A method and system for local discharge location of a power transmission tower

Abstract

The present invention relates to the field of power technology, and in particular to a method and system for local discharge positioning of a power pole tower. Among them, the method for local discharge positioning of the power pole tower includes the following steps: obtaining a first database; obtaining a visible light image and an energy response image of the target power pole tower; based on the structure and terminals of the target power pole tower in the visible light image, matching the target three-dimensional structure model of the target power pole tower from the first database, and positioning the positions of the terminals on the target three-dimensional structure model; based on the energy response image corresponding to the visible light image, obtaining the discharge probability of the terminals in the visible light image; according to the positions of each terminal on the target three-dimensional structure model and the discharge probability, positioning the abnormal discharge terminals in the target power pole tower. The present invention is based on model database matching and multi-view data fusion, and can accurately locate the abnormal discharge terminals on the power pole tower, which helps to improve the overall power supply stability and reliability of the power system.

Description

Technical Field

[0001] The present invention relates to the field of power technology, and in particular to a method and system for local discharge location of power transmission towers. Background Art

[0002] As a key facility of the power transmission network, power transmission towers may experience partial discharges due to electrical overload, aging, or external environmental factors (such as lightning strikes, moisture, etc.). Failure to detect and handle them in a timely manner may lead to equipment damage, outages, or even fires. Therefore, timely and accurate detection of the discharge points on power transmission towers is crucial for ensuring the safe and stable operation of the power system.

[0003] Currently, technicians mainly use handheld ultraviolet imagers and ultrasonic imagers to detect discharges on power transmission towers. Workers usually stand on the ground and use these instruments to photograph the discharge points on the towers, and use ultraviolet or thermal images to identify the discharge sources.

[0004] However, due to the relatively long distance between the workers and the towers, and the image overlap during upward shooting, it is difficult to accurately locate the discharge points, especially for complex structures and high-altitude discharge points. Moreover, the existing multi-lens design has image alignment errors, resulting in difficulty in matching the thermal map and the ultraviolet imaging image, which affects the positioning accuracy. At the same time, the accuracy of the thermal map of the existing instruments is relatively low in the case of long distance and complex environments (such as high temperature, strong light, or bad weather), and the contrast between the discharge point and the background is low, increasing the difficulty of identification. Summary of the Invention

[0005] In view of the deficiencies of the prior art and the requirements of practical applications, the present invention provides a method and system for local discharge location of power transmission towers, aiming to accurately locate the discharge area of the terminals of power transmission towers, improve the accuracy of discharge detection, and thus effectively reduce the risk of power system failures caused by discharges.

[0006] In an embodiment provided by the present invention, the method for local discharge location of power transmission towers includes the following steps:

[0007] Obtain a first database, which stores a number of power transmission tower structure model samples, and any one of the power transmission tower structure model samples is a three-dimensional structure model;

[0008] Obtain the visible light image and the energy response image of the target power transmission tower. The visible light image includes the local structure image or the overall structure image of the target power transmission tower, and includes at least one terminal image, and the energy response image is aligned with the visible light image in terms of image;

[0009] Based on the target power transmission tower structure and terminals in the visible light image, match the target three-dimensional structure model of the target power transmission tower from the first database, and locate the positions of the terminals on the target three-dimensional structure model;

[0010] Based on the energy response image corresponding to the visible light image, obtain the discharge probability of the terminals in the visible light image;

[0011] According to the positions of each terminal on the target three-dimensional structure model and the discharge probability, locate the abnormally discharging terminals in the target power transmission tower.

[0012] In this embodiment or other embodiments, the obtaining of the first database includes the following steps:

[0013] Obtain the power transmission tower structure feature data in the current power industry. The power transmission tower structure feature data includes the structure features of the power transmission tower and the terminal configuration parameters of the power transmission tower under the same structure features. The terminal configuration parameters include the number of terminals and the spatial coordinates of the terminals;

[0014] Based on the power transmission tower structure feature data, construct three-dimensional structure models of different types of power transmission towers, and mark the spatial coordinates and spatial coordinate numbers of each terminal in each three-dimensional structure model;

[0015] Summarize the three-dimensional structure models of different types of power transmission towers to obtain the first database.

[0016] In this embodiment or other embodiments, based on the target power transmission tower structure in the visible light image, matching the target three-dimensional structure model of the target power transmission tower from the first database includes the following steps:

[0017] Based on the target power transmission tower structure, obtain the structure features of the target power transmission tower;

[0018] Using the structure features, match the target three-dimensional structure model in the first database through image recognition technology.

[0019] In this embodiment or other embodiments, based on the terminals in the visible light image, locating the positions of the terminals on the target three-dimensional structure model includes the following steps:

[0020] Obtain the image coordinates of each terminal in the visible light image, and generate a data group to be matched. One data to be matched in the data group to be matched is the image coordinate of a terminal;

[0021] Obtain all the terminal configuration parameters of the power transmission tower under the target three-dimensional structure model, and generate one or more groups of matching data sets based on all the terminal configuration parameters. One matching data in the matching data set is the spatial coordinate of a terminal under a certain terminal configuration parameter of the target three-dimensional structure model;

[0022] Based on the data set to be matched, obtain the target matching data set from one or more groups of matching data sets. Each image coordinate in the data set to be matched is mapped to a spatial coordinate in the target matching data set through a unique mapping method, and the spatial coordinates mapped by each image coordinate in the data set to be matched in the target matching data set are different;

[0023] Locate the spatial coordinate number of each image coordinate in the data set to be matched through the matching data corresponding to the data to be matched in the target matching data set through the unique mapping method.

[0024] In this embodiment or other embodiments, obtaining the visible light image and the energy response image of the target power transmission tower includes the following steps:

[0025] Provide a first image acquisition device for acquiring the visible light image of the target power transmission tower;

[0026] Provide a second image acquisition device for acquiring an energy response image aligned with the visible light image.

[0027] In this embodiment or other embodiments, obtaining the discharge probability of the terminal in the visible light image based on the energy response image corresponding to the visible light image includes the following steps:

[0028] Based on the image resolution of the energy response image, construct a single-pixel point discharge distribution model. The single-pixel point discharge distribution model is a Gaussian energy distribution response matrix based on the central matrix element, and the size of the energy distribution response matrix is the same as the image resolution;

[0029] Obtain the discharge probability of the terminal in the visible light image according to the energy response value of each matrix element in the energy distribution response matrix and each image pixel in the energy response image.

[0030] In this embodiment or other embodiments, the central matrix element of the energy distribution response matrix has the maximum ultraviolet energy response coefficient , and the values of the remaining elements of the energy distribution response matrix are , where represents the matrix element coordinate, represents the central matrix element coordinate, Represents the ultraviolet energy diffusion coefficient;

[0031] The discharge probability of the terminal in the visible light image satisfies the following calculation model: , where Represents the coordinate of the matrix element The ultraviolet energy response coefficient at the position, Represents the energy response value at the image pixel coordinate at the position in the energy response image, Represents the discharge response standard value.

[0032] In this embodiment or other embodiments, locating the abnormal discharge terminal in the target power transmission tower according to the position and discharge probability of each terminal in the target three-dimensional structure model includes the following steps:

[0033] Based on the visible light images and the corresponding energy response images at different shooting angles, obtain the discharge probability of each terminal at different shooting angles;

[0034] Respectively obtain the final discharge probability of each terminal according to the discharge probability of each terminal at different shooting angles, and set the terminal whose final discharge probability exceeds the prefabricated value as the abnormal discharge terminal, and locate the abnormal discharge terminal on the target power transmission tower according to the position of the abnormal discharge terminal.

[0035] In this embodiment or other embodiments, the final discharge probability of any terminal satisfies the following calculation model: , , ;

[0036] Among them, i represents the serial number, and i takes any integer between 1 and n, , Represents the discharge probability of the terminal at the first shooting angle, Represents the discharge probability of the terminal at the second shooting angle, Represents the discharge probability of the terminal at the nth shooting angle, Represents the discharge probability of the terminal at the ith shooting angle, Represents the final discharge probability, Represents The expectation of, Represents the discharge probability of the terminal at the ith shooting angle Based on the posterior probability of the initial discharge probability , Represents the discharge probability The marginal likelihood function value of, Represents the initial discharge probability of the terminal, Represents the posterior discharge probability of the terminal, Indicates the discharge probability The value of the likelihood function.

[0037] In this embodiment or other embodiments, the present invention is based on the power pole tower partial discharge locating method, and also provides a power pole tower partial discharge locating system; the power pole tower partial discharge locating system includes an input device, a processor, a memory and an output device; wherein the input device, the processor, the memory and the output device are interconnected; the memory is used to store a computer program, the computer program includes program instructions, and the processor is configured to call the program instructions to execute the power pole tower partial discharge locating method provided in this embodiment.

[0038] The advantages of the method and system for locating partial discharge of power towers provided by the present invention are:

[0039] The present invention uses an energy response image corresponding to a visible light image to evaluate the discharge probability of each terminal on the visible light image, thereby identifying the abnormal discharge terminal in the visible light image; at the same time, based on image recognition technology, the present invention identifies and matches a three-dimensional structural model with the same structural features of a power tower in the visible light image from a first database built based on current power industry data, thereby matching the terminals on the visible light image with the terminals on the target three-dimensional structural model, thereby achieving accurate positioning of the abnormal discharge terminals on the visible light image, avoiding the mismatching problem caused by inaccurate image alignment, viewing angle deviation or lens distortion in traditional detection and positioning methods;

[0040] Furthermore, the present invention also makes the abnormal discharge detection with the terminal as the center of the image more accurate by setting a single pixel point discharge distribution model at the center of the matrix; and improves the micro discharge detection capability, that is, it can effectively capture smaller discharge signals, especially in the case of complex structures or long-distance shooting, and can still accurately identify abnormal discharge terminals;

[0041] Furthermore, the present invention also effectively supplements the deficiencies of a single perspective by fusing multi-perspective data taken at different angles, thereby achieving higher-precision discharge source positioning, further reducing the error caused by single-perspective detection, and improving the robustness and accuracy of the overall positioning.

[0042] Furthermore, the present invention helps the power sector to implement preventive maintenance by accurately locating the discharge area, effectively avoiding the chain reaction caused by the failure to discover the discharge fault in time, thereby reducing the risk of large-scale power outages and equipment damage, reducing the high maintenance costs and safety hazards caused by sudden failures, and improving the overall power supply stability and reliability of the power system.

[0043] Furthermore, the power pole partial discharge positioning system architecture provided by the present invention realizes the automated detection, analysis, and positioning of power pole partial discharges by seamlessly integrating input devices, processors, memories, and output devices. It can also respond to and process complex image data in real time, greatly improving the efficiency and safety of power maintenance, and providing strong technical support for the long-term stable operation of the power system. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 is a flowchart of the power pole partial discharge positioning method provided by an embodiment of the present invention;

[0045] Figure 2 is a schematic structural diagram of the dual-mode image acquisition device provided by an embodiment of the present invention;

[0046] Figure 3 is a schematic diagram of the power pole partial discharge positioning system provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0047] In the following description, specific details such as specific system structures and technologies are presented for the purpose of illustration rather than limitation, in order to thoroughly understand the embodiments of the present application.

[0048] Those skilled in the art should understand that the present application can also be implemented in other embodiments without these specific details. In the description of the present application, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to avoid unnecessary details from interfering with the description of the present application. Additionally, terms such as "first" and "second" are only used for distinguishing descriptions and should not be construed as indicating or implying relative importance.

[0049] Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings, and other embodiments can also be obtained.

[0050] In one embodiment, please refer to Figure 1 , Figure 1 is a flowchart of the power pole partial discharge positioning method provided by an embodiment of the present invention.

[0051] For the precise positioning of partial discharges in power poles, as Figure 1 shown, the power pole partial discharge positioning method provided by the present invention includes the following steps:

[0052] S01. Obtain the first database.

[0053] The first database obtained in step S01 stores a number of power pole structure model samples. Any one of the power pole structure model samples is a three-dimensional structure model. The first database also stores one or more groups of terminal matching parameters corresponding to each power pole structure model sample.

[0054] Further, the obtaining of the first database in step S01 includes the following steps:

[0055] S011. Obtain the power pole structure feature data in the current power industry. The power pole structure feature data includes the structure features of the power pole and the terminal configuration parameters of the power pole under the same structure features. The terminal configuration parameters include the number of terminals and the terminal space coordinates.

[0056] It can be realized that the power pole structure feature data in the current power industry can be obtained through existing public data such as power system design specifications, that is, all the structure types of power poles in the current power industry and one or more terminal configuration parameters under each structure type are obtained.

[0057] Further, the structure types include but are not limited to straight poles, angle towers, guyed poles, cat head poles, etc.; among them, based on the cat head pole, the following terminal configuration parameters include but are not limited to:

[0058] For a 1-meter cat head pole, 3*2 cables are suspended (3-phase cables, with 2 insulating terminals configured for each phase); for a 1.1-meter cat head pole, 3*2 cables are suspended (3-phase cables, with 2 insulating terminals configured for each phase); for a 1.2-meter cat head pole, 3*2 cables are suspended (3-phase cables, with 2 insulating terminals configured for each phase);

[0059] For a 1-meter cat head pole, 6*2 cables are suspended (6-phase cables, with 2 insulating terminals configured for each phase); for a 1.1-meter cat head pole, 6*2 cables are suspended (6-phase cables, with 2 insulating terminals configured for each phase); for a 1.2-meter cat head pole, 6*2 cables are suspended (6-phase cables, with 2 insulating terminals configured for each phase);

[0060] For a 1-meter cat head pole, 6*4 cables are suspended (6-phase cables, with 4 insulating terminals configured for each phase); for a 1.1-meter cat head pole, 6*4 cables are suspended (6-phase cables, with 4 insulating terminals configured for each phase); for a 1.2-meter cat head pole, 6*4 cables are suspended (6-phase cables, with 4 insulating terminals configured for each phase).

[0061] It can be understood that for a power pole under any specific type and terminal configuration parameters, step S011 also obtains other relevant data, such as arm length, angle, etc., for subsequent restoration of a three-dimensional structure model with accurate proportions in 3D modeling software.

[0062] S012. Based on the structural feature data of the power transmission tower, construct 3D structural models of different types of power transmission towers, and mark the spatial coordinates and spatial coordinate numbers of each terminal in each 3D structural model.

[0063] Specifically, based on the structural feature data of the power transmission tower, step S012 can be specifically implemented through CAD tools and 3D modeling software (such as SolidWorks, AutoCAD, etc.).

[0064] It can be known that the above or other 3D modeling software can accurately construct the 3D model of the power transmission tower according to the set dimensions and parameters; further, in the 3D structural model of each type of power transmission tower, the spatial coordinates of each terminal relative to the 3D structural model of the power transmission tower can be obtained under different terminal configuration parameters, and one or more terminals under the same terminal configuration parameters can be numbered based on different spatial coordinates of the same 3D structural model of the power transmission tower.

[0065] S013. Summarize the 3D structural models of different types of power transmission towers to obtain the first database.

[0066] Further, based on the 3D structural models of different structural types constructed in step S012 and a set or multiple sets of terminal matching parameters corresponding to each structural type, construct the first database, and the first database satisfies the following representation model , where represents the first database, is the first 3D structural model in the first database, is the second 3D structural model in the first database, is the third 3D structural model in the first database.

[0067] Furthermore, based on any 3D structural model in the first database , its corresponding terminal configuration parameter set is as follows: , where represents the serial number, r takes positive integers from 1 to R, and R is the number of samples in the first database, represents the terminal configuration parameter set corresponding to the th 3D structural model, represents the first set of configuration parameters of the th 3D structural model, represents the second set of configuration parameters of the th 3D structural model, represents the third set of configuration parameters of the

[0068] Furthermore, based on the Any set of configuration parameters in the three-dimensional structure model corresponds to the following specific configuration data: where g represents the serial number, and g takes positive integers from 1 to G, represents the g-th set of configuration parameters of the three-dimensional structure model, represents the spatial coordinates of the first terminal in the g-th set of configuration parameters of the three-dimensional structure model, represents the spatial coordinates of the m-th terminal in the g-th set of configuration parameters of the three-dimensional structure model.

[0069] S02. Obtain the visible light image and the energy response image of the target power transmission tower.

[0070] In this embodiment, the visible light image includes the local structure image or the overall structure image of the target power transmission tower, and includes at least one terminal image, and the energy response image is aligned with the visible light image.

[0071] It should be noted that the alignment of the energy response image and the visible light image in this embodiment means that the images shown by the pixels of the two images are consistent in space. Specifically, it can be understood that: if the visible light image shows the surface of a terminal, the energy response image also shows the energy response corresponding to the surface of this terminal.

[0072] Furthermore, obtaining the visible light image and the energy response image of the target power transmission tower in step S02 includes the following steps:

[0073] S021. Provide a first image acquisition device, which is used to obtain the visible light image of the target power transmission tower.

[0074] It can be understood that the first image acquisition device provided in this embodiment can be an imaging device with high resolution such as a digital camera or an industrial camera, which is used to collect the local visible light image or the overall visible light image of the target power transmission tower.

[0075] S022. Provide a second image acquisition device, which is used to obtain the energy response image aligned with the visible light image.

[0076] It can be understood that the second image acquisition device provided in this embodiment can be a photoelectric sensor such as a matrix ultraviolet light sensor or a matrix infrared light sensor and the corresponding data processing device, which is used to obtain the energy response intensity of the specific wavelength light corresponding to any position in the visible light image by adjusting the shooting angle or setting the optical path relationship related to the first image acquisition device.

[0077] Further, to facilitate the image alignment between the energy response image and the visible light image and improve the detection efficiency, the present embodiment provides a dual-mode image acquisition device integrating a first image acquisition device and a second image acquisition device. The two are split by a semi-reflective and semi-transmissive lens, so as to realize different types of imaging for the same incident light.

[0078] Please refer to Figure 2 , Figure 2 which is a schematic structural diagram of the dual-mode image acquisition device provided by the embodiment of the present invention. As Figure 2 shown, the dual-mode image acquisition device integrates the first image acquisition device 10 and the second image acquisition device 20 in a housing 00. An incident light port 30 is provided on the housing 00, and a light splitting device 40 is provided inside the incident light port 30 for splitting the incident light into a first light and a second light, so as to respectively enter the first image acquisition device 10 and the second image acquisition device 20, and then generate a visible light image and an energy response image respectively.

[0079] In some other embodiments, the first image acquisition device, the second image acquisition device, the dual-mode image acquisition device and other types of image acquisition devices can all realize image acquisition through operation modes such as drones, inspection robots or manual operation.

[0080] S03. Based on the target power transmission tower structure and terminals in the visible light image, match the target three-dimensional structure model of the target power transmission tower from the first database, and locate the positions of the terminals on the target three-dimensional structure model.

[0081] In the present embodiment, the matching of the target three-dimensional structure model of the target power transmission tower from the first database based on the target power transmission tower structure in step S03 includes the following steps:

[0082] S0311. Based on the target power transmission tower structure, obtain the structural features of the target power transmission tower.

[0083] It can be understood that the structural features in step S011 include but are not limited to the overall or partial shape, overall or partial construction method, overall or partial support method, overall or partial terminal position distribution, etc. that can be collected from the visible light image.

[0084] Further, the structural features of any target power transmission tower can be identified and extracted by means of manual work, image recognition technologies (such as convolutional neural network (CNN, etc.), object detection technologies (YOLO series, Faster R-CNN, etc.), etc.).

[0085] S0312. Using the structural features, match the target three-dimensional structure model in the first database through image recognition technology.

[0086] Furthermore, based on one or more structural features collected in step S0311, step S0312 can be implemented through existing image matching technologies such as object detection technology (YOLO series, etc.) and feature matching algorithms (such as SURF, SIFT).

[0087] In this embodiment, the step of positioning the position of the terminal on the target three-dimensional structure model based on the terminal in the visible light image in step S03 includes the following steps:

[0088] S0321. Obtain the image coordinates of each terminal in the visible light image and generate a set of data to be matched. One piece of data to be matched in the set of data to be matched is the image coordinate of one terminal.

[0089] Similarly, image recognition technology can be used to identify each terminal from the visible light image based on the shape or structural features of the terminal, and detect the two-dimensional image coordinates of each terminal in this visible light image, which can be specifically represented as: , where represents the image coordinate of the first terminal in the visible light image, represents the image coordinate of the second terminal in the visible light image, represents the image coordinate of the third terminal in the visible light image.

[0090] S0322. Obtain all the terminal configuration parameters of the power transmission tower under the target three-dimensional structure model, and based on all the terminal configuration parameters, generate one or more sets of matching data groups respectively. One piece of matching data in the matching data group is the spatial coordinate of one terminal under a certain terminal configuration parameter of the target three-dimensional structure model.

[0091] Refer to any set of terminal configuration parameters of the three-dimensional structure model in the above-mentioned embodiment of this example: , and the three-dimensional structure model any set of configuration parameters in .

[0092] It can be understood that one set of matching data generated in step S0322 is the spatial coordinate part in one set of configuration parameters of the above-mentioned three-dimensional structure model .

[0093] S0323. Based on the data group to be matched, obtain a target matching data group from one or more groups of matching data groups. Each image coordinate in the data group to be matched is mapped to a spatial coordinate in the target matching data group through a unique mapping method, and the spatial coordinates mapped by each image coordinate in the data group to be matched in the target matching data group are different.

[0094] It should be noted that the mapping method described in this embodiment specifically refers to a coordinate transformation method, that is, converting two-dimensional image coordinates into three-dimensional spatial coordinates, and the specific parameters include but are not limited to the internal or external parameters of the image acquisition device, projection matrix, perspective transformation matrix, etc.

[0095] Furthermore, the solution of the unique mapping method can be realized by setting an objective function and solving the objective function. It should be noted that since the terminal configuration parameter group under any structural type is limited, for each image coordinate in the data group to be matched, there is always a group of matching data groups, and some or all of the spatial coordinates in it correspond one-to-one with those in the data group to be matched, and the mapping methods between any corresponding relationships are the same.

[0096] In some other embodiments, based on the local structural features of the power transmission tower in the visible light image, and identifying a three-dimensional structural model that matches it, therefore, directly match the two-dimensional image coordinates of the terminals in the visible light image with the data in the first database one by one; similarly, since the structural types of the power transmission towers are limited, and at the same time the terminal configuration parameter groups under any structural type are limited, for each image coordinate in the data group to be matched, there is always a group of matching data groups, and some or all of the spatial coordinates in it correspond one-to-one with those in the data group to be matched, and the mapping methods between any corresponding relationships are the same, or all the mapping methods have the smallest difference.

[0097] S0324. Locate the spatial coordinate number of each image coordinate in the data group to be matched through the matching data in the target matching data group that corresponds to the data to be matched through the unique mapping method.

[0098] It can be understood that under a set of terminal configuration parameters corresponding to any three-dimensional structural model, any terminal has a unique spatial coordinate number based on its different spatial coordinates from the other terminals; therefore, step S0324 can further locate the corresponding image coordinates according to the spatial coordinate numbers that match each image coordinate in the target matching data group.

[0099] S04. Based on the energy response image corresponding to the visible light image, obtain the discharge probability of the terminals in the visible light image.

[0100] In actual situations, abnormal discharges may occur not only at the terminals on the power pole, but also at other parts of the power pole.

[0101] Therefore, it should be noted that the abnormal discharge point identified in the embodiments of the present invention is the terminal of the power pole. To reduce the influence of other abnormal discharges on the implementation of this embodiment, the image centers of the visible light images and their corresponding energy response images obtained in this embodiment are both a terminal of the power pole.

[0102] Furthermore, to obtain the discharge probability of each terminal on the power pole, the step of obtaining the discharge probability of the terminal in the visible light image based on the energy response image corresponding to the visible light image in step S04 includes the following steps:

[0103] S041. Based on the image resolution of the energy response image, construct a single-pixel point discharge distribution model. The single-pixel point discharge distribution model is a Gaussian energy distribution response matrix based on the central matrix element, and the size of the energy distribution response matrix is the same as the image resolution.

[0104] Even further, the central matrix element of the energy distribution response matrix has the maximum ultraviolet energy response coefficient , and the values of the remaining elements of the energy distribution response matrix are , where represents the matrix element coordinates, represents the central matrix element coordinates, represents the ultraviolet energy diffusion coefficient.

[0105] Among them, the ultraviolet energy diffusion coefficient is a parameter used to describe how ultraviolet light spreads in space. It is used in the Gaussian distribution model to control the rate of change of energy intensity with distance, and it can be set using historical detection data and actual test experience.

[0106] S042. According to the energy response value of each matrix element in the energy distribution response matrix and each image pixel in the energy response image, obtain the discharge probability of the terminal in the visible light image.

[0107] In this embodiment, at least one terminal is included in any visible light image, and image acquisition is performed with one terminal as the image center; further, the discharge probability of the terminal corresponding to the image center in the visible light image satisfies the following calculation model: , where represents the ultraviolet energy response coefficient at the matrix element coordinates , represents the energy response value at the image pixel coordinates in the energy response image, Represents the standard value of the discharge response.

[0108] Among them, the standard value of the discharge response Is the standard value of the ultraviolet light response intensity for judging whether there is abnormal discharge at this position, and it can be set according to historical detection data and actual test experience.

[0109] S05. Locate the abnormal discharge terminals in the target power pole tower according to the positions of each terminal in the target three-dimensional structure model and the discharge probability.

[0110] Furthermore, to reduce the error caused by single-view detection, the step of locating the abnormal discharge terminals in the target power pole tower according to the positions of each terminal in the target three-dimensional structure model and the discharge probability in step S05 includes the following steps:

[0111] S051. Obtain the discharge probability of each terminal at different shooting angles based on the visible light images and the corresponding energy response images at different shooting angles.

[0112] S052. Obtain the final discharge probability of each terminal according to the discharge probability of each terminal at different shooting angles respectively.

[0113] Furthermore, based on the discharge probability of the terminal at different shooting angles, its final discharge probability satisfies the following calculation model: , , ; where i represents the serial number, and i takes any integer between 1 and n, , Represents the discharge probability of the terminal at the first shooting angle, Represents the discharge probability of the terminal at the second shooting angle, Represents the discharge probability of the terminal at the nth shooting angle, Represents the discharge probability of the terminal at the i-th shooting angle, Represents the final discharge probability, Represents The expectation of, Represents the discharge probability of the terminal at the i-th shooting angle Based on the initial discharge probability The posterior probability of, Represents the discharge probability The marginal likelihood function value of, Represents the initial discharge probability of the terminal, Represents the posterior discharge probability of the terminal, Represents the discharge probability The likelihood function value of.

[0114] In this embodiment, the initial discharge probability of each terminal is 50%, that is, each terminal may discharge. In other embodiments, the initial discharge probability of any terminal can also be set according to the actual situation or empirical values.

[0115] S053. Set the terminals with the final discharge probability exceeding the threshold as abnormal discharge terminals, and locate the abnormal discharge terminals on the target power pole tower according to the positions of the abnormal discharge terminals.

[0116] Furthermore, by combining with the target three-dimensional structure model, the positions of the abnormal discharge terminals can be directly displayed in the virtual three-dimensional model, enabling maintenance personnel to more intuitively understand the positions of the terminals and reducing the time and energy consumption during on-site maintenance.

[0117] In this embodiment, based on the above power pole tower partial discharge location method, the present invention also provides a power pole tower partial discharge location system.

[0118] Please refer to Figure 3 , Figure 3 , which is a schematic diagram of the power pole tower partial discharge location system provided by the embodiment of the present invention. As Figure 3 shown, the power pole tower partial discharge location system provided by the embodiment of the present invention includes an input device, a processor, a memory, and an output device.

[0119] Furthermore, the input device, the processor, the memory, and the output device are interconnected; the memory is used to store a computer program, the computer program includes program instructions, and the processor is configured to call the program instructions to execute the above power pole tower partial discharge location method.

[0120] In the above embodiments, the descriptions of the various embodiments have their own emphases. For parts not detailedly described or recorded in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0121] It should be noted that the above embodiments can be freely combined as needed. The above is only the preferred implementation mode of the present invention; it should be noted that for those of ordinary skill in the art, without departing from the principle of the present invention, several improvements and refinements can be made, and these improvements and refinements should also be regarded as the protection scope of the present invention.

Claims

1. A method for localizing partial discharge of a power transmission tower, characterized in that Including the following steps: Obtain a first database, in which a number of power pole structure model samples are stored, and any one of the power pole structure model samples is a three-dimensional structure model; Obtain a visible light image and an energy response image of the target power pole. The visible light image includes a partial structure image or an overall structure image of the target power pole and includes at least one terminal image, and the energy response image is aligned with the visible light image; Based on the structure and terminals of the target power pole in the visible light image, match the target three-dimensional structure model of the target power pole from the first database, and locate the positions of the terminals on the target three-dimensional structure model; Based on the energy response image corresponding to the visible light image, obtain the discharge probability of the terminals in the visible light image, including the following steps: Based on the image resolution of the energy response image, construct a single-pixel point discharge distribution model, which is a Gaussian energy distribution response matrix based on the central matrix element, and the size of the energy distribution response matrix is the same as the image resolution; According to the energy response values of each matrix element in the energy distribution response matrix and each image pixel in the energy response image, obtain the discharge probability of the terminals in the visible light image; According to the positions and discharge probabilities of each terminal on the target three-dimensional structure model, locate the abnormally discharging terminals in the target power pole, including the following steps: Based on the visible light images and the corresponding energy response images at different shooting angles, obtain the discharge probabilities of each terminal at different shooting angles; Respectively obtain the final discharge probability of each terminal according to the discharge probabilities of each terminal at different shooting angles, and set the terminals with the final discharge probability exceeding the preset value as abnormally discharging terminals, and locate the abnormally discharging terminals on the target power pole according to the positions of the abnormally discharging terminals.

2. The method for local discharge location of a power pole according to claim 1, wherein The obtaining of the first database includes the following steps: Obtain the power pole structure feature data in the current power industry. The power pole structure feature data includes the structure features of the power pole and the terminal configuration parameters of the power pole under the same structure features. The terminal configuration parameters include the number of terminals and the spatial coordinates of the terminals; Based on the power pole structure feature data, construct three-dimensional structure models of different types of power poles, and mark the spatial coordinates and spatial coordinate numbers of each terminal in each three-dimensional structure model; Summarize the three-dimensional structure models of different types of power poles to obtain the first database.

3. The method for local discharge positioning of a power transmission tower according to claim 2, wherein, Based on the structure of the target power pole in the visible light image, matching the target three-dimensional structure model of the target power pole from the first database includes the following steps: Based on the structure of the target power pole, obtain the structure features of the target power pole; Use the structure features to match the target three-dimensional structure model in the first database through image recognition technology.

4. The method for local discharge positioning of a power transmission tower according to claim 3, characterized in that, Based on the terminals in the visible light image, locating the positions of the terminals on the target three-dimensional structure model includes the following steps: Obtain the image coordinates of each terminal in the visible light image to generate a group of data to be matched, where one piece of data to be matched in the group of data to be matched is the image coordinate of one terminal. Obtain all the terminal configuration parameters of the power transmission tower under the target three-dimensional structure model, and based on all the terminal configuration parameters, generate one or more groups of matching data groups respectively. One piece of matching data in the matching data group is the spatial coordinate of one terminal under one terminal configuration parameter of the target three-dimensional structure model. Based on the group of data to be matched, obtain the target matching data group from one or more groups of matching data groups. Each image coordinate in the group of data to be matched is mapped to a spatial coordinate in the target matching data group through a unique mapping method, and the spatial coordinates mapped by each image coordinate in the group of data to be matched in the target matching data group are different. Through the matching data corresponding to the data to be matched in the target matching data group through the unique mapping method, locate the spatial coordinate number of each image coordinate in the group of data to be matched.

5. The method for local discharge positioning of a power transmission tower according to claim 1, characterized in that, The obtaining of the visible light image and the energy response image of the target power transmission tower includes the following steps: Provide a first image acquisition device, which is used to obtain the visible light image of the target power transmission tower. Provide a second image acquisition device, which is used to obtain an energy response image that is image-aligned with the visible light image.

6. The method for local discharge positioning of a power transmission tower according to claim 1, characterized in that, The central matrix element of the energy distribution response matrix has the maximum ultraviolet energy response coefficient α(x c , y c ), and the values of the remaining elements of the energy distribution response matrix are where (x, y) represents the coordinates of the matrix element, and (x c , y c ) represents the coordinates of the central matrix element, and σ represents the ultraviolet energy diffusion coefficient; The discharge probability of the terminal in the visible light image satisfies the following calculation model: P = (∑ (X，Y) [α(x, y)·I(x, y)]) / E std , where α(x, y) represents the ultraviolet energy response coefficient at the matrix element coordinate (x, y), I(x, y) represents the energy response value at the image pixel coordinate (x, y) in the energy response image, and E std represents the discharge response standard value.

7. The method for local discharge positioning of a power transmission tower according to claim 1, characterized in that, The final discharge probability of any terminal satisfies the following calculation model: P total = E[P(D total |d i )], Among them, i represents the serial number, and i takes any integer between 1 and n. D = {d1, d2,..., d n}, where d1 represents the discharge probability of the terminal at the first shooting angle, d2 represents the discharge probability of the terminal at the second shooting angle, d n represents the discharge probability of the terminal at the nth shooting angle, and d i represents the discharge probability of the terminal at the ith shooting angle. P total represents the final discharge probability. E[P(D total |d i )] represents the expectation of P(D total |d i ). P(D total |d i ) represents the posterior probability of the discharge probability d i of the terminal based on the initial discharge probability D total . P(d i ) represents the marginal likelihood function value of the discharge probability d i . P(D total ) represents the initial discharge probability of the terminal. P(D|D total ) represents the posterior discharge probability of the terminal. P(d i |D total ) represents the likelihood function value of the discharge probability d i .

8. A partial discharge location system for a power transmission tower, characterized in that, The partial discharge location of the power transmission tower includes an input device, a processor, a memory, and an output device; wherein, the input device, the processor, the memory, and the output device are interconnected; the memory is used to store a computer program, the computer program includes program instructions, and the processor is configured to call the program instructions to execute the partial discharge location method of the power transmission tower according to any one of claims 1-7.

Images