KR20210153006A - Apparatus for detecting corona discharge - Google Patents

Abstract

Disclosed is a device for detecting corona discharge, which comprises: an operation switch for operating a detection device; an image measurement unit for photographing a power facility through visible light and ultraviolet light; an environment measurement unit for measuring a measurement environment of the image measurement unit; a control unit for storing a photographed image photographed by the image measurement unit and an environment measurement value measured by the environment measurement unit in accordance with an operation of the operation switch, identifying the power facility from a visible light image for each frame for the photographed image through image analysis, accumulating an ultraviolet image for each frame to generate an accumulated ultraviolet image to synthesize the same with the visible light image, and storing integrated data matching the environment measurement value to analyze and output a corona discharge state; and a storage unit for storing the photographed image and the environment measurement value measured by the image measurement unit and the environment measurement unit, and storing a synthesized image and the integrated data generated by the control unit. Therefore, determination reliability can be improved.

Description

Corona discharge detection device {APPARATUS FOR DETECTING CORONA DISCHARGE}

The present invention relates to a corona discharge detection device, and more particularly, by recognizing power equipment through an image of a camera that shoots a visible light region and an ultraviolet region, labeling each frame, forming an ultraviolet cumulative image, and analyzing the image. It relates to a corona discharge detection device capable of quantitatively detecting a corona discharge.

In general, when a defect occurs in a high voltage power facility for power, a corona is generated by partial discharge in the defective part, and if this corona discharge continues, light emission, audible noise, communication disorder, ozone, etc. Power equipment leads to insulation breakdown. In particular, ozone is the biggest factor that accelerates corrosion of power equipment.

As a conventional technology for detecting partial discharge, a method using a sensor capable of detecting vibration, ultrasonic waves, various wavelengths, heat, etc. has been proposed to detect and analyze various phenomena appearing in omens.

Recently, during the diagnosis of overhead transmission lines, deterioration evaluation of overhead transmission lines, insulators, and connection facilities is being performed through weekly corona measurement.

Background art of the present invention is disclosed in Korean Patent Application Laid-Open No. 10-2012-0016327 (published on February 24, 2012, corona discharge image detection method and system).

As such, the daytime corona inspection of overhead power transmission lines is an indirect measurement method that detects by image processing a part of the UV (ultraviolet) band among the electromagnetic wave bands emitted by the corona using an ultraviolet camera, and is effective as a detection method in the live state.

However, there is no quantitative standard to determine whether or not there is a defect, so the knowledge of an expert to determine whether or not there is a defect by examining the discharge pattern is required, and there is a problem in that it is difficult to provide an immediate judgment in the field.

In addition, the weekly corona inspection equipment through the UV camera can be divided into portable equipment that the measurer takes directly, and equipment that can be mounted on transportation means such as cars, trains, helicopters, and drones to take pictures.

In particular, the UV camera mounted on the transportation means can determine the state of the facility through analysis by an expert after acquiring the weekly corona image.

However, there is a problem in that it is difficult for a measurer to read an image measured in a high-speed transportation means in real time, and it is difficult to make an accurate judgment due to noise in the measured image.

In addition, in the case of overhead power transmission lines, it is difficult to take close-up photos using an ultraviolet camera because they are located in places that are difficult to access by vehicles and manpower, such as mountainous areas and flat land, rather than downtown. When it occurs, it is difficult to trace the location where the discharge occurs. In addition, when using aerial photography such as drones rather than manpower measurement, there is a problem in that it is difficult for the measurer to judge the image in the field.

The present invention has been devised to improve the above problems, and an object of the present invention according to one aspect is to recognize power equipment through an image of a camera that shoots visible and ultraviolet regions, label each frame, and then accumulate ultraviolet rays To provide a corona discharge detection device capable of quantitatively detecting a weekly corona discharge by forming an image and analyzing the image.

A corona discharge detection device according to an aspect of the present invention includes an operation switch for operating the detection device; an image measurement unit for photographing power equipment through visible light and ultraviolet light; an environment measurement unit for measuring the measurement environment of the image measurement unit; According to the operation of the operation switch, the captured image captured by the image measurement unit and the environmental measurement value measured through the environment measurement unit are stored, and power equipment is identified from the visible light image frame by frame for the captured image through image analysis, and each frame a controller for accumulating ultraviolet images to generate an accumulated ultraviolet image, synthesizing it with a visible light image, storing integrated data matching environmental measurement values, and analyzing and outputting the corona discharge state; and a storage unit that stores the captured images and environmental measurement values measured by the image measurement unit and the environment measurement unit, and stores the synthesized image and integrated data generated by the control unit.

In the present invention, the image measurement unit, an optical camera for taking a visible light image; and an ultraviolet camera for capturing an ultraviolet image.

In the present invention, the environment measuring unit includes: a position detecting unit for detecting the measuring position of the image measuring unit; a gyro sensor unit for detecting a measurement direction of the image measurement unit; and a distance detection unit for measuring the distance to the power equipment measured by the image measurement unit; It is characterized in that it includes any one or more.

In the present invention, the control unit is characterized in that the noise is removed by removing pixels whose frequency of corona discharge is less than or equal to a set number of times for the accumulated UV image.

In the present invention, the control unit is characterized in that the noise is removed by removing less than a noise threshold for an ultraviolet image whose measurement distance is within a set distance from the environmental measurement value.

In the present invention, the control unit is characterized in that it calculates the UV-generating point through the number of clusters by substituting the measurement distance and the noise threshold into the DBSCAN (Density-Based Spatial Clustering of Applications with Noise) algorithm.

A corona discharge detection method according to another aspect of the present invention comprises the steps of: receiving, by a control unit, a photographed image captured by an image measurement unit according to an operation of an operation switch and an environmental measurement value measured by an environment measurement unit and storing the received image in a storage unit; The control unit calling the visible light image stored in the storage unit to identify the power facility; generating, by the controller, the ultraviolet image stored in the storage unit and accumulating the ultraviolet image for each frame to generate an accumulated ultraviolet image; storing, by the control unit, integrated data obtained by synthesizing the visible light image and the ultraviolet image, and matching the environmental measurement value and the identification result of the power facility; and analyzing and outputting, by the controller, the corona discharge state based on the stored integrated data.

In the present invention, the photographed image is a visible light image photographed through an optical camera; and an ultraviolet image taken through an ultraviolet camera.

In the present invention, the environmental measurement value is characterized in that it includes any one or more of a measurement position of the image measurement unit, a measurement direction of the image measurement unit, and a measurement distance with a power facility measured by the image measurement unit.

In the present invention, the generating of the accumulated ultraviolet image may further include, by the controller, removing the noise by removing pixels whose frequency of corona discharge is less than or equal to a set number of times for the accumulated ultraviolet image.

In the present invention, the step of generating the accumulated ultraviolet image may further include, by the controller, removing the noise by removing less than the noise threshold from the ultraviolet image whose measurement distance is within a set distance from the environmental measurement value.

In the present invention, the step of generating the accumulated UV image is a step of calculating, by the control unit, the location of UV generation through the number of clustered clusters by substituting the measurement distance and the noise threshold into the DBSCAN (Density-Based Spatial Clustering of Applications with Noise) algorithm. It is characterized in that it further includes;

Corona discharge detection device according to an aspect of the present invention recognizes power equipment through the image of a camera that shoots visible light region and ultraviolet region, labels for each frame, forms an ultraviolet accumulated image, and analyzes the image, so that the weekly corona discharge can be quantitatively detected, so the inspection task can be simplified and the judgment reliability can be improved.

1 is a block diagram illustrating a corona discharge detection apparatus according to an embodiment of the present invention.
2 is an exemplary view showing the accumulation state of ultraviolet images in the corona discharge detection apparatus according to an embodiment of the present invention.
3 is an exemplary view illustrating a noise removal state of an accumulated UV image in the corona discharge detection apparatus according to an embodiment of the present invention.
4 is an exemplary diagram illustrating an output state of integrated data in the corona discharge detection apparatus according to an embodiment of the present invention.
5 is a flowchart illustrating a corona discharge detection method according to an embodiment of the present invention.

Hereinafter, a corona discharge detecting apparatus and method according to the present invention will be described with reference to the accompanying drawings. In this process, the thickness of the lines or the size of the components shown in the drawings may be exaggerated for clarity and convenience of explanation. In addition, the terms to be described later are terms defined in consideration of functions in the present invention, which may vary according to the intention or custom of the user or operator. Therefore, definitions of these terms should be made based on the content throughout this specification.

1 is a block diagram showing a corona discharge detection apparatus according to an embodiment of the present invention, and FIG. 2 is an exemplary view showing the accumulation state of ultraviolet images in the corona discharge detection apparatus according to an embodiment of the present invention, FIG. 3 is an exemplary view showing the noise removal state of the ultraviolet accumulated image in the corona discharge detection apparatus according to an embodiment of the present invention, and FIG. 4 is an output state of the integrated data in the corona discharge detection apparatus according to an embodiment of the present invention. It is an example diagram shown.

As shown in FIG. 1 , the corona discharge detection apparatus according to an embodiment of the present invention includes an operation switch 10 , an image measurement unit 20 , an environment measurement unit 30 , a control unit 40 , and a storage unit 50 . ) may be included.

The operation switch 10 is a switch for allowing the user to operate the detection device to detect corona discharge of the power equipment.

The image measurement unit 20 may include an optical camera 210 for photographing a visible ray image and an ultraviolet camera 220 for photographing an ultraviolet image in order to photograph the power equipment through visible light and ultraviolet light.

The environment measurement unit 30 includes a position detection unit 310 that detects the measurement position of the image measurement unit 20 based on GPS to measure the measurement environment of the image measurement unit 20 , and a measurement direction of the image measurement unit 20 . It may include any one or more of a gyro sensor unit 320 that detects , and a distance detection unit 330 that measures a distance to a power facility measured by the image measurement unit 20 based on a non-contact laser sensor.

That is, UV rays have a characteristic that their brightness (illuminance) decreases in inverse proportion to the square of the distance as the distance from the light source increases. provided so that

The control unit 40 stores the captured image captured by the image measurement unit 20 and the environment measurement value measured through the environment measurement unit 30 according to the operation of the operation switch 10 , and the captured image through image analysis For each frame, the power equipment is identified from the visible light image, and the ultraviolet image is accumulated for each frame to create an accumulated ultraviolet image, and then synthesized with the visible light image. can be analyzed and output.

The control unit 40 stores in the storage unit 50 a real image captured image that is a visible light image captured through the optical camera 210 of the image measurement unit 20 and an ultraviolet light captured image captured by the ultraviolet camera 220 . do.

Afterwards, for image analysis, first, the real image captured image stored in the storage unit 50 is read, divided into frames, labeling is processed for each frame, and a convolutional neural network module (CNN: Convolution Neural Network) in each frame is performed. Recognizes power equipment from labeled frames using

In this case, the control unit 40 may recognize by correcting the shake and resolution of each frame when recognizing the power equipment, but may discard the frame if the correction is performed more than a set number of times.

In addition, the control unit 40 reads the ultraviolet photographed image, divides it into frames, and processes labeling for each frame of each image. Here, assuming that the resolution of the image is 800 × 600, the pixels in which corona occurred and pixels in which corona occurred can be expressed as shown in FIG. It can be seen that many corona discharges occur.

The controller 40 may perform this process for the mode frame to generate an accumulated UV image, and separately store coordinate values for UV pixel data detected in the image of each label in order to calculate the UV generating location.

In addition, the control unit 40 assumes that data having a numerical value less than or equal to a certain numerical value for the accumulated ultraviolet image is external noise, and removes ultraviolet noise that may occur during daytime measurement by removing this noise, as shown in FIG. 3 . Likewise, only ultraviolet data of the measurement target can be acquired.

For example, if it is assumed that the corona discharge that occurs less than 50 times in the accumulated ultraviolet image is noise and the value of the corresponding pixel is replaced with 0 and removed, the accumulated ultraviolet light image as shown in FIG. 2 is shown in FIG. As shown, an intrinsic corona discharge signal can be obtained by removing noise.

On the other hand, since the ultraviolet camera 210 of the image measurement unit 20 is inaccurate when the measurement distance is more than usually 100 [m], the controller 40 reflects the measurement distance as an environmental measurement value to 100 [m] For image data within a distance, noise may be removed by calculating a threshold value of about 10% of the total frame labeling information and then subtracting it from the accumulated UV image. In this case, the data exceeding the measurement distance can be classified separately without deducting it.

In addition, the control unit 40 uses the accumulated UV pixel data stored coordinate values while generating the UV accumulated image, and substitutes the measurement distance and the noise threshold to the DBSCAN (Density-Based Spatial Clustering of Applications with Noise) algorithm to determine the number of clustered clusters. Through this, it is possible to calculate the location where the ultraviolet light is generated.

Even though the UV emitted from the power equipment has the same amount of radiation, if the measurement distance is farther away, the UV pixel data is clustered into one point. ), and after clustering, the number of clusters from which noise has been removed can be calculated.

The controller 40 may generate an accumulated ultraviolet ray image, synthesize it with a visible ray image, store integrated data matching the environmental measurement value, analyze the corona discharge state, and output it through the output unit 60 .

As shown in FIG. 4 , the control unit 40 displays the measurement position, direction, and distance based on the environmental measurement value of the integrated data through the output unit 60 , and displays brief information about the analyzed detection point. When you click on an area, the stored UV accumulated image and visible light image are displayed.

The storage unit 50 may store the captured images and environmental measurement values measured by the image measurement unit 20 and the environment measurement unit 30 , and store the synthesized image and the integrated data generated by the control unit 40 .

As described above, according to the corona discharge detection device according to the embodiment of the present invention, the power equipment is recognized through the image of the camera photographing the visible light region and the ultraviolet region, and after labeling each frame, an ultraviolet accumulated image is formed to form an image By analyzing the weekly corona discharge, it is possible to quantitatively detect the weekly corona discharge, thereby simplifying the inspection task and improving the determination reliability.

5 is a flowchart illustrating a corona discharge detection method according to an embodiment of the present invention.

As shown in FIG. 5 , in the corona discharge detection method according to an embodiment of the present invention, first, the control unit 40 determines the operating state of the operation switch 10 for operating the corona discharge detection device ( S10 ).

When the operation switch 10 is operated in step S10, the control unit 40 receives the captured image captured by the image measurement unit 20 and the environmental measurement value measured through the environment measurement unit 30, and the storage unit ( 50) (S20).

Here, the image measurement unit 20 may include an optical camera 210 for photographing a visible ray image and an ultraviolet camera 220 for photographing an ultraviolet image in order to photograph the power equipment through visible light and ultraviolet light.

In addition, the environment measurement unit 30 includes a position detection unit 310 that detects the measurement position of the image measurement unit 20 based on GPS, in order to measure the measurement environment of the image measurement unit 20 , of the image measurement unit 20 . It may include any one or more of a gyro sensor unit 320 for detecting a measurement direction and a distance detection unit 330 for measuring a distance to a power facility measured by the image measurement unit 20 based on a non-contact laser sensor.

After receiving and storing the captured image and the environmental measurement value in step S20, the control unit 40 calls the visible light image stored in the storage unit 50 to identify the power equipment (S30).

The control unit 40 reads the real image captured image stored in the storage unit 50, divides it into frames, processes labeling for each frame, and then operates a convolutional neural network module (CNN) in each frame. It recognizes the power equipment from the labeled frames using it and accumulates the frames to create an image.

In this case, the control unit 40 may recognize by correcting the shake and resolution of each frame when recognizing the power equipment, but may discard the frame if the correction is performed more than a set number of times.

After identifying the power equipment in step S30, the control unit 40 calls the ultraviolet image stored in the storage unit 50 and accumulates the ultraviolet image for each frame to generate the accumulated ultraviolet image (S40).

That is, the controller 40 reads the ultraviolet-ray photographed image, divides it into frames, and processes the labeling for each frame of each image. Here, assuming that the resolution of the image is 800 × 600, the pixels in which corona occurred and pixels in which corona occurred can be expressed as shown in FIG. It can be seen that many corona discharges occur.

Accordingly, the controller 40 may perform this process for the mode frame to generate an accumulated UV image, and separately store coordinate values for UV pixel data detected in the image of each label in order to calculate the UV generation location.

In addition, the control unit 40 assumes that data having a numerical value less than or equal to a certain numerical value for the accumulated ultraviolet image is external noise, and removes ultraviolet noise that may occur during daytime measurement by removing this noise, as shown in FIG. 3 . Likewise, only ultraviolet data of the measurement target can be acquired.

For example, if it is assumed that the corona discharge that occurs less than 50 times in the accumulated ultraviolet image is noise and the value of the corresponding pixel is replaced with 0 and removed, the accumulated ultraviolet light image as shown in FIG. 2 is shown in FIG. As shown, an intrinsic corona discharge signal can be obtained by removing noise.

On the other hand, the control unit 40 reflects the measurement distance as an environmental measurement value, calculates a threshold value through about 10% of the total frame labeling information for image data within a distance of 100 [m], and then subtracts the noise from the accumulated UV image can be removed. In this case, the data exceeding the measurement distance can be classified separately without deducting it.

In addition, the control unit 40 uses the accumulated UV pixel data stored coordinate values while generating the UV accumulated image, and substitutes the measurement distance and the noise threshold to the DBSCAN (Density-Based Spatial Clustering of Applications with Noise) algorithm to determine the number of clustered clusters. Through this, it is possible to calculate the location where the ultraviolet light is generated.

Even though the UV emitted from the power equipment has the same amount of radiation, if the measurement distance is farther away, the UV pixel data is clustered into one point. ), and after clustering, the number of clusters from which noise has been removed can be calculated.

After generating the accumulated UV image in step S40, the control unit 40 synthesizes the visible light image and the UV image (S50), and creates and stores integrated data that matches the environmental measurement value and the identification result of the power facility with the synthesized image do (S60).

After storing the integrated data in step S60, the control unit 40 displays the measurement position, direction and distance based on the environmental measurement value through the output unit 60 through the output unit 60 as shown in FIG. 4, and the analyzed detection When brief information about a location is displayed and the corresponding location is clicked, the stored UV accumulation image and visible light image are displayed to output a detection result (S70).

As described above, according to the corona discharge detection method according to the embodiment of the present invention, the power equipment is recognized through the image of the camera photographing the visible light region and the ultraviolet region, and after labeling each frame, an ultraviolet accumulated image is formed to form an image By analyzing the weekly corona discharge, it is possible to quantitatively detect the weekly corona discharge, thereby simplifying the inspection task and improving the determination reliability.

Implementations described herein may be implemented in, for example, a method or process, an apparatus, a software program, a data stream, or a signal. Although discussed only in the context of a single form of implementation (eg, discussed only as a method), implementations of the discussed features may also be implemented in other forms (eg, as an apparatus or program). The apparatus may be implemented in suitable hardware, software and firmware, and the like. A method may be implemented in an apparatus such as, for example, a processor, which generally refers to a computer, a microprocessor, a processing device, including an integrated circuit or programmable logic device, or the like. Processors also include communication devices such as computers, cell phones, portable/personal digital assistants (“PDAs”) and other devices that facilitate communication of information between end-users.

Although the present invention has been described with reference to the embodiment shown in the drawings, this is merely an example, and those skilled in the art to which various modifications and equivalent other embodiments are possible. will understand

Therefore, the true technical protection scope of the present invention should be defined by the following claims.

10: operation switch 20: image measurement unit
30: environmental measurement unit 40: control unit
50: storage unit 60: output unit
210: optical camera 220: ultraviolet camera
310: position detection unit 320: gyro sensor unit
330: distance detection unit

Claims (1)

an operation switch for operating the detection device;
an image measurement unit for photographing power equipment through visible light and ultraviolet light;
an environment measurement unit for measuring the measurement environment of the image measurement unit;
According to the operation of the operation switch, the captured image captured by the image measurement unit and the environmental measurement value measured through the environment measurement unit are stored, and the power equipment is removed from the visible light image frame by frame for the captured image through image analysis. After identifying, accumulating ultraviolet images for each frame to create an accumulated ultraviolet image, synthesizing it with a visible light image, storing the integrated data matching the environmental measurement value and the identification result of the power facility, and analyzing the corona discharge state a control unit to output; and
A storage unit for storing the photographed image and the environmental measurement value measured by the image measurement unit and the environment measurement unit, and for storing the synthesized image and the integrated data generated by the control unit; including,
Further comprising an output unit for outputting a result of analyzing the corona discharge state based on the integrated data stored in the control unit,
The control unit reads the visible ray image stored in the storage unit, divides it into frames, processes labeling for each frame, and uses a convolutional neural network module (CNN) in each frame. The power equipment is recognized from the frame, and the noise is removed by removing less than the noise threshold for the ultraviolet image whose measurement distance is within the set distance from the environmental measurement value, and the measurement is performed using the DBSCAN (Density-Based Spatial Clustering of Applications with Noise) algorithm. By substituting the distance and the noise threshold, the location of UV generation is calculated through the number of clustered clusters,
The environment measuring unit includes a position detecting unit detecting a measurement position of the image measuring unit, a gyro sensor detecting a measuring direction of the image measuring unit, and a distance detecting unit measuring a distance to the power equipment measured by the image measuring unit and
The image measurement unit may include: an optical camera for capturing the visible light image; and
Including; an ultraviolet camera for taking the ultraviolet image;
The control unit, corona discharge detection device, characterized in that to remove the noise by removing the pixels having a frequency of corona discharge occurring in the accumulated ultraviolet rays image is less than or equal to a set number of times.

Images