Digital (SD) card, a Flash Card, etc. equipped on the electronic device 1. Further, the memory 11 also includes the internal storage unit of the electronic device 1 and the external storage device. The memory 11 can not only be used to store application software installed in the electronic device 1 and various types of data, such as the code of the distributed AI intelligent remote inspection method program, but also be used to temporarily store data that has been output or will be output.

[0205] In some embodiments, the processor 10 may be composed of an integrated circuit. For example, it may be composed of a single packaged integrated circuit, or may be composed of multiple packaged integrated circuits with the same or different functions, including one or more central processing units (CPUs), microprocessors, digital processing chips, graphics processors, and combinations of various control chips, etc. The processor 10 is the control core (Control Unit) of the electronic device, connecting various components of the entire electronic device through various interfaces and lines, and by running or executing programs or modules stored in the memory 11 (such as the program of the distributed AI intelligent remote inspection method based on edge computing, etc.), and calling the data stored in the memory 11, to execute various functions of the electronic device 1 and process data.

[0206] The bus 12 may be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus, etc. This bus 12 can be divided into an address bus, a data bus, a control bus, etc. The bus 12 is set to enable connection communication between the memory 11 and at least one processor 10, etc.

[0207] Figure 3 Only the electronic device with components is shown. Those skilled in the art can understand that, Figure 3 the shown structure does not constitute a limitation on the electronic device 1, and it may include fewer or more components than shown, or combine certain components, or have different component arrangements.

[0208] For example, although not shown, the electronic device 1 may further include a power source (such as a battery) for supplying power to each component. Preferably, the power source can be logically connected to the at least one processor 10 through a power management device, so as to implement functions such as charge management, discharge management, and power consumption management through the power management device. The power source may also include any components such as one or more DC or AC power sources, a recharge device, a power failure detection circuit, a power converter or inverter, and a power status indicator. The electronic device 1 may also include various sensors, a Bluetooth module, a Wi-Fi module, etc., which will not be elaborated here.

[0209] Furthermore, the electronic device 1 may further include a network interface. Optionally, the network interface may include a wired interface and / or a wireless interface (such as a WI-FI interface, a Bluetooth interface, etc.), which is usually used to establish a communication connection between the electronic device 1 and other electronic devices.

[0210] Optionally, the electronic device 1 may further include a user interface, which may be a display, an input unit (such as a keyboard), and optionally, the user interface may also be a standard wired interface or a wireless interface. Optionally, in some embodiments, the display may be an LED display, a liquid crystal display, a touch liquid crystal display, and an OLED (Organic Light-Emitting Diode) toucher, etc. Among them, the display may also be appropriately referred to as a display screen or a display unit, which is used to display the information processed in the electronic device 1 and to display a visual user interface.

[0211] The distributed AI intelligent remote inspection method program based on edge computing stored in the memory 11 in the electronic device 1 is a combination of multiple instructions, and when running in the processor 10, it can achieve:

[0212] Obtain an inspection area set, sort the inspection area set by area using a preset inspection order to obtain an ordered inspection area set, where an ordered inspection area includes a substation, and the substation includes: multiple transmission lines;

[0213] Successively extract an ordered inspection area from the ordered inspection area set, and perform the following operations on each of the extracted ordered inspection areas:

[0214] Obtain the electromagnetic interference intensity of the ordered inspection area, and set an optimal inspection path according to the electromagnetic interference intensity;

[0215] Collect the ambient brightness, confirm the status of the lighting module according to the ambient brightness to obtain the lighting module status, where the lighting module includes: a decrement counter, an all-zero detector, and a period counter;

[0216] Receive an intelligent remote inspection instruction, and perform line inspection on the multiple transmission lines in the ordered inspection area according to the intelligent remote inspection instruction, the lighting module status, and the optimal inspection path to obtain a line image set, where one line image includes one or more transmission lines;

[0217] Perform distortion correction on each line image in the line image set to obtain a corrected image set, successively extract a corrected image from the corrected image set, and perform the following operations on each of the extracted corrected images:

[0218] Identify the insulation skin break points of the corrected image to obtain a break area set, and obtain the voltage intensity set of the break area set, where the break area set includes one or more break areas;

[0219] If a voltage intensity greater than a preset voltage intensity threshold is confirmed to exist in a concentrated manner, a pre-constructed alarm signal is sent to a pre-constructed maintenance unit. When the maintenance unit receives the alarm signal, the damaged area is repaired using the maintenance unit to obtain a repaired area, and the repaired areas are aggregated to obtain a group of repaired areas;

[0220] Aggregate the groups of repaired areas corresponding to the orderly inspection areas to obtain a set of groups of repaired areas;

[0221] Construct a maintenance database based on the set of groups of repaired areas, and complete distributed AI intelligent remote inspection based on edge computing based on the maintenance database.

[0222] Specifically, for the specific implementation method of the above instructions by the processor 10, reference can be made to Figures 1 to 3 the description of the relevant steps in the corresponding embodiment, which will not be elaborated here.

[0223] Further, if the modules / units integrated in the electronic device 1 are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. The computer-readable storage medium can be volatile or non-volatile. For example, the computer-readable medium can include: any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a mobile hard disk, a magnetic disk, an optical disc, a computer memory, a read-only memory (ROM, Read-Only Memory).

[0224] The present invention also provides a computer-readable storage medium, and the readable storage medium stores a computer program, and when the computer program is executed by a processor of an electronic device, it can implement:

[0225] Obtain a set of inspection areas, perform area sorting on the set of inspection areas using a preset inspection order to obtain an ordered set of inspection areas, where an ordered inspection area includes a substation, and the substation includes: a plurality of transmission lines;

[0226] Successively extract an ordered inspection area from the ordered set of inspection areas, and perform the following operations on each of the extracted ordered inspection areas:

[0227] Obtain the electromagnetic interference intensity of the ordered inspection area, and set an optimal inspection path according to the electromagnetic interference intensity;

[0228] Collect the ambient brightness, confirm the status of the lighting module according to the ambient brightness to obtain the lighting module status, where the lighting module includes: a decrement counter, an all-zero detector, and a cycle counter;

[0229] Receive the intelligent remote inspection instruction, and perform line inspections on multiple transmission lines in the orderly inspection area according to the intelligent remote inspection instruction, the status of the lighting module, and the optimal inspection path to obtain a line image set, where one line image includes one or more transmission lines;

[0230] Perform distortion correction on each line image in the line image set to obtain a corrected image set. Extract one corrected image from the corrected image set in sequence, and perform the following operations on the extracted corrected images:

[0231] Identify the insulation skin break points of the corrected image to obtain a break area set, and obtain a voltage intensity set of the break area set, where the break area set includes one or more break areas;

[0232] If it is confirmed that there is a voltage intensity in the voltage intensity set that is greater than the preset voltage intensity threshold, send a pre-constructed alarm signal to the pre-constructed maintenance unit. When the maintenance unit receives the alarm signal, use the maintenance unit to repair the break area to obtain a repaired area, and summarize the repaired areas to obtain a repaired area group;

[0233] Summarize the repaired area groups corresponding to the orderly inspection area to obtain a repaired area group set;

[0234] Construct a maintenance database based on the repaired area group set, and complete the distributed AI intelligent remote inspection based on edge computing based on the maintenance database.

[0235] In several embodiments provided by the present invention, it should be understood that the disclosed devices, systems, and methods can be implemented in other ways. For example, the system embodiments described above are only illustrative, and there can be other division methods in actual implementation.

[0236] The modules described as separate components may or may not be physically separated. The components shown as modules may or may not be physical units, that is, they can be located in one place, or they can be distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0237] In addition, in each embodiment of the present invention, the functional modules can be integrated in one processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit. The above integrated units can be implemented in the form of hardware, or in the form of hardware plus software functional modules.

[0238] For those skilled in the art, it is obvious that the present invention is not limited to the details of the above exemplary embodiments, and can be implemented in other specific forms without departing from the spirit or basic characteristics of the present invention.

[0239] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A distributed AI intelligent remote inspection method based on edge computing, characterized in that: The method comprises: Obtain an inspection area set, and use a preset inspection order to sort the inspection area set to obtain an ordered inspection area set, wherein an ordered inspection area includes a substation, and the substation includes: a plurality of transmission lines; An ordered inspection area is extracted from the ordered inspection area set in sequence, and the following operations are performed on the extracted ordered inspection areas: Obtain the electromagnetic interference intensity of the orderly inspection area and set the optimal inspection path according to the electromagnetic interference intensity; The step of setting the optimal inspection path according to the electromagnetic interference intensity includes: The electromagnetic intensity attenuation value is obtained according to the electromagnetic interference intensity, and the inspection field length and inspection field width of the pre-built UAV are determined based on the electromagnetic intensity attenuation value, wherein the UAV includes: a camera, and the inspection field length and inspection field width are expressed as: ， ， ， in, Indicates the length of the inspection field of view. Indicates the inspection field width. represents the proportional adjustment coefficient, Indicates the length of the camera, Indicates the width of the camera, Indicates the preset flight altitude. Indicates the relative height difference of the camera. Represents the electromagnetic intensity attenuation value, Indicates the electromagnetic interference demarcation point, Indicates the line measurement distance; Calculate the optimal inspection path for the ordered inspection area using the pre-built optimal path formula, inspection field length, and inspection field width; Collecting the ambient brightness, confirming the state of the lighting module according to the ambient brightness, and obtaining the state of the lighting module, wherein the lighting module includes: a minus-one counter, a zero detector, and a cycle counter; receiving an intelligent remote inspection instruction, and performing line inspection on a plurality of power transmission lines in the orderly inspection area according to the intelligent remote inspection instruction, the state of the lighting module and the optimal inspection path, to obtain a line image set, wherein one line image includes one or more power transmission lines; Each line image in the line image set is subjected to distortion correction to obtain a correction image set, and one correction image is extracted from the correction image set in turn, and the following operations are performed on each of the extracted correction images: Identify insulation damage points on the corrected image to obtain a damage area set, and obtain a voltage intensity set of the damage area set, wherein the damage area set includes one or more damage areas; If it is confirmed that there is a voltage intensity greater than a preset voltage intensity threshold in the voltage intensity concentration, a pre-constructed alarm signal is sent to a pre-constructed maintenance unit. After the maintenance unit receives the alarm signal, the maintenance unit is used to repair the damaged area to obtain a maintenance area, and the maintenance areas are summarized to obtain a maintenance area group. Summarizing the maintenance area groups corresponding to the orderly inspection areas to obtain a maintenance area group set; Build a maintenance database based on maintenance area groups, and complete distributed AI intelligent remote inspection based on edge computing based on the maintenance database.

2. The distributed AI intelligent remote inspection method based on edge computing according to claim 1 is characterized in that: The obtaining of electromagnetic interference intensity in the orderly inspection area includes: Obtain regional environmental parameters and regional current intensity of the orderly inspection area, and obtain regional air density values ​​based on the regional environmental parameters, wherein the regional environmental parameters include: regional temperature and regional air pressure; Determine whether the regional air density value is within the preset theoretical density range; If it is confirmed that the regional air density value is within the theoretical density interval, the regional air density value is corrected using the regional current intensity to obtain a corrected density air value; The modified density air value is used to calculate the electromagnetic interference intensity of multiple power transmission lines in the orderly inspection area, wherein the calculation formula of the electromagnetic interference intensity is as follows: ， ， in, Indicates the corrected density air value, Indicates the preset standard temperature. represents the regional temperature, represents the regional air pressure, Indicates the preset standard air pressure. represents the regional current intensity, Indicates the electromagnetic interference intensity. Indicates the preset insulation roughness coefficient. Indicates the preset transmission line radius.

3. The distributed AI intelligent remote inspection method based on edge computing as claimed in claim 2, characterized in that: The optimal path formula is as follows: ， in, represents the optimal inspection path, Indicates the total inspection length of the drone. Indicates the line spacing of the transmission line, Represents the number of transmission lines.

4. The distributed AI intelligent remote inspection method based on edge computing as claimed in claim 3 is characterized in that: The step of determining the state of the lighting module according to the ambient brightness to obtain the state of the lighting module includes: Get the ambient brightness value according to the ambient brightness; If the ambient brightness value is within the preset standard inspection brightness range, a start signal is obtained, the lighting module is turned on based on the start signal, and after confirming that the turned-on lighting module is the preset turn-on lighting module, the all-zero detector is confirmed as the initial state indicator; If the ambient brightness value is not within the preset standard inspection brightness range, a shutdown signal is obtained, the lighting module is turned off based on the shutdown signal, and after confirming that the turned-off lighting module is the preset turned-off lighting module, the all-zero detector is confirmed as the initial state indicator; The initial value of the subtraction counter is set based on the initial state indicator and the cycle counter to obtain the initial value J of the subtraction counter, and a decrement operation is performed on the initial value J of the subtraction counter using a pre-constructed frequency-divided clock signal to obtain a decrement output value; If the decrement output value is not zero, the decrement output value is used as the initial value J of the decrement counter, and the step of performing the decrement operation on the initial value J of the decrement counter using the pre-built divided clock signal is returned until the decrement output value is zero, then it is confirmed that the initial state indicator is the target all-zero detector, and the lighting module state is confirmed based on the target all-zero detector.

5. The distributed AI intelligent remote inspection method based on edge computing as claimed in claim 4 is characterized in that: The step of performing distortion correction on each line image in the line image set to obtain a corrected image set includes: Extract one line image from the line image set in turn, and perform the following operations on each of the extracted line images: Construct a plane rectangular coordinate system according to the line image, obtain the coordinates of each pixel point in the line image according to the plane rectangular coordinate system, and obtain a pixel point coordinate set; Extract one pixel point coordinate from the pixel point coordinate set one by one, and perform the following operations on the extracted pixel point coordinates: The pixel coordinates are distorted and corrected to obtain the corrected pixel coordinates, where the corrected pixel coordinates are expressed as: ， in, Indicates the coordinates of the corrected pixel points. Represents the pixel coordinates, Indicates the coordinate deviation value, represents the radial distortion parameter, represents the tangential distortion parameter, Indicates the horizontal coordinate of the corrected pixel coordinates, Indicates the ordinate of the corrected pixel coordinates, Represents the horizontal coordinate of the pixel point, Indicates the vertical coordinate of the pixel; Summarizing the coordinates of the corrected pixel points to obtain a corrected pixel point coordinate set, and acquiring a corrected image based on the corrected pixel point coordinate set; The corrected images are aggregated to obtain a corrected image set.

6. The distributed AI intelligent remote inspection method based on edge computing as claimed in claim 5, characterized in that: The step of identifying insulation damage points on the corrected image to obtain a damaged area set includes: Identifying a target power transmission line from the corrected image, wherein the corrected image includes a plurality of target power transmission lines; Perform the following operations on the target transmission lines: Dividing the target transmission line by using a preset division range to obtain multiple line segments; Extract one route segment from multiple route segments one by one, and perform the following operations on each of the extracted route segments: Obtain a standard grayscale interval, filter the line segment to obtain a filtered line segment, perform grayscale conversion on the filtered line segment to obtain a grayscale line segment, wherein the grayscale line segment includes: a plurality of grayscale pixels, and the grayscale pixel includes: a grayscale pixel value, wherein the grayscale pixel value is the grayscale value of the grayscale pixel; The following operations are performed on each of the multiple grayscale pixels: Determine whether the grayscale pixel value corresponding to the grayscale pixel is within the standard grayscale interval; If the grayscale pixel value corresponding to the grayscale pixel is not in the standard grayscale interval, the grayscale pixel is recorded as an abnormal pixel; Summarize the abnormal pixels to obtain multiple abnormal pixels, confirm the number of abnormal pixels among the multiple abnormal pixels, and obtain the number of abnormal pixels; Compare the number of abnormal pixels with the preset standard number of pixels; If the number of abnormal pixels is greater than the preset number of standard pixels, a unit damaged area composed of multiple abnormal pixels is obtained; If the number of abnormal pixels is not greater than the preset number of standard pixels, returning to the step of extracting one line segment from the multiple line segments in sequence until all the multiple line segments are extracted; The unit damaged areas are summarized to obtain a unit damaged area group corresponding to the target transmission line, and the unit damaged area groups are summarized to obtain a damaged area set corresponding to the corrected image.

7. The distributed AI intelligent remote inspection method based on edge computing according to claim 6 is characterized in that: The step of obtaining the voltage intensity set of the damaged area set includes: The damaged areas are extracted from the damaged area set in sequence, and the following operations are performed on the extracted damaged areas: De-noise the damaged area to obtain a denoised area image; Extract edge points from the denoised area image to obtain an edge point set, calculate the leakage voltage component of each edge point in the edge point set, and sum the leakage voltage components of each edge point in the edge point set to obtain a total leakage voltage component; The voltage intensity of the damaged area is calculated based on the total leakage voltage component, where the voltage intensity is calculated as follows: ， in, Indicates the voltage intensity, represents pi, represents the capacitance parameter of the transmission line, Indicates the edge point concentration The leakage voltage component of each edge point is Indicates the number of edge points in the edge point set; The voltage intensities are summed up to obtain a voltage intensity set.

8. The distributed AI intelligent remote inspection method based on edge computing as claimed in claim 7, characterized in that: Denoising the damaged area to obtain a denoised area image includes: The pre-built Gaussian kernel formula is used to perform convolution operation on each damaged image pixel in the damaged area image to obtain a denoised pixel point set, where the Gaussian kernel formula is as follows: ， in, represents the denoised pixel, represents the standard deviation of the Gaussian function, represents the exponential function with base e as base, Represents the horizontal coordinate of the denoised pixel, Represents the ordinate of the denoised pixel point, Indicates the center position of the Gaussian kernel; A denoised area image is confirmed according to the denoised pixel point set.

9. A system using the distributed AI intelligent remote inspection method based on edge computing as claimed in claim 1, characterized in that: The system comprises: The inspection area management module is used to obtain an inspection area set, and use a preset inspection order to sort the inspection area set to obtain an ordered inspection area set, wherein an ordered inspection area includes a substation, and the substation includes: multiple transmission lines; The inspection path planning module is used to extract an orderly inspection area from the orderly inspection area set in turn, and perform the following operations on the extracted orderly inspection areas: obtain the electromagnetic interference intensity of the orderly inspection area, set the optimal inspection path according to the electromagnetic interference intensity, collect the ambient brightness, confirm the state of the lighting module according to the ambient brightness, and obtain the state of the lighting module, wherein the lighting module includes: a minus-one counter, a zero detector and a cycle counter, receive intelligent remote inspection instructions, and perform line inspection on multiple power transmission lines in the orderly inspection area according to the intelligent remote inspection instructions, the state of the lighting module and the optimal inspection path to obtain a line image set, wherein one line image includes one or more power transmission lines; The line image processing module is used to perform distortion correction on each line image in the line image set to obtain a correction image set, extract a correction image from the correction image set in turn, and perform the following operations on the extracted correction images: identify insulation damage points on the correction image to obtain a damage area set, and obtain a voltage intensity set of the damage area set, wherein the damage area set includes one or more damage areas; The line maintenance management module is used to send a pre-built alarm signal to a pre-built maintenance unit if it is confirmed that there is a voltage intensity greater than a preset voltage intensity threshold in the voltage intensity concentration. When the maintenance unit receives the alarm signal, it uses the maintenance unit to repair the damaged area to obtain the maintenance area, and summarizes the maintenance area to obtain a maintenance area group, summarizes the maintenance area groups corresponding to the ordered inspection areas to obtain a maintenance area group set, builds a maintenance database based on the maintenance area group set, and completes distributed AI intelligent remote inspection based on edge computing based on the maintenance database.

Images