Power equipment remote monitoring system and method based on network security protocol

Abstract

The invention provides a power equipment remote monitoring system and method based on a network security protocol, and the system comprises an inspection unmanned aerial vehicle which carries a pan-tilt camera, a control system, an image processor and a signal transmission system, and when the inspection unmanned aerial vehicle hovers at a hovering point, the pan-tilt camera is controlled to perform image collection after automatic focusing; dividing the hovering point locations into common hovering point locations and insulator chain hovering point locations according to different detection targets; at the insulator chain hovering point position, the pan-tilt camera performs image acquisition according to a preset angle, the image processor detects whether the insulator chain exists according to the image acquired by the pan-tilt camera, further detects the number and the suspension type of the insulator chain, and detects whether the self-explosion insulator exists based on the number and the suspension type of the insulator chain. The characteristics of the insulator string are utilized, the advantages of the inspection unmanned aerial vehicle are fully played, and rapid and accurate real-time detection can be carried out.

Description

Technical Field

[0001] The present invention belongs to the field of power equipment safety monitoring, and in particular relates to a power equipment remote monitoring system and method based on a network security protocol. Background Art

[0002] The power grid is the foundation and carrier of the power market. The power grid consists of substations and transmission lines of different voltage levels. Its function is to transmit, control and distribute electric energy. Cross-regional and long-distance power transmission and trading can better regulate the balance of power and promote the optimal allocation of resources across the country. The high-voltage transmission network, which is composed of transmission lines erected by high-voltage power towers, is the energy artery of the power grid. The operating status of the high-voltage transmission network directly determines the safety of the power system. Safety monitoring of the high-voltage transmission network plays an important role in ensuring the stable operation of the power system. The high-voltage power towers and transmission lines of the high-voltage transmission network have been operating in outdoor environments for a long time. Various defects have appeared under the influence of huge tension, temperature changes, lightning, flying object impact, aging, rust, etc. If they are not discovered and repaired in time, they will eventually lead to failures in the high-voltage transmission network, causing large-scale power outages and huge economic losses.

[0003] Traditionally, inspections of high-voltage transmission networks are mainly conducted by inspection workers climbing up towers for regular inspections and maintenance. Some also use telescopes for remote observation and inspections, and some use helicopter inspections to collect images for inspection. However, high-voltage power towers along the high-voltage transmission network are mostly located in mountainous areas, hilly areas and other areas with complex geographical conditions. Manual inspections often require hardworking and experienced inspection workers to climb mountains and towers to check for safety hazards. Inspection work is difficult, inefficient and involves unsafe factors. Helicopter inspections are costly and have low flexibility in image acquisition.

[0004] At present, traditional manual inspections have been gradually replaced by drone inspections, which play an important role in the inspection of power equipment, especially the inspection of high-voltage transmission networks. Drones used for high-voltage transmission network inspections are usually equipped with pan-tilt cameras, which can include cameras in the visible light band and non-visible light band, and some are also equipped with ultrasonic equipment, etc. Through inspections, it is possible to detect the tilt of the tower base of high-voltage power towers, bending and deformation of tower materials, hanging foreign objects, overheating, loose or missing nuts / bolts, pin shedding, insulator defects, etc.

[0005] The use of drones to inspect high-voltage transmission networks usually requires planning inspection routes based on the high-voltage transmission network sections to be inspected. The drones fly along the inspection routes and collect corresponding images. Power workers understand the status of the high-voltage transmission network based on the collected images, judge safety hazards, or obtain safety hazard information through digital image processing. However, in the current inspection of high-voltage transmission network drones, the route is usually planned before the inspection, the inspection is carried out according to the predetermined route, and the captured images and detected data are uploaded for defect detection. There is no subdivision of defects according to the defective parts (such as hardware, tower materials, insulators), nor is there targeted defect detection based on the different characteristics of different defective parts. During the inspection process, there is no autonomous adjustment of the shooting angle, etc. After the inspection, it may be found that the images collected at some points cannot be used for defect identification, and the advantages of the inspection drones are not fully utilized. Summary of the invention

[0006] In order to solve the above problems in the prior art, the present invention provides a remote monitoring system and method for electric power equipment based on a network security protocol.

[0007] The present invention first provides a method for remote monitoring of electric power equipment based on a network security protocol, comprising planning an inspection route and a hovering point of an inspection drone according to a high-voltage transmission network segment to be inspected, wherein when the inspection drone hovers at the hovering point, a pan / tilt camera carried on the inspection drone automatically focuses to perform image acquisition, and uploads the image to a remote work platform and / or an image processor carried on the inspection drone; Specifically, the hovering points are divided into ordinary hovering points and insulator string hovering points according to different detection targets; at the insulator string hovering points, the gimbal camera collects images at a preset angle, and the image processor detects whether there is an insulator string based on the image collected by the gimbal camera. If there is an insulator string, the number of insulator strings is further detected. If the number of insulator strings is greater than 1, the suspension type of multiple insulator strings is further detected, and the number and suspension type of the insulator strings are fed back to the control system; the suspension types include multiple insulator strings being parallel and adjacent, and multiple insulator strings forming a certain angle; the image processor further detects whether there are self-destructing insulators based on the number and suspension type of insulator strings.

[0008] Furthermore, when the number of insulator strings in the image collected at the hovering point of the insulator string is 1, the image processor extracts an image of one insulator string according to a preset size, rotates the extracted insulator string image 180 degrees to obtain a deformed image, and calculates the similarity between the extracted insulator string image and its deformed image through image comparison. If the similarity is lower than a first threshold, it is determined that a self-explosive insulator appears in the insulator string.

[0009] Furthermore, when the number of insulator strings in the image collected at the hovering point of the insulator string is greater than 1 and the multiple insulator strings form a certain angle, the image processor extracts the image of each insulator string according to a preset size, and rotates each extracted image of the insulator string by 180 degrees to obtain respective deformed images, and calculates their respective similarities by comparing the image of each insulator string with their respective deformed images, and compares each calculated similarity with a second threshold value, and determines that self-explosive insulators appear in the insulator strings corresponding to the similarities below the second threshold value.

[0010] Further, when the number of insulator strings in the image collected at the hovering point of the insulator string is greater than 1 and the multiple insulator strings are parallel and adjacent, the image processor extracts the image of each insulator string according to a preset size, and marks them as image 1, image 2, ..., image n in sequence, and constructs a first image set {image 1, image 2, ..., image n}, where n is an integer greater than or equal to 2; Compare any two images in the first image set, calculate the similarity, obtain a first similarity sequence, further calculate the variance of the first similarity sequence, compare the variance of the first similarity sequence with a first variance threshold, and when the variance of the first similarity sequence is greater than the first variance threshold, determine that a self-destructing insulator appears; Alternatively, m images in the first image set are selected and rotated 180 degrees to obtain a second image set {image a1, image a2, ..., image am}, where m is greater than or equal to 1 and less than or equal to n, the first image set and the second image set are combined into a third image set {image 1, image 2, ..., image n, image a1, image a2, ..., image am}, any two images in the third image set are compared, the similarity is calculated, a second similarity series is obtained, the variance of the second similarity series is further calculated, the variance of the second similarity series is compared with a second variance threshold, and when the variance of the second similarity series is greater than the second variance threshold, it is determined that a self-destructing insulator has occurred.

[0011] Furthermore, when the variance of the first similarity series is less than or equal to the first variance threshold, any one image in the first image set is further selected, and the image is rotated 180 degrees to obtain a deformed image, and the selected image is compared with the deformed image to calculate the similarity. If the similarity is greater than a third threshold, it is determined that there is no self-destructing insulator.

[0012] Furthermore, if the calculated similarity is less than a fifth threshold, the image processor feeds back a signal to the control system, and the control system controls the gimbal camera to adjust its position to perform secondary image acquisition according to the feedback signal.

[0013] Furthermore, at the hovering point of the insulator string, after the image processor completes the relevant detection, it sends a detection completion signal to the control system. After the control system receives the detection completion signal, it controls the inspection drone to go to the next point.

[0014] Furthermore, the insulators in the insulator string are glass insulators.

[0015] The present invention also provides a remote monitoring system for electric power equipment based on a network security protocol, including an inspection drone equipped with a pan-tilt camera, a control system, an image processor, and a signal transmission system. The pan-tilt camera is used to collect high-definition pictures or videos during inspections. The control system controls the flight status of the inspection drone and the position of the pan-tilt camera according to instructions. The image processor is used to process pictures collected by the pan-tilt camera. The signal transmission system transmits signals and data. The system is used to implement the aforementioned remote monitoring method for electric power equipment.

[0016] Furthermore, the transmission of signals and data is based on the 5G network security protocol.

[0017] Compared with the prior art, the present invention improves the image processing mode of drone inspection for remote monitoring of medium and high voltage transmission networks of power equipment, specifically including: The present invention classifies the hovering points of the drone inspection, specifically into ordinary points and insulator string points according to different detection targets, and performs image acquisition at ordinary points according to preset photo positions; at insulator string points, after determining the number and arrangement type of insulator strings using its own image processor, insulator self-explosion detection is performed, with fast detection speed and high detection efficiency. It can also determine whether it is necessary to perform secondary image acquisition after angle adjustment at the hovering point based on the detection results, to prevent the discovery of detection black holes caused by the angle and quality of the collected images after the inspection is completed.

[0018] According to the arrangement characteristics of the insulator strings in the high-voltage transmission network, according to the number and arrangement type of the insulator strings, the present invention can quickly determine whether there is a case of self-explosion of the insulators through a simple algorithm, and the detection speed is fast. During detection, the present invention uses the image of the insulator string in the current image and the 180-degree flipped image, and does not require the system to pre-store images, saving storage space. Without comparing with the previous image, the insulator string can be pre-inspected during the initial inspection, and self-explosive insulators can be found in time. The present invention utilizes the characteristics of the insulator string, gives full play to the advantages of the inspection drone, and can perform fast and accurate real-time detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 It is a flow chart of classifying the hovering points according to the present invention; Figure 2 is a flow chart of the present invention for detecting based on the number and suspension type of insulator strings; Figure 3 It is a flow chart of another embodiment of the present invention when the number of insulator strings is greater than 1 and multiple insulator strings are parallel and adjacent. DETAILED DESCRIPTION

[0020] In order to facilitate the understanding of the present invention, the present invention will be described more fully below with reference to the relevant drawings. Preferred embodiments of the present invention are given in the drawings. However, the present invention can be implemented in many different forms and is not limited to the embodiments described herein. On the contrary, the purpose of providing these embodiments is to make the disclosure of the present invention more thoroughly understood. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those generally understood by those skilled in the art of the present invention. The terms used herein are only for the purpose of describing specific embodiments and are not intended to limit the present invention.

[0021] In high-voltage transmission networks, high-voltage power towers are used to erect transmission lines. Different types of towers are selected according to different voltage levels, the layout of transmission lines and lightning arresters, etc. The towers mainly bear wind loads, line tension and other loads. Due to the tension effect, the transmission line generates huge tension in the horizontal direction, and the terminal structure needs to be designed to bear the tension. The insulators are installed on the terminal structure, so that the terminal structure and the live transmission line are connected together but in an electrically isolated state. Usually, one insulator can withstand a voltage of 10,000-15,000 volts. In high-voltage transmission networks, multiple insulators are usually connected to form an insulator string. Under the simultaneous effects of dirt accumulation, moisture and electric field, the surface leakage current of the insulator in the insulator string is too large. When air breakdown occurs, the arc generated will corrode the insulator shed. When the erosion depth is deep, it will cause the insulator to explode, resulting in the loss of insulators in the insulator string, and the insulators adjacent to the self-exploding insulators will also have defects such as cracks and incompleteness. In the inspection of high-voltage transmission networks, it is of great significance to promptly detect defects in insulator strings.

[0022] In the prior art, drones collect images when inspecting high-voltage transmission networks, and power workers judge insulator string defects based on the images, or detect insulator string defects through image algorithms. However, manual judgment is time-consuming and labor-intensive, and relies on the experience of relevant personnel. The algorithm used by image algorithms to detect insulator defects is relatively complex. In other words, the detection method for insulator string defects in the prior art neither fully utilizes the advantages of drone inspections nor fully utilizes the special arrangement and special structural form of insulator strings in high-voltage transmission networks.

[0023] The remote monitoring system and method for electric power equipment provided by the present invention can quickly and easily detect defects in insulator strings, especially detect insulator self-explosion. The remote monitoring system for electric power equipment provided by the present invention includes an inspection drone equipped with a pan-tilt camera, a control system, an image processor, and a signal transmission system. The pan-tilt camera is used to collect high-definition pictures or videos during inspections. The control system controls the flight status of the inspection drone and the orientation of the pan-tilt camera according to instructions. The image processor is used to process the pictures collected by the pan-tilt camera, and the signal transmission system transmits signals and data.

[0024] Furthermore, the transmission of signals and data is based on the 5G network security protocol, which can ensure information security while ensuring high-speed transmission.

[0025] The remote monitoring method of electric power equipment provided by the present invention comprises: The inspection route and hovering point of the inspection drone are planned according to the high-voltage transmission network section to be inspected. When the inspection drone hovers at the hovering point, the gimbal camera automatically focuses to collect images, and uploads the images to the remote work platform and / or the image processor on the inspection drone.

[0026] like Figure 1 As shown, the hovering points are divided into common hovering points and insulator string hovering points according to different detection targets. Usually, a high-voltage power tower in a high-voltage transmission network section to be detected includes multiple common hovering points and multiple insulator string hovering points.

[0027] At common hovering points, the gimbal camera collects images at preset angles, such as the points for photographing the tower base and the tower head. After the collected images are uploaded to the remote work platform, power workers can browse the images to discover safety hazards, or use the image algorithms on the remote work platform to investigate safety hazards.

[0028] A high-voltage power iron tower usually includes multiple strings of insulators, which are arranged at different positions. The hanging directions and the number of insulators of the insulator strings at different positions are the same or different. These insulator strings can be classified according to the hanging direction as follows: a single insulator string is hung in a certain direction, multiple insulator strings are hung in a parallel and adjacent arrangement, and multiple insulator strings are hung at a certain angle. In addition, the number of insulators on an insulator string ranges from a few to dozens. And on a high-voltage power iron tower, when multiple insulator strings are arranged in parallel and adjacent, the number of insulators of these insulator strings is equal and the insulators on different strings correspond in sequence, and the structural morphology is extremely similar. When an insulator in an insulator string explodes, an insulator is missing at the position of the self-exploding insulator. Based on this, the present invention distinguishes the defect detection of insulator strings in high-voltage transmission networks from other defect detections. At the hovering point of the insulator string, based on the special properties of the insulator string, the image processor carried by the inspection drone itself is used to quickly and accurately detect the self-explosion of the insulator.

[0029] Specifically, Figure 2 As shown, in the present invention, at the hovering point of the insulator string, the pan-tilt camera collects images according to a preset angle and uploads the images to the image processor carried by the inspection drone itself. The image processor quickly detects whether there is an insulator string based on the image collected by the pan-tilt camera. If there is an insulator string, the number of insulator strings is further detected. If the number of insulator strings is greater than 1, the suspension type of multiple insulator strings is further detected, and the number and suspension type of the insulator strings are fed back to the control system. The image processor further detects whether there is a self-explosive insulator based on the number and suspension type of the insulator strings. The suspension type includes multiple insulator strings that are parallel and adjacent, and multiple insulator strings that are at a certain angle. Among them, whether they are adjacent is determined by the distance between the parallel insulator strings, and the insulators in the insulator strings are glass insulators.

[0030] Specifically, when the image processor detects that the number of insulator strings in the image captured at the hovering point of the insulator string is 1, the image processor extracts an image of one insulator string according to a preset size, and rotates the image 180 degrees to obtain a deformed image, and calculates the similarity between the extracted insulator string image and its deformed image through image comparison. If the similarity is lower than a first threshold, it is determined that a self-destructing insulator appears in the insulator string. At the same time, it is considered that the image captured by the gimbal camera at a preset angle at the hovering point of the insulator string meets the relevant requirements, and the detection result is fed back to the control system. The control system controls the signal transmission system to upload the detection result and the image captured by the gimbal camera to the remote work platform.

[0031] When the image processor detects that the number of insulator strings in the image collected at the hovering point of the insulator string is greater than 1 and the multiple insulator strings are at a certain angle, that is, different insulator strings are suspended in different directions, the image processor extracts the image of each insulator string in the multiple insulator strings according to the preset size, and rotates the image of each insulator string by 180 degrees to obtain their respective deformed images, and calculates their respective similarities by comparing the image of each insulator string with their respective deformed images. Each calculated similarity is compared with the second threshold in turn, and it is determined that the insulator string corresponding to the similarity below the second threshold has a self-explosive insulator, and the position of the insulator string where the self-explosive insulator appears is recorded. At the same time, it is considered that the image collected by the gimbal camera at the preset angle at the hovering point of the insulator string meets the relevant requirements, and the detection result is fed back to the control system, and the control system controls the signal transmission system to upload the detection result and the image collected by the gimbal camera to the remote work platform.

[0032] When the image processor detects that the number of insulator strings in the image collected at the hovering point of the insulator string is greater than 1 and the multiple insulator strings are parallel and adjacent, the image processor extracts the image of each insulator string in the multiple insulator strings according to a preset size to obtain multiple insulator string images. The multiple insulator string images are marked as image 1, image 2, ..., image n in sequence, and an image set {image 1, image 2, ..., image n} is constructed, where n is an integer greater than or equal to 2. Compare all any two images in the image set {image 1, image 2, ..., image n} and calculate the similarity, that is, compare image 1 with image 2 to image n in sequence to obtain multiple similarities S12, S13 ... S1n, compare image 2 with image 3 to image n in sequence to obtain multiple similarities S23, S24 ... S2n, ..., compare image n-1 with image n to obtain similarity S(n-1)n. Through the above calculation, multiple similarities will be obtained, and a similarity sequence [S12, S13, ..., S1n, S23, S24, ..., S2n, ..., S(n-1)n] will be constructed. The variance of the similarity sequence is calculated, and the variance is compared with a first variance threshold. If the calculated variance is greater than the first variance threshold, it is determined that a self-destructing insulator has occurred. At the same time, it is believed that at the hovering point of the insulator string, the image captured by the gimbal camera at the preset angle meets the relevant requirements, and the detection results are fed back to the control system. The control system controls the signal transmission system to upload the detection results and the image captured by the gimbal camera to the remote work platform.

[0033] When the number of insulator strings is greater than 1 and multiple insulator strings are parallel and adjacent, considering the impact of the insulator self-explosion on its adjacent insulators, resulting in the simultaneous self-explosion of multiple insulator strings at the same position, when the variance of the similarity calculated according to the above method is less than or equal to the first variance threshold, it cannot be directly considered that there is no self-exploding insulator. It should also be further selected that one of the multiple insulator string images extracted is rotated 180 degrees to obtain a deformed image, and the selected insulator string image is compared with its deformed image. If the similarity between the two is calculated to be greater than the third threshold, it is considered that there is no self-exploding insulator in the insulator string, and it can be further considered that there is no self-exploding insulator in the multiple insulator strings, so that the detection result is more accurate.

[0034] More specifically, for high-voltage power towers, when the presence of self-explosion of insulators is detected, it is necessary to promptly install new insulators at the positions of the self-explosion insulators or replace the entire insulator string. In order to further confirm which insulator string or which insulator strings the self-explosion insulator specifically appears on when multiple insulator strings are parallel and adjacent, as a preferred embodiment, image 1, image 2, ... image n are respectively rotated 180 degrees to obtain respective deformed images image 1', image 2', ... image n', and by comparing the image of each insulator string with the respective deformed images, their respective similarities are calculated. Each calculated similarity is compared with the fourth threshold, and it is determined that the self-explosion insulator appears in the insulator string corresponding to the similarity lower than the fourth threshold, and the position of the insulator string where the self-explosion insulator appears is recorded.

[0035] like Figure 3As shown, in another embodiment of the present invention, when the number of insulator strings is greater than 1 and multiple insulator strings are parallel and adjacent, in order to make the judgment more accurate, multiple insulator string images are marked as image 1, image 2, ..., image n in sequence, and a first image set {image 1, image 2, ..., image n} is constructed, where n is an integer greater than or equal to 2. After the m images in the first image set {image 1, image 2, ..., image n} are rotated 180 degrees, m deformed images are obtained, and a second image set {image a1, image a2, ..., image am} is constructed. The first image set {image 1, image 2, ..., image n} and the second image set {image a1, image a2, ..., image am} are merged to obtain a third image set {image 1, image 2, ..., image n, image a1, image a2, ..., image am}. All any two images in the third image set are compared to calculate similarity to obtain a similarity series, and the variance is calculated, and the variance is compared with the second variance threshold. For example, when there are three insulator strings, the image processor extracts the image of each insulator string in the insulator string according to the preset size to obtain image 1, image 2, and image 3, and rotates image 2 and image 3 by 180 degrees to obtain deformed images image 2' and image 3', and constructs image 1, image 2, image 3, image 2', and image 3' into a third image set {image 1, image 2, image 3, image 2', and image 3'}, and sequentially compares image 1 with image 2, image 3, image 2', and image 3' to obtain multiple similarities, and sequentially compares image 2 with image 3, image 2', and image 3' to obtain multiple similarities, and sequentially compares image 3 with image 2' and image 3' to obtain multiple similarities, and compares image 2' with image 3' to obtain similarities, and constructs a similarity series. Calculate the variance of the similarity series, and compare the variance with the second variance threshold. If the calculated variance is greater than the second variance threshold, it is considered that there is a self-exploding insulator. The number m of insulator string images that undergo a 180-degree rotation process is greater than or equal to 1 and less than or equal to the total number of extracted insulator string images. In a specific embodiment, the number m of insulator string images that undergo a 180-degree rotation process is greater than one-half of the total number of extracted insulator string images. In another specific embodiment, the number m of insulator string images that undergo a 180-degree rotation process is positively correlated with the number of insulators on the insulator string.

[0036] In an embodiment of the present invention, the size of the extracted insulator string image is 100*600 or other suitable size that can cover a complete insulator string. In addition, at the hovering point of the insulator string, after the image processor completes the relevant detection, it sends a detection completion signal to the control system, and after the control system receives the detection completion signal, it controls the inspection drone to go to the next point.

[0037] In addition, in an embodiment of the present invention, if the calculated similarity is less than a fifth threshold value, it indicates that the extracted insulator string image does not meet the image processing requirements, and the image processor feeds back a signal to the control system. The control system controls the gimbal camera to adjust the orientation for secondary image acquisition according to the feedback signal, and the image processor performs relevant processing based on the secondary acquired image.

[0038] The first threshold, the second threshold, the third threshold, the fourth threshold, the fifth threshold, the first variance threshold and the second variance threshold of the present invention are set based on experience and can be fine-tuned. The method for the image processor to judge the similarity of the extracted insulator string images includes judging the image similarity based on texture and shape information, judging the image similarity based on shape and edge information in the binary image, and judging the image similarity based on a histogram.

[0039] The technical features of the above-described embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above-described embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0040] The above-mentioned embodiments only express several embodiments of the present invention, and the description is relatively specific and detailed, but it cannot be understood as limiting the scope of the patent of the present invention. It should be pointed out that for ordinary technicians in this field, several modifications and improvements can be made without departing from the concept of the present invention, which all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be based on the attached claims.

Claims

1. A remote monitoring method for power equipment based on a network security protocol, comprising planning the inspection route and hovering point of an inspection drone according to the high-voltage transmission network segment to be inspected, wherein when the inspection drone hovers at the hovering point, the gimbal camera carried by the inspection drone automatically focuses to collect images, and uploads the images to a remote work platform and / or an image processor carried by the inspection drone; Features: The hovering points are divided into common hovering points and insulator string hovering points according to different detection targets; at the insulator string hovering points, the gimbal camera collects images at a preset angle, and the image processor detects whether there is an insulator string based on the image collected by the gimbal camera. If there is an insulator string, the number of insulator strings is further detected. If the number of insulator strings is greater than 1, the suspension type of multiple insulator strings is further detected, and the number and suspension type of the insulator strings are fed back to the control system; the suspension types include multiple insulator strings being parallel and adjacent, and multiple insulator strings being at a certain angle; the image processor further detects whether there is a self-explosive insulator based on the number and suspension type of the insulator strings.

2. The method for remote monitoring of electric power equipment according to claim 1, characterized in that: When the number of insulator strings in the image collected at the hovering point of the insulator string is 1, the image processor extracts an image of one insulator string according to a preset size, rotates the extracted insulator string image 180 degrees to obtain a deformed image, and calculates the similarity between the extracted insulator string image and its deformed image through image comparison. If the similarity is lower than a first threshold, it is determined that a self-explosive insulator appears in the insulator string.

3. The remote monitoring method for electric power equipment according to claim 1, characterized in that: When the number of insulator strings in the image collected at the hovering point of the insulator string is greater than 1 and the multiple insulator strings form a certain angle, the image processor extracts the image of each insulator string according to a preset size, and rotates the extracted image of each insulator string by 180 degrees to obtain respective deformed images, and calculates their respective similarities by comparing the image of each insulator string with their respective deformed images, and compares each calculated similarity with a second threshold value to determine whether a self-explosive insulator exists in the insulator string corresponding to the similarity lower than the second threshold value.

4. The remote monitoring method for electric power equipment according to claim 1, characterized in that: When the number of insulator strings in the image collected at the hovering point of the insulator string is greater than 1 and the multiple insulator strings are parallel and adjacent, the image processor extracts the image of each insulator string according to a preset size, and marks them as image 1, image 2, ..., image n in sequence, and constructs a first image set {image 1, image 2, ..., image n}, where n is an integer greater than or equal to 2; Compare any two images in the first image set, calculate the similarity, obtain a first similarity sequence, further calculate the variance of the first similarity sequence, compare the variance of the first similarity sequence with a first variance threshold, and when the variance of the first similarity sequence is greater than the first variance threshold, determine that a self-destructing insulator appears; Alternatively, m images in the first image set are selected and rotated 180 degrees to obtain a second image set {image a1, image a2, ..., image am}, where m is greater than or equal to 1 and less than or equal to n, the first image set and the second image set are combined into a third image set {image 1, image 2, ..., image n, image a1, image a2, ..., image am}, any two images in the third image set are compared, the similarity is calculated, a second similarity series is obtained, the variance of the second similarity series is further calculated, the variance of the second similarity series is compared with a second variance threshold, and when the variance of the second similarity series is greater than the second variance threshold, it is determined that a self-destructing insulator has occurred.

5. The method for remote monitoring of electric power equipment according to claim 4, characterized in that: When the variance of the first similarity series is less than or equal to the first variance threshold, any one image in the first image set is further selected, and the image is rotated 180 degrees to obtain a deformed image. The selected image is compared with the deformed image to calculate the similarity. If the similarity is greater than the third threshold, it is determined that there is no self-destructing insulator.

6. The remote monitoring method for electric power equipment according to any one of claims 2 to 4, characterized in that: If the calculated similarity is less than the fifth threshold, the image processor feeds back a signal to the control system, and the control system controls the gimbal camera to adjust its position for secondary image acquisition according to the feedback signal.

7. The remote monitoring method for electric power equipment according to any one of claims 2 to 5, characterized in that: At the hovering point of the insulator string, after the image processor completes the relevant detection, it sends a detection completion signal to the control system. After receiving the detection completion signal, the control system controls the inspection drone to move to the next point.

8. The remote monitoring method for electric power equipment according to any one of claims 1 to 7, characterized in that: The insulators in the insulator string are glass insulators.

9. A remote monitoring system for power equipment based on a network security protocol, comprising an inspection drone equipped with a pan-tilt camera, a control system, an image processor, and a signal transmission system. The pan-tilt camera is used to collect high-definition pictures or videos during inspections. The control system controls the flight status of the inspection drone and the position of the pan-tilt camera according to instructions. The image processor is used to process pictures collected by the pan-tilt camera. The signal transmission system transmits signals and data. The system is used to implement the remote monitoring method for power equipment described in any one of claims 1 to 8.

10. The remote monitoring system for electric power equipment according to claim 9, characterized in that: The transmission of signals and data is based on 5G network security protocols.

Images