KR20230142330A - Package Switchboard degradation preventive system and a method of preventing degradation using the same - Google Patents

Abstract

The present invention relates to an integrated switchgear deterioration prevention system and, more specifically, to an integrated switchgear deterioration prevention system which adjusts white balance of a high-performance sensor image to solve the existing limitations of preventing deterioration with a thermal imaging image sensor of a switchgear including white-colored switchboards and components, thereby accurately predicting and preventing deterioration.

Description

Integrated switchboard degradation prevention system and deterioration prevention method using the same {Package Switchboard degradation preventive system and a method of preventing degradation using the same}

The present invention relates to an integrated switchgear deterioration prevention system. More specifically, in order to solve the conventional limitations of preventing deterioration with a thermal image sensor of a switchgear including white-colored switchboards and components, the white balance of the high-performance sensor image is provided. This relates to an integrated switchgear deterioration prevention system that can predict and prevent deterioration more accurately by adjusting it.

In general, a switchboard is a facility that allows consumers to safely use power produced and transformed at power plants, substations, etc., and is a facility that safely blocks and distributes power by installing transformers, circuit breakers, and instrument transformers. In particular, in industrial complexes with large-scale factories, the power transformed through high-low-voltage switchboards and extra-high-voltage switchboards is used as a power source for the production process system.

These switchboards are complicatedly installed with various electrical components and wiring, so there is a high risk of fire due to overcurrent or short circuit. If various risk factors inside the switchgear are detected before a large fire occurs, a manager is immediately dispatched. Appropriate measures must be taken.

Conventionally, in order to measure the temperature of switchgear or power equipment and the degree of deterioration due to aging, a separate temperature measuring device or thermal imaging camera is used to measure the temperature by directly projecting it to the area to be measured, and periodically to check the deterioration state. In most cases, the condition is estimated by measuring and inspecting the external condition, so there is a problem in preventing accidents due to the soundness and deterioration of the equipment of the switchgear.

In particular, in the case of switchboards containing white-colored switchboards and components, even when deterioration images are captured with a thermal imaging sensor, it is difficult to obtain a background color that is accurately distinguished from blue or red wiring, making accurate deterioration diagnosis difficult.

As related prior art, Registered Patent No. 10-2154854 (Switchboard Monitoring System Using Big Data and Artificial Intelligence) is installed on the switchboard to collect power information from the switchboard, and uses an IOT sensor to collect data inside the switchboard. A digital power information collection unit that collects data, a customer installation server that receives data stored in the digital power information collection unit, and a database that receives the data transmitted to the customer installation server and converts each customer's switchboard power information into a database by type, Receives the data input through the big data analysis server and the customer installation server, which derives a prediction model for abnormal signs of the switchgear for each type of DB power information, and the DB data transmitted from the big data analysis server, and based on this, the current A technology including an artificial intelligence server that self-analyzes the status of data transmitted to a consumer installation server is disclosed.

However, the information provided in big data is inevitably very limited, and in the process of updating big data, updates are made depending on measured values, making it difficult to increase accuracy.

KR 2021-0073426 KR 2021-0030724 KR 2020-0163322 JP 2020-076017 KR 2014-0031175 US 14-239945 KR 2009-0086972 KR 2007-0084622 KR 2007-0076475 EP 2001-102145

The present invention was made to solve the above problems, and the present invention provides an integrated distribution panel deterioration prevention system that can correct white balance by appropriately estimating the light source appearing in an image or video and suppressing color fading to prevent deterioration. The purpose is to do this.

In order to solve the above problem, the present invention includes a real image capturing step of capturing a real image corresponding to the closed space of the switchgear using a real image sensor 200; A thermal image capturing step of capturing a thermal image corresponding to the enclosed space of the switchgear using the thermal image sensor 100; A composite step of generating a composite image by extracting edges of the real image and synthesizing the thermal image to overlap the edges; A matching step of comparing and matching the real image with a plurality of images or images stored in a database and then designating an image or image with a shape closest to the real image as a problem area; a display step of externally displaying the composite image and displaying the problem area designated in the matching step as a name corresponding to the enclosed space of the switchgear; The database stores temperature distribution data for a plurality of problem area images or images, temperature distribution for the thermal image corresponding to the enclosed space of the switchboard, and temperature distribution for the problem area image or video specified in the matching step. a temperature comparison step of comparing temperature distribution data stored in the database; A warning step of generating an alarm when the temperature distribution of the thermal image corresponding to the enclosed space of the switchgear compared in the temperature comparison step is outside the allowable range of the temperature distribution data of the problem area image or video; Includes, and designation as a problem area in the matching step is based on the real image, determines and designates the area where ultraviolet wavelengths are emitted within the distribution system of the real image sensor as the problem area.

In the matching step, a color gain for adjusting the white balance is calculated for the YCbCr color space information of the image or video in the problem area, and the step of adjusting the white balance according to the calculated color gain is further included.

The present invention uses a real image sensor to compare and match the image corresponding to the enclosed space of the switchgear with a plurality of images or images stored in a database, and then selects the image or image with the closest shape to the actual image. A real image color space conversion unit 110 that specifies the occurrence area and converts the real image information into YCbCr color space information; A real image white balance adjustment unit 120 that calculates a color gain for adjusting the white balance for the YCbCr color space information converted by the real image color space conversion unit and adjusts the white balance according to the calculated color gain; A database in which the distance and illuminance value of the light source and the color temperature corresponding to the color gain are recorded, and the color temperature value corresponding to the distance and illuminance value of the light source input to the input unit and the color gain applied in the real image white balance adjustment unit are recorded in the database. It includes a real image color temperature detector 130 that calculates the color temperature using .

and a problem area unit that determines a problem area in which ultraviolet rays are emitted within the switchgear system, based on an image taken of the problem area.

Receives thermal images of the problem area from a thermal image sensor,

It further includes an image control module 160 that receives the corresponding signal change when the PIR sensor 150 detects minute movement of switchgear wires in an enclosed space.

The technology disclosed in this specification can have the following effects. However, since it does not mean that a specific embodiment must include all of the following effects or only the following effects, the scope of rights of the disclosed technology should not be understood as being limited thereby.

The present invention, achieved as described above, can distinguish between white color (ground color inside the switchgear) and wiring color by adjusting the white balance in advance, and can detect the color change of the flame due to the flame as quickly as possible.

In addition, the performance of white balance adjustment can be improved by accurately detecting the white area of the image containing white, which is the background color or component color of the enclosed space of the switchgear.

In addition, by using the Y, Cb and Cr data of the input image frame, auto white balance is performed using pixels included in Cb and Cr in different ranges according to the brightness component (luminance (Y) component), so that the input image frame is Image distortion can be minimized even if it is a single color or contains multiple pixels in a specific color area.

In addition, due to the difference in the physical separation distance between the thermal image sensor and the real image sensor, the distortion of the thermal image and the real image is automatically corrected, which has the effect of accurately determining the temperature of the target while recognizing the target the user is looking at. there is.

Additionally, there is a memory saving effect because images are captured only from problem areas.

In addition, the PIR sensor can detect minute movements of objects (parts) such as switchgear wires with minimal noise, improving the accuracy and precision of object recognition.

Even without prior data on the integrated switchgear, the deterioration state can be predicted step by step according to the condition scenario, and predictive information can be provided to managers to prepare in advance.

Figure 1 is a diagram showing the overall external appearance of an integrated switchgear deterioration prevention system according to an embodiment of the present invention.
Figure 2 is a diagram showing the internal parts, wiring, and background color of an integrated switchgear deterioration prevention system according to an embodiment of the present invention.
Figure 3 is a diagram showing various devices that can be installed in an integrated switchgear deterioration prevention system according to an embodiment of the present invention.
Figure 4 is a diagram showing modules related to a thermal image sensor according to another embodiment of the present invention.
Figure 5 is a diagram showing modules related to an ultraviolet image sensor according to another embodiment of the present invention.
Figure 6 is a diagram showing a thermal image sensor and a PIR sensor according to another embodiment of the present invention.
Figure 7 is a diagram showing an image control module and a monitoring unit according to another embodiment of the present invention.
Figure 8 is a diagram showing a look-up table, etc. according to another embodiment of the present invention.
Figure 9 is a diagram showing a configuration learned by a learning module according to another embodiment of the present invention.
Figure 10 is a diagram showing an analysis module, etc. according to another embodiment of the present invention.

In order to fully understand the present invention, preferred embodiments of the present invention will be described with reference to the accompanying drawings. Embodiments of the present invention may be modified in various forms, and the scope of the present invention should not be construed as being limited to the embodiments described in detail below. This example is provided to more completely explain the present invention to those with average knowledge in the art. Therefore, the shapes of elements in the drawings may be exaggerated to emphasize a clearer description. It should be noted that identical members in each drawing may be indicated by the same reference numerals. Additionally, detailed descriptions of known functions and configurations that are judged to unnecessarily obscure the gist of the present invention are omitted.

As shown in Figures 1 to 3, the interior of the integrated switchboard deterioration prevention system of the present invention is composed of a load switch (1), a VCB (vacuum circuit breaker) (2), a pull-out chassis (3), a bushing, etc.

As shown in FIG. 4, the image corresponding to the enclosed space of the switchgear is compared and matched with a plurality of images or videos stored in the database using a real image sensor, and then the image or video with the closest shape to the real image is matched. A real image color space conversion unit 110 that converts real image information into YCbCr color space information by designating an image or video as a problem area, and white on the YCbCr color space information converted by the real image color space converter. A real image white balance adjustment unit 120 that calculates a color gain for adjusting the balance and adjusts the white balance according to the calculated color gain, a database in which the distance and illuminance value of the light source and the color temperature corresponding to the color gain are recorded, and , It is equipped with a real image color temperature detection unit 130 that calculates the color temperature value corresponding to the distance and illuminance value of the light source input to the input unit and the color gain applied by the real image white balance adjustment unit using information recorded in the database.

As shown in FIG. 5, the ultraviolet image sensor 200 includes a problem area unit that determines a problem area in which ultraviolet wavelengths are emitted within the switchgear system, based on an image taken of the problem area, and the ultraviolet image After determining the problem area in the sensor through the problem area unit 210, the mode of the thermal imaging image sensor is switched from the deactivated mode to the activated mode.

The thermal image sensor 100 can capture a thermal image only in the problem area and store it in the memory 220.

As shown in FIG. 6, the thermal image sensor is implemented as an infrared temperature sensor and includes a PIR sensor 150 that transmits a signal change to the image control module 160 when detecting fine tremor in the enclosed space; Includes.

According to the high-sensitivity PIR sensor 150, minute movements of objects (parts) such as switchgear wires can be detected while noise is minimized, thereby improving the accuracy and precision of object recognition.

When a temperature change greater than a preset temperature occurs in a thermal image in which heat distribution according to brightness change and saturation change is displayed, the thermal imaging image sensor 100 displays the brightness change and saturation change differently at the corresponding location, and displays the thermal image at the corresponding location. is transmitted to the image control module.

As shown in FIG. 7, the image control module 160 receives an output waveform from the PIR sensor, and when a change in the output waveform of the PIR sensor exceeds a preset threshold, a fine movement occurs in the enclosed space. A monitoring unit 170 that determines that something has occurred; A thermal image is received from the thermal image sensor 100.

As shown in FIG. 8, a real image white balance adjustment unit that adjusts the white balance according to the calculated color gain, an input unit 122 capable of inputting the distance and illuminance value of the light source, the distance and illuminance value of the light source, and The look-up table 123 records the color temperature corresponding to the color gain, and the distance and illuminance value of the light source input to the input unit and the color temperature value corresponding to the color gain applied by the real image white balance adjuster 120 are recorded in the look-up table. It is provided with a color temperature detection unit 130 that calculates the color temperature using information.

As shown in Figure 9, when the brightness or saturation in the thermal image changes more than a preset threshold, a temperature monitoring unit 181 determines that a temperature change has occurred in the enclosed space; an image acquisition unit 182 that controls the photographing module according to the temperature monitoring unit and monitoring results from the temperature monitoring unit to obtain an image captured in the enclosed space; and temporarily stores the image received from the thermal image sensor 100, and receives it from the PIR sensor together with the output signal from the PIR sensor 150 and the thermal image sensor 100 at the time of the corresponding shooting. and an image storage unit 183 that transmits the image to the manager terminal 184, wherein the manager terminal 184 includes the captured image received from the image storage unit 183, the output signal of the PIR sensor, and the Thermal images from the thermal image sensor are used as learning data and stored in the database.

In addition, the learning module 185 defines each captured image, output signal, and thermal image as one segment, and constructs a deep learning model that learns the learning data by targeting a plurality of state scenarios of the enclosed space for each defined segment. , the captured image received from the image storage unit 183, the output signal of the PIR sensor, the thermal image of the thermal image sensor, and the deterioration state scenario that is the output of the deep learning model are output together.

As shown in FIG. 10, the PIR sensor 150, ultraviolet image sensor 200, and thermal image sensor 100 measure microdischarge and temperature generated from power devices inside the switchgear.

High-sensitivity PIR sensors detect minute movements of objects (parts) such as switchgear wires. Each sensor is provided in plural numbers inside the switchboard according to its focusing area.

Next, the A/D conversion module 191 converts the measurement results of each ultraviolet image sensor 200 and the thermal image sensor 100 into digital signals.

The analysis module 192 analyzes and processes the measurement results converted into digital signals to generate temperature distribution data.

The communication unit 193 receives measurement results from the A/D conversion module, and the notification unit 194 provides an alarm sound or message to the manager when the measurement result exceeds the preset temperature and at the same time it is confirmed whether discharge is detected. .

Subsequently, the analysis processing unit 195 generates temperature distribution data including analysis information (switchboard measurement amount, measurement temperature, and measurement time) according to the measurement results of each sensor and trend information of the degree of deterioration according to this analysis information, and generates temperature distribution data. Output data to administrator.

Hereinafter, the method or device for practicing the present invention will be described in detail.

The present invention uses a thermal imaging image sensor to compare and match the image corresponding to the closed space of the switchgear with a plurality of images or images stored in a database, and then compares and matches the image or image corresponding to the closed space of the switchgear with the image or image having the closest shape to the actual image. It includes a matching step that designates as the problem area.

In addition, a real image color space conversion unit 110 converts real image information into YCbCr color space information, and calculates a color gain for adjusting white balance for the YCbCr color space information converted by the real image color space conversion unit. .

Therefore, the performance of white balance adjustment can be improved by accurately detecting the white area of the image containing white, which is the background color or component color of the enclosed space of the switchgear.

In one embodiment, the image control module defines each captured image, output signal, thermal image, and white balance adjusted color temperature value as one segment, and targets and learns a plurality of state scenarios of the enclosed space for each defined segment. A deep learning model that has learned the data is built, and the captured image received from the image control module, the output signal of the PIR sensor, the thermal image of the thermal image sensor, and the state scenario that is the output of the deep learning model are output together.

Therefore, by using the Y, Cb and Cr data of the input image frame and performing auto white balance using pixels included in Cb and Cr in different ranges according to the brightness component (luminance (Y) component), the input image frame is Image distortion can be minimized even if it is a single color or contains multiple pixels in a specific color area.

Specifically, a real image capturing step of capturing a real image corresponding to the enclosed space of the switchgear using a thermal image sensor; A thermal image capturing step of capturing a thermal image corresponding to the enclosed space of the switchgear using a thermal image sensor; A composite step of generating a composite image by extracting edges of the real image and synthesizing the thermal image to overlap the edges; A matching step of comparing and matching the real image with a plurality of images or images stored in a database, and then designating the image or image with the closest shape to the real image as a problem area; a display step of externally displaying the composite image and displaying the problem area designated in the matching step as a name corresponding to the enclosed space of the switchgear; The database stores temperature distribution data for a number of problem area images or images, temperature distribution for the thermal image corresponding to the enclosed space of the switchgear, and temperature distribution for the problem area image or video specified in the matching step. a temperature comparison step of comparing temperature distribution data stored in the database; A warning step of generating an alarm when the temperature distribution of the thermal image corresponding to the enclosed space of the switchgear compared in the temperature comparison step is outside the allowable range of the problem area image or temperature distribution data of the image.

The ultraviolet image sensor can photograph problem areas, which are at least some areas of the switchgear system, in an active mode.

A real image capturing step of capturing a real image corresponding to the enclosed space of the switchgear using a thermal image sensor; A thermal image capturing step of capturing a thermal image corresponding to the enclosed space of the switchgear using a thermal image sensor; A composite step of generating a composite image by extracting edges of the real image and synthesizing the thermal image to overlap the edges; It includes a matching step of comparing and matching the real image with a plurality of images or images stored in a database, and then designating the image or image with the closest shape to the real image as the problem area.

Meanwhile, the thermal image sensor can capture thermal images only from the problem area and store them in memory.

Therefore, by automatically correcting the distortion of the thermal image and the real image due to the difference in the physical separation distance between the thermal image sensor and the real image sensor, the effect of accurately determining the temperature of the target while recognizing what the user is looking at is achieved. there is.

Additionally, there is a memory saving effect because images are captured only from problem areas.

In addition, the high-sensitivity PIR sensor 150 can detect minute movements of objects (parts) such as switchgear wires while minimizing noise, thereby improving the accuracy and precision of object recognition.

The thermal image sensor displays the thermal image with different brightness and saturation changes at the corresponding location when a temperature change greater than a preset temperature occurs in a thermal image that displays heat distribution according to brightness and saturation changes, and displays the thermal image as the image. Sent to the control module.

In one embodiment, the image data captured by the thermal image sensor 100 is separated into a luminance signal and a color signal and received as input data through the image control module 160; And, color correction and white balance by correcting the input data with a color correction conversion matrix generated by inputting the color signal for the predetermined color of the image data into a color correction conversion matrix expressed as a function of the color signal based on the predetermined color. It includes the step of outputting the final result video carried out together.

A real image capturing step of capturing a real image corresponding to the enclosed space of the switchgear using a thermal image sensor 100; A thermal image capturing step of capturing a thermal image corresponding to the enclosed space of the switchgear using a thermal image sensor; A real image color space conversion unit converts real image information into YCbCr color space information, and a color gain for adjusting white balance is calculated for the YCbCr color space information converted by the real image color space conversion unit.

In addition, the present invention can calculate color temperature using information obtained through experiment in advance and recorded in a look-up table, thereby reducing color temperature detection error.

The learning module 185 defines each captured image, output signal, and thermal image as one segment, and constructs a deep learning model that learns the learning data by targeting a plurality of state scenarios of the enclosed space for each defined segment, The captured image received from the image storage unit 183, the output signal of the PIR sensor, the thermal image of the thermal image sensor, and the deterioration state scenario that is the output of the deep learning model are output together.

Therefore, even without prior data on the integrated switchgear, the deterioration state can be predicted step by step according to the condition scenario, and predictive information can be provided to managers to prepare in advance.

After the matching step, a temperature comparison step, a warning step, and a supplementation step may be further included, which are arranged in parallel with the display step.

At this time, the database stores temperature distribution data for multiple images or videos.

The temperature comparison step is performed after the matching step, and compares the temperature distribution of the thermal image of the switchboard with temperature distribution data stored in the database for the target image or video specified in the matching step.

The warning step generates an alarm when the temperature distribution of the thermal image of the switchboard compared in the temperature comparison step is outside the allowable range of temperature distribution data of the target designation image or video.

The temperature distribution data of the image or video stored in the database includes a high temperature specific location that specifies the location where the highest temperature occurs.

In the temperature comparison step, the location where the highest temperature occurs in the thermal image of the switchboard is compared with the high temperature specific location in the target-designated image or video, and in the warning step, the location where the highest temperature occurs in the thermal image of the switchboard is compared. If the high temperature is not in a specific location in the target image or video, an alarm is generated.

Additionally, the temperature distribution data of the image or video stored in the database includes a low temperature specific location that specifies the location where the lowest temperature occurs.

In the temperature comparison step, the location where the lowest temperature occurs in the thermal image of the switchboard is compared with the low temperature specific location in the target-designated image or video, and in the warning step, the location where the lowest temperature occurs in the thermal image of the switchboard is compared. If it is not a low temperature specific location in the target designated image or video, an alarm is generated.

In the supplementary step, when an alarm occurs in the warning step and it is input that there is no problem through an external input, this is stored and the alarm is no longer generated in the future under the conditions that generated the alarm.

As described above, the present invention extracts the edge through an image sensor, composites the thermal image inside the edge, and displays it on the outside, allowing the user to clearly recognize the shape of the photographed switchboard and easily determine the heat distribution of each part. .

In addition, by comparing the temperature distribution of the photographed switchboard with the temperature distribution data stored in the database, it is possible to easily determine which part of the switchboard is overheated or short-circuited.

In addition, the present invention generates an external alarm when overheating beyond the allowable range begins to occur through the temperature distribution of the distribution board being photographed, so that a fire, etc. can be notified in advance before it occurs, allowing for proactive response and prevention of major accidents. there is.

Additionally, the image control module includes a shading estimation unit that estimates the shading intensity of the image, a light source color estimation unit that stores in advance the achromatic judgment range for each shading intensity, and sets the achromatic judgment range corresponding to the estimated shading intensity to a color space; , a correction coefficient calculation unit that calculates the color coordinates of the pixels in the image and calculates the white balance gain based on the pixel value of the pixel whose color coordinates are included in the set achromatic judgment range, and uses the calculated white balance gain to calculate the color coordinates of the image. A correction unit for correcting white balance is provided.

Therefore, the present invention can correct white balance by appropriately estimating the light source and suppressing color fading.

100: thermal image sensor
110: Real image color space conversion unit
120: Real image white balance adjustment unit
122: input unit
123: Lookup table
130: Real image color temperature detection unit
150: PIR sensor
160: Image control module
170: monitoring unit
181: Temperature monitoring unit
182: Image acquisition unit
184: Administrator terminal
183: video storage unit
185: Learning module
191: A/D conversion module
192: Analysis module
193: Department of Communications
194: Notification unit
195: Analysis processing department
200: ultraviolet image sensor
210: Problem area
220: memory

Claims (5)

A real image capturing step of capturing a real image corresponding to the enclosed space of the switchgear using a real image image sensor 200; A thermal image capturing step of capturing a thermal image corresponding to the enclosed space of the switchgear using the thermal image sensor 100; A composite step of generating a composite image by extracting edges of the real image and synthesizing the thermal image to overlap the edges; A matching step of comparing and matching the real image with a plurality of images or images stored in a database and then designating an image or image with a shape closest to the real image as a problem area; a display step of externally displaying the composite image and displaying the problem area designated in the matching step as a name corresponding to the enclosed space of the switchgear; The database stores temperature distribution data for a plurality of problem area images or images, temperature distribution for the thermal image corresponding to the enclosed space of the switchboard, and temperature distribution for the problem area image or video specified in the matching step. a temperature comparison step of comparing temperature distribution data stored in the database; A warning step of generating an alarm when the temperature distribution of the thermal image corresponding to the enclosed space of the switchgear compared in the temperature comparison step is outside the allowable range of the temperature distribution data of the problem area image or video; Includes,
Designation as a problem area in the matching step is,
A deterioration prevention method using an integrated switchgear deterioration prevention system, characterized in that, based on the real image, the area where ultraviolet wavelengths are emitted within the switchboard system of the real image sensor is determined and designated as a problem area. In claim 1,
In the matching step,
An integrated switchgear deterioration prevention system further comprising the step of calculating a color gain for adjusting the white balance for the YCbCr color space information of the image or video in the problem area, and adjusting the white balance according to the calculated color gain. Deterioration prevention method using .
Using a real image sensor, the image corresponding to the enclosed space of the switchgear is compared and matched with a plurality of images or videos stored in the database, and then the image or video with the closest shape to the actual image is selected as the problem area. specify,
A real image color space conversion unit 110 that converts real image information into YCbCr color space information;
Calculate the color gain for adjusting the white balance for the YCbCr color space information converted by the real image color space conversion unit,
A real image white balance adjustment unit 120 that adjusts the white balance according to the calculated color gain;
A database in which the distance and illuminance value of the light source and the color temperature corresponding to the color gain are recorded, and the color temperature value corresponding to the distance and illuminance value of the light source input to the input unit and the color gain applied in the real image white balance adjustment unit are recorded in the database. An integrated switchgear deterioration prevention system comprising a real image color temperature detection unit 130 that calculates using .
In claim 3,
An integrated switchgear deterioration prevention system comprising a problem area unit that determines a problem area in which ultraviolet rays are emitted within the switchgear system, based on images taken of the problem area.
In claim 4
Receives thermal images of the problem area from a thermal image sensor,
An integrated switchgear deterioration prevention system further comprising: an image control module 160 that receives the corresponding signal change when the PIR sensor 150 detects minute movements of switchgear wires in an enclosed space.