KR20140076800A

KR20140076800A - Diagnosing disorder system for new renewable energy generator and method thereof

Info

Publication number: KR20140076800A
Application number: KR1020120145247A
Authority: KR
Inventors: 김성호; 문대선
Original assignee: 주식회사 가온솔루션
Priority date: 2012-12-13
Filing date: 2012-12-13
Publication date: 2014-06-23

Abstract

The present invention relates to a system and a method for diagnosing a malfunction using an infrared camera in a new renewable energy generator. The present invention includes: a camera unit for photographing a thermal image; and a control unit which converts the thermal image photographed by the camera unit into a binary image, controls the camera unit in order to perform a focusing until an edge image has a maximum white pixel after converting the binary image into the edge image again, detects a region of interest (ROI) for an observation by using the thermal image photographed by the camera unit and a pre-stored template, and enables the camera unit to diagnose a malfunction of the ROI according to temperatures which are detected in the thermal image which photographs the ROI.

Description

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a fault diagnosis system for a renewable energy generator,

The present invention relates to a fault diagnosis technique using an infrared camera, and more particularly, to a fault diagnosis technique using an infrared camera, in which a thermal image is photographed using an infrared camera to diagnose a failure of a renewable energy generator such as sunlight, wind power, To a system capable of diagnosing a fault of a wind power generator through an image and a method therefor.

Renewable energy refers to the energy used to transform existing fossil fuels or to convert and use renewable energy including sunlight, water, geothermal, and biological organisms. These new and renewable energy features future energy sources for a sustainable energy supply system. New and renewable energy has become increasingly important due to unstable oil prices and regulatory compliance with the Convention on Climate Change. In Korea, there are eight renewable energy sources (solar, photovoltaic, biomass, wind, hydro, geothermal, marine and waste energy), three new energy sources (fuel cell, coal liquefied gasification, hydrogen energy) 11 sectors are designated as renewable energy.

Among renewable energy, photovoltaic generators are composed of solar cells, accumulators, and power conversion devices. When sunlight is attached to a solar cell that is a junction of a p-type semiconductor and an n-type semiconductor, holes and electrons are generated in the solar cell by the energy of the sunlight. At this time, the holes are collected toward the p-type semiconductor and the electrons are collected toward the n-type semiconductor, so that a current flows when a potential difference occurs. The advantage of photovoltaic power generation is that there is no pollution, it can be developed only where it is needed, and it is easy to maintain. On the other hand, there is a disadvantage that the amount of electric power depends on the amount of sunshine, the installation site is limited, the initial investment cost and the power generation cost are high. Therefore, there is a need for a technique for tracking the position of the sun to receive more sunlight, and for this reason, there is also a need for a technique for tracking the movement of the solar generator and diagnosing its failure.

On the other hand, a large wind turbine includes a blade, a main shaft, a gear box, a generator, a power converter, a transformer, and the like. In addition, a pitch controller, a yaw controller and various cooling systems are used. In order to standardize the CMS (Condition Monitoring System) related to the wind power generation system, the European Commission has established a consultation body called Novel Integration Condition Monitoring (NIMO) to standardize the CMS. The causes of failure of each element are frequent occurrence in the order of gearbox, generator, main bearing and blade. CMS based on current vibration sensor and current sensor is effective for diagnosis of core parts. However, the failure detection of cooling systems related to pitch controllers, yaw controllers, generators and transformers mainly uses temperature sensors. If the failure of the cooling system is not recognized due to the failure or malfunction of the temperature sensor, the failure can be spread to the power plant of the wind power generator. Therefore, there is a demand for a more improved fault diagnosis system.

It is an object of the present invention, taken in view of the above points, to continuously detect a specific part of a generator for renewable energy, that is, an area of interest by using an infrared camera, And a method for the same.

It is still another object of the present invention to provide a system and method for performing more accurate fault diagnosis by performing focusing of an infrared camera to perform accurate tracking on a region of interest.

According to an aspect of the present invention, there is provided a fault diagnosis system using an infrared camera in a renewable energy generator, the fault diagnosis system comprising: a camera unit for photographing a thermal image; And a step of comparing the converted edge image with a template storing an edge of a region of interest (ROI) to track the ROI, And a controller for measuring the temperature of the tracked ROI from the thermal image of the ROI to diagnose the failure.

The controller may measure the temperature of the tracked ROI from the thermal image of the tracked ROI and diagnose the failure as a temperature change over a predetermined period of time according to the ambient environment of the accumulated ROI .

The control unit converts the thermal image into an edge image and performs focusing by moving the lens of the camera unit so that the number of white pixels obtained from the edge image is maximized.

Wherein the control unit moves the lens of the camera unit by a set distance in either one of forward and backward directions until the number of white pixels of the edge image reaches a maximum, If the number of white pixels of the edge image at the position after the movement is less than the number of pixels of the edge image at the position before the movement in the one direction, The control unit controls the lens to move in the direction of the optical axis.

Wherein the control unit determines the number of white pixels of the edge image at the post-movement position to be greater than the number of pixels of the edge image at the pre- The moving distance is changed so as to be inversely proportional to the rate of change of the number of pixels of white pixels of the edge image so that the lens is continuously moved in one direction.

According to another aspect of the present invention, there is provided a fault diagnosis method using an infrared camera in a renewable energy generator, the method comprising: capturing a thermal image; Comparing the converted edge image with a template storing an edge of a region of interest (ROI), and tracking the ROI, after converting the thermal image into an edge-only edge image; And measuring the temperature of the tracked ROI from the thermal image of the tracked ROI to diagnose the failure.

Wherein the diagnosing comprises measuring the temperature of the tracked ROI from the thermal image of the ROI being tracked and diagnosing the range of temperature change according to the ambient environment of the accumulated ROI, .

Converting the thermal image into an edge image and performing focusing by moving the lens of the camera unit so that the number of white pixels obtained from the edge image is maximized before the tracking step.

Wherein the focusing is performed by shifting the lens of the camera unit by a set distance in either one of forward and backward directions until the number of white pixels of the edge image reaches a maximum, Wherein if the number of pixels is greater than or equal to the number of pixels of the edge image at the pre-movement position, the lens is continuously moved in the one direction, and if the number of white pixels of the edge image at the post- And the lens is moved in a direction opposite to the one direction.

Wherein the step of performing focusing comprises: if the number of white pixels of the edge image at the post-movement position is equal to or greater than the number of pixels of the edge image at the pre-movement position, The moving distance is changed so as to be in inverse proportion to the rate of change of the number of pixels of white pixels of the edge image at the post-moving position, and the lens is continuously moved in the one direction.

According to the present invention, by continuously measuring the temperature of a generator for renewable energy using an infrared camera, it is possible to diagnose faults stably compared to a conventional simple temperature sensor.

In addition, more accurate tracking of a region of interest can be achieved by focusing focusing on a case where the number of white pixels is the maximum in an edge image in which edges of an object are detected. Therefore, the reliability of the fault diagnosis is improved.

Particularly, by tracking an area of interest through a template using only an edge of a region of interest (ROI), which is a part for diagnosing a failure, which is photographed by an infrared camera, compared with the conventional tracking method using RGB , Illumination, light, etc. can be provided.

1 is a diagram for explaining a schematic configuration of a fault diagnosis system for a wind turbine according to an embodiment of the present invention;
2 is a block diagram illustrating a structure of a fault diagnosis apparatus according to an embodiment of the present invention;
3 is a flowchart illustrating a focusing method according to an embodiment of the present invention;
4 and 5 are graphs for explaining a focusing method according to an embodiment of the present invention;
FIGS. 6 and 7 are views illustrating a focusing operation according to an embodiment of the present invention; FIG.
FIG. 8 is a flowchart illustrating a method of setting a region of interest according to an embodiment of the present invention; FIG.
9 is a diagram illustrating a method of setting a region of interest according to an embodiment of the present invention;
FIG. 10 is a flowchart illustrating a fault diagnosis method using a thermal image according to an embodiment of the present invention; FIG. And,
11 is a view illustrating a tracking method for performing a fault diagnosis according to an embodiment of the present invention.

Prior to the detailed description of the present invention, the terms or words used in the present specification and claims should not be construed as limited to ordinary or preliminary meaning, and the inventor may designate his own invention in the best way It should be construed in accordance with the technical idea of the present invention based on the principle that it can be appropriately defined as a concept of a term to describe it. Therefore, the embodiments described in the present specification and the configurations shown in the drawings are merely the most preferred embodiments of the present invention, and are not intended to represent all of the technical ideas of the present invention. Therefore, various equivalents It should be understood that water and variations may be present.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. Note that, in the drawings, the same components are denoted by the same reference symbols as possible. Further, the detailed description of known functions and configurations that may obscure the gist of the present invention will be omitted. For the same reason, some of the elements in the accompanying drawings are exaggerated, omitted, or schematically shown, and the size of each element does not entirely reflect the actual size.

1 is a diagram for explaining a schematic configuration of a fault diagnosis system for a wind turbine according to an embodiment of the present invention.

Referring to FIG. 1, a fault diagnosis system according to an embodiment of the present invention includes a fault diagnosis apparatus 100, a fault diagnosis server 200, and a user terminal 300.

The fault diagnosis apparatus 100 is constituted as a part of a generator for renewable energy such as solar power, a wind turbine generator or the like, or is installed in proximity to such a generator to diagnose a failure of the generator. In particular, according to an embodiment of the present invention, the fault diagnosis apparatus 100 includes an infrared camera to track a plurality of ROIs (Region Of Interest) of the generator, And records (stores) them continuously. The temperature of the plurality of regions of interest will tend to be. For example, the temperature of any one of the regions of interest will have a particular tendency depending on the environment, such as weather, season, day, and so on. Therefore, the fault diagnosis apparatus 100 accumulates this tendency. The information accumulated and stored as such tendency is referred to as "trend information ". If the temperature of a particular region of interest (ROI) is stored and stored for a predetermined period of time, the failure is diagnosed. For example, in winter, at 3 pm, in clear, cloudless weather, it is assumed that the ROI changes in a range of temperatures between 35 and 40 degrees Celsius, and the trend is recorded. At this time, if the temperature of the region of interest (ROI) is out of the range (35 to 40 degrees) and continues for a predetermined time (for example, 20 minutes or more) If the apparatus 100 is maintained at a temperature of 45 degrees or more for 20 minutes or more, it can be judged as a failure.

The fault diagnosis apparatus 100 may track the ROI as described above and continuously transmit the recorded results to the fault diagnosis server 200. [ That is, the fault diagnosis apparatus 100 can transmit an image of a generator including ROIs, a temperature of each of a plurality of ROIs measured by the fault diagnosis apparatus 100, and the like. In particular, when the failure diagnosis apparatus 100 determines that a failure has occurred, the failure diagnosis apparatus 100 can transmit to the failure diagnosis server 200 an identifier indicating a region of interest determined as a failure and an indicator indicating that a failure has occurred.

The fault diagnosis server 200 communicates with the fault diagnosis apparatus 100 via the network 10 and remotely controls the fault diagnosis apparatus 100 or receives information on the generator from the fault diagnosis apparatus 100. [ In the embodiment of the present invention, the network 10 is preferably an IP network such as the Internet. However, the present invention is not limited thereto, and any kind of network can be applied to the present invention as long as data communication is possible. The fault diagnosis server 200 provides the user terminal 300 connected to the fault diagnosis server 200 with information on the generator received from the fault diagnosis apparatus. The fault diagnosis server 200 communicates with the fault diagnosis apparatus 100 to detect the images photographed by the fault diagnosis apparatus 100 and the temperature of each of a plurality of ROIs measured by the fault diagnosis apparatus 100 . In addition, the fault diagnosis server 200 receives an identifier indicating a region of interest (ROI) determined by the fault diagnosis apparatus 100 as a failure and an indicator indicating that a fault has occurred. The failure diagnosis server 200 detects the temperature of each of the plurality of ROIs measured by the failure diagnosis apparatus 100 and the temperature of each of the ROIs measured by the failure diagnosis apparatus 100, Make it possible to view the fault. As an example of an interface for such browsing, a web page format can be used. In this respect, the fault diagnosis server 200 may be a web server. To this end, the fault diagnosis server 200 determines whether or not the temperature of each of the plurality of ROIs measured by the fault diagnosis apparatus 100, the ROI of a specific ROI measured by the fault diagnosis apparatus 100, , &Lt; / RTI > and the like. In addition, the interface for browsing may not only be a web page format but also a client / server interface. In addition, an interface through a direct connection with the fault diagnosis server 200 may be provided.

In addition, the fault diagnosis server 200 may provide an interface for allowing a user of the connected user terminal 300 to remotely control the fault diagnosis apparatus 100. [ These interfaces can also provide both web page format, client / server format, and direct format.

The user terminal 300 is connected to the fault diagnosis server 200 through the network 10 to browse the information collected by the fault diagnosis apparatus 100 and if necessary, , And transmits a command for controlling the fault diagnosis apparatus 100 to the fault diagnosis server 200. Then, the failure diagnosis server 200 can transmit the command to the failure diagnosis apparatus 100 to control the failure diagnosis apparatus 100. [ The user terminal 300 may be any type of device that can access the fault diagnosis server 200 through a network such as a smart phone, a personal computer, a PDA, or the like.

2 is a block diagram illustrating a structure of a fault diagnosis apparatus according to an embodiment of the present invention.

2, the fault diagnosis apparatus 100 includes a camera unit 110, a motion unit 120, a storage unit 130, a communication unit 140, and a control unit 150.

The camera unit 110 photographs a thermal image of a portion including at least one region of interest (ROI) of the generator, and provides the photographed thermal image to the control unit 150. The camera unit 110 according to the embodiment of the present invention performs autofocusing under the control of the controller 150. [ The camera unit 110 includes a camera module and a focusing module. The camera module is an infrared (IR) camera, and images taken through it are thermal images. The focusing module is for performing focusing by moving the lens of the camera module. According to an embodiment of the present invention, in order to measure a more accurate object temperature, the focusing module moves the lens of the camera module to perform focusing. The focusing module includes a motor for rotating the focusing ring for moving the lens forward or backward relative to the subject including the region of interest of the camera module, and a device for driving the motor. The focusing module operates under the control of the controller 150.

The movement unit 120 is mounted on a lower portion of the camera unit 110 to move the camera unit 110 up and down and left and right so that the camera unit 110 tracks and captures ROIs of interest. The movement unit 120 may represent a so-called pan / tilt, but any device capable of physically moving the camera unit 110 can be used. At this time, the movement unit 120 physically moves the camera unit 110 under the control of the control unit 150. [ To this end, the motion unit 120 includes a motor and a device for driving the motor under the control of the controller 150.

The storage unit 130 is for storing a thermal image of at least one ROI captured by the camera unit 110, a temperature of at least one ROI measured thereby, and the like. In addition, the storage unit 130 stores a template, which is information necessary for the camera unit 110 to track a plurality of regions of interest. This template is an image of the detected edge of the region of interest and will be described in more detail below. Also, the storage unit 130 stores the trend of the temperature according to the surrounding environment in the steady state of each of the plurality of ROIs as accumulated data. Here, the surrounding environment may be a season, a weather, a humidity, an external temperature, a time, etc. when measuring the temperature of the region of interest, and a trend is a range of temperatures that can be varied depending on such a surrounding environment. It is referred to as "trend information" in which the tendency of the temperature change according to the surrounding environment is cumulatively stored.

The communication unit 140 is for communicating with the fault diagnosis server 200 and communicates with the fault diagnosis server 200 via the network 10. [ The communication unit 140 receives the image, data, message, etc. according to the embodiment of the present invention from the control unit 150 and transmits the received image, data, message, etc. to the fault diagnosis server 200 through the network 10. The communication unit 140 can receive the command from the failure diagnosis server 200 and transmit it to the control unit 150. [ To this end, the communication unit 140 includes modules for enabling communication using a communication protocol according to the type of the network 10. [

The control unit 150 controls the camera unit 110 to perform autofocusing and controls the movement unit 120 to move the camera unit 110 so that the camera unit 110 can move a plurality of interest areas (ROI) of the subject. The control unit 150 collects the thermal image from the camera unit 110 in real time and stores the collected thermal image in the data storage module 450. The control unit 150 measures the temperature of a specific part of the degradation image, for example, a region of interest (ROI) from the thermal image, compares the measured temperature of the region of interest with previously stored trend information, can do. In addition, the control unit 150 may map the measured temperature to a corresponding region of interest on the degraded image, and store the mapped region in the storage unit 130.

According to the embodiment of the present invention, when photographing a thermal image through the camera unit 110, focusing must be performed to measure a more accurate temperature, extract and track a more accurate region of interest. This focusing method will be described below.

FIG. 3 is a flowchart for explaining a focusing method according to an embodiment of the present invention. FIGS. 4 and 5 are graphs for explaining a focusing method according to an embodiment of the present invention, and FIGS. 6 and 7 are graphs FIG. 7 is a screen example for explaining focusing according to the embodiment.

Focusing according to the focusing method described below can be performed at any time during the operation of the camera unit 110 as needed. For example, it can be performed any time before setting the ROI, tracking the ROI, or shooting the ROI.

Prior to the description of the drawings, the present invention is directed to a system and method for imaging at least one specific portion of a generator, i.e., a region of interest (ROI), via an infrared camera, Is judged to be faulty. At this time, the camera unit 110 stores a template derived only of edges of a plurality of ROIs stored in advance, and tracks a plurality of ROIs through a stored template. At this time, the present invention sets a more accurate template and performs focusing to track the ROI more precisely and accurately.

In the embodiment described below, focusing is performed by moving the lens of the infrared camera. At this time, the controller 150 performs focusing by rotating the focusing ring for moving the lens of the camera unit 110 forward or backward with respect to a subject (for example, an area of interest) in a clockwise or counterclockwise direction.

In addition, the thermal image photographed through the camera unit 110 of the present invention is a gray-level image that displays the temperature of the subject through shading. 6 (A) shows a thermal image which is a gray-level image. The control unit 110 converts such a gray level image (hereinafter referred to as "gray image") into an image (hereinafter referred to as "binary image") displayed only at two levels, which are displayed only in black and white, When converting into an image, the threshold value is adjusted to finally convert the edge of the object into the detected image (hereinafter referred to as "edge image").

When the binary image is converted into a binary image, the threshold value is adjusted so that various white pixel portions and black pixel portions are changed as shown in FIGS. 6 (B), 6 (C) and 6 Lt; / RTI > In this binarization step, the threshold value is adjusted to find the edge of the object, and the detected edge is returned to generate an edge image composed of the returned edge. FIG. 6E shows an example of an edge image. As shown, the converted edge image is displayed in white with the edge and black in the other part. The focusing according to the embodiment of the present invention focuses when the number of white pixels in the edge image shown in (C) of FIG. 6 is the maximum, that is, when the edge appears most accurately.

The fault diagnosis apparatus 100 has high probability of being installed inside the generator. In this case, since the light is insufficient, it is difficult to use the contrast detection technique and the phase difference detection technique as they are. For example, since there is not enough light in the nugle where the core parts of the wind turbine are mounted, contrast detection and phase difference detection techniques can not be used as they are. For this reason, in the present invention, as described above, the thermal image is converted into the binary image, and the focusing is performed using the edge image converted from the binary image. Hereinafter, a focusing method according to an embodiment of the present invention will be described in more detail.

3, the control unit 150 receives a thermal image photographed at an initial position of the lens from the camera unit 110 in step S305, converts the received thermal image into an edge image, Measure the number. Here, the process of obtaining an edge image from a thermal image is as described above. In step S310, the control unit 150 controls the camera unit 110 to move the lens in either one of the forward and backward directions by a predetermined distance.

Then, the control unit 150 receives the thermal image photographed at the moved position from the camera unit 110 in step S315, converts the received thermal image into an edge image, and measures the number of white pixels of the edge image .

Then, the controller 150 determines whether the number of white pixels of the edge image at the moved position is equal to or greater than the number of white pixels of the edge image at the position before the movement. If the number of white pixels in the edge image is equal to or greater than the number of white pixels in the edge image at the position where the number of white pixels is shifted, the process proceeds to step S330. Otherwise, if the number of white pixels is decreased, The flow advances to step S325.

If the number of white pixels is decreased, the control unit 150 moves away from focusing, and the control unit 150 proceeds to step S325 and changes the moving direction of the lens so as to move to step S315.

On the other hand, if the number of white pixels in the edge image is equal to or greater than the number of white pixels in the edge image at the position where the number of white pixels is shifted, The change rate of the number of white pixels of the edge image is measured.

Then, in step S335, the controller 150 determines whether the rate of change of the number of white pixels of the edge image at the pre-movement position and the post-movement position is less than a predetermined value.

If the rate of change is not less than the predetermined value, that is, if the rate of change is equal to or greater than a preset value, the controller 150 moves the lens by a distance in inverse proportion to the rate of change of white pixels, The flow advances to step S315. On the other hand, if the rate of change is less than the predetermined value, the controller 150 determines that focusing is completed in step S345 and ends the process.

With respect to the above-described step S335, there is a difference of about 200 from the maximum value of the number of the whitish pixels at the time when the actual ideal focusing is completed. Therefore, if the white pixel value is within a predetermined range through the experimentally obtained value, that is, if the white pixel value is within the predetermined range, it can be determined that the focusing is completed.

Referring to the graph of FIG. 4, the horizontal axis represents the position of the lens, and the vertical axis represents the number of white pixels. The control unit 150 controls the camera unit 110 to move the focusing ring when the lens is moved, and the moving distance is adjusted through the time for operating the motor that rotates the focusing ring clockwise or counterclockwise . Therefore, the moving distance of the lens is proportional to the time of moving the focusing ring, and is also proportional to the angle of movement of the focusing ring in the clockwise or counterclockwise direction.

Here, the rate of change

) Can be expressed by the following equation (1).

As shown in Equation 1,

Is the rate of change of white pixels in the position before and after the movement.

That is, a moving distance, and corresponds to a time at which the focusing ring rotates when the focusing ring for moving the lens rotates at a constant speed.

Represents the number of white pixels of the edge image at the position before movement,

Represents the number of white pixels of the edge image at the position after the movement.

The moving distance of the lens is moved inversely proportional to the rate of change based on the rate of change described above. The moving distance of the lens can be calculated according to the following equation (2).

here,

Represents a moving distance calculated according to a rate of change of a white pixel due to a previous movement,

Represents the previous movement distance,

silver

This change rate,

.

5, the lens is moved in the direction in which the white pixels extend in the edge image, and whenever the movement is made, the change in the white pixel with respect to the moved distance (Equation 1) Calculates the next movement distance according to the change rate, and moves the lens by the calculated movement distance. As shown in the graph of FIG. 5 (I), when this process is repeated, the lens is sequentially moved from the first position to the ninth position (1 to 9)

It will also gradually decrease. As a result, in the ninth position 9,

Is stopped at the ninth position (9). Thus, focusing is completed. That is, as shown in (I) of FIG. 5, the rate of change (slope) of the number of white pixels is obtained using a hill climbing algorithm, and the lens is moved so as to be in inverse proportion to the rate of change using the obtained rate of change The focus is searched by adjusting the distance (time for operating the motor that rotates the focusing ring clockwise or counterclockwise).

At this time, in the graph shown in (J) of FIG. 5, it may happen that the focusing point can not be found, as in the case of moving from the ninth position (9) to the tenth position (10). That is, as described in steps S320 and S325, the number of white pixels may be reduced. In this case, the change rate

Has a positive value and is changed to a negative value. Alternatively, when focusing is performed in the opposite direction, negative values can be changed to positive values. In this case, the moving direction of the lens is changed to move the lens.

Thus, when the lens moves near the maximum value to be focused, the focusing can be terminated if it is less than a preset change rate. That is, with respect to the above-described step S335, there is a difference of about 200 from the maximum value of the number of the whitish pixels at the time when the actual ideal focusing is completed. Therefore, if the white pixel value is within a predetermined range through the experimentally obtained value, that is, if the white pixel value is within the predetermined range, it can be determined that the focusing is completed.

7 (A) shows an edge image in which focusing is not completely performed, and FIG. 7 (B) shows an edge image after focusing is completed. As can be seen, when the focusing is completed, it can be seen that more points and lines of white pixels appear before the focusing is completed. As can be seen, if focusing is performed completely, more accurate tracking can be performed, and this tracking method will be described in more detail below.

FIG. 8 is a flowchart illustrating a method of setting a region of interest according to an embodiment of the present invention. FIG. 9 is a diagram illustrating a method of setting a region of interest according to an embodiment of the present invention.

The region of interest (ROI) is an area for observing the temperature for diagnosis of a failure of a generator for renewable energy, and this region of interest is selected by a manager designating a specific region. When the failure diagnosis server 200 receives a thermal image from the failure diagnosis apparatus 100, the user (manager) of the user terminal 300 connected to the failure diagnosis server 200 browses the thermal image, You can specify the region of interest on the degradation phase. Then, the fault diagnosis server 200 transmits the designated area of interest to the fault diagnosis apparatus 100, and accordingly, the fault diagnosis apparatus 100 performs the following operations.

6, 8 and 9, the controller 150 receives a thermal image from the camera unit 110 in step S810. Such a thermal image is shown in FIG. 6 (A) in advance of the gray level image. Then, the controller 150 converts the gray level image into an edge image in step S820. More specifically, in step S820, the controller 150 converts a thermal image, which is a gray image, into a binary image having only two colors of black and white, and converts the binary image into an edge image again. When converting from a binary image to an edge image, as shown in (B), (C) and (D) of FIG. 6, the edge of the binary image is adjusted to detect an edge, do. That is, the controller 150 detects edges from the binary image, and converts the edges into an edge image. An edge image composed of only such an edge is shown in (E) of FIG. In addition, such an edge image can be detected from a binary image using a canny edge detection method.

After converting the edge image into the edge image, the controller 400 stores the ROI in the edge image as a template in step S840. This completes the process of setting the region of interest.

9 (L) represents a thermal image, FIG. 9 (M) represents an edge image obtained from a thermal image, and FIG. 9 (N) represents a template of an ROI selected in an edge image . Thus, the template is extracted from the edge image and consists of the edges of the region of interest. Then, we will explain how to trace the area of interest through these templates and diagnose the fault.

FIG. 10 is a flowchart illustrating a method for diagnosing a fault using a thermal image according to an embodiment of the present invention, and FIG. 11 is a view illustrating a tracking method for performing a fault diagnosis according to an embodiment of the present invention.

Referring to FIG. 10, in step S1010, the controller 150 of the failure diagnosis apparatus 100 sets at least one region of interest (ROI), which is a specific region of the generator. Here, the ROI is an area where temperature is to be measured in order to diagnose a failure, and the setting of the ROI is stored as a template composed of only the edges of the ROI. Step 1010 is the same as the embodiment described in Fig.

After the ROI is set, the controller 150 controls the camera unit 110 and the motion unit 120 to continuously track at least one ROI in step S1020. In particular, the ROI acquires an edge image from a thermal image taken from the camera unit 110, and then matches the template for the ROI with the obtained edge image to track the ROI . In particular, when a plurality of regions of interest are set in advance, they can be tracked by matching with templates stored in advance. At this time, the tracking of the region of interest must first convert the thermal image into an edge image in preparation for matching with the edge stored as a template. That is, the control unit 400 converts the thermal image captured by the camera unit 200 into a binary image. Then, the control unit 400 detects an edge in the binary image and converts it into an edge image consisting only of an edge of the object. When the current thermal image is converted into an edge image, the controller 400 compares the edge image of the transformed edge image with the edge of the template stored in advance, determines the matching area as a region of interest (ROI), and performs tracking do.

As shown in FIG. 11, an example of a screen in which templates previously stored in an edge image are matched is shown.

Template matching is a pattern matching technique that uses templates to track specific objects.

In the embodiment of the present invention, a template matching function provided by OpenCV (Opne Computer Vision) can be used. OpenCV is an open source computer vision C library. The template matching algorithm used in OpenCV can provide a total of six matching methods as shown in Table 1 below.

CV_TM_SQDIFF Square-root matching CV_TM_SQDIFF_NORMED Normalization of Squared Difference Matching CV_TM_CCORR Correlation matching CV_TM_CCORR_NORMED Normalization of Correlation Matching CV_TM_CCOEFF Correlation coefficient matching CV_TM_CCOEFF_NORMED Normalization of Correlation Coefficient Matching

In the embodiment of the present invention, the similarity is calculated by comparing the photographed image while shifting the template to be searched from the image photographed by the camera unit 110 from left to right and from top to bottom by one pixel using correlation coefficient matching And the result is converted into a correlation coefficient so that the point at which the correlation coefficient becomes equal to or greater than a predetermined number (for example, 0.7) can be continuously tracked.

Then, the controller 150 measures (records) the temperature of the ROI of the degraded image photographed in step S1030. In step S1040, the controller 150 compares the pre-stored trends with the measured temperatures to diagnose the failure. The temperature of a particular area of interest will have a tendency depending on factors such as the external environment. For example, the temperature of any one of the regions of interest may have a specific tendency depending on the environment such as weather, season, and sunrise. Accordingly, the control unit 150 continuously measures the temperature of each ROI, continuously measures the temperature of the ROI and the surrounding environment (season, weather, humidity, external temperature, time, etc.) (Stored). That is, the controller 150 can diagnose that the temperature of the region of interest (ROI) has previously stored and accumulated, for a preset period of time, to be out of order by referring to the trend information. For example, in the summer, at 2 pm, and at a humidity of 70%, it is assumed that a particular region of interest (ROI) varies in the range of temperatures between 50 and 55 degrees Celsius and that trend is recorded. If the temperature of the region of interest (ROI) is out of the range (50 to 55 degrees) and continues for a predetermined period of time (for example, 20 minutes or more), that is, 20 If the temperature is maintained above 56 ° C, it can be judged as a failure.

In step S1050, the control unit 150 transmits the thermal image of the region of interest, the measured temperature, and the failure or the like to the failure diagnosis server 200 through the communication unit 140. [

The fault diagnosis method using the thermal image according to the present invention can be implemented in the form of software readable by various computer means and recorded in a computer-readable recording medium. Here, the recording medium may include program commands, data files, data structures, and the like, alone or in combination. Program instructions to be recorded on a recording medium may be those specially designed and constructed for the present invention or may be available to those skilled in the art of computer software. For example, the recording medium may be an optical recording medium such as a magnetic medium such as a hard disk, a floppy disk and a magnetic tape, a compact disk read only memory (CD-ROM), a digital video disk (DVD) A magneto-optical medium such as a floppy disk and a ROM, a random access memory (RAM), a flash memory, a solid state disk (SSD), a hard disk drive (HDD) And hardware devices specifically configured to store and perform the same program instructions. Examples of program instructions may include machine language code such as those generated by a compiler, as well as high-level language code that may be executed by a computer using an interpreter or the like. Such hardware devices may be configured to operate as one or more software modules to perform the operations of the present invention, and vice versa.

While the present invention has been described with reference to several preferred embodiments, these embodiments are illustrative and not restrictive. It will be understood by those skilled in the art that various changes and modifications may be made without departing from the spirit of the invention and the scope of the appended claims.

10: Network 100: Fault diagnosis device
110: camera unit 120:
130: storage unit 140: communication unit
150: Control section 200: Fault diagnosis server
300: user terminal

Claims

In a fault diagnosis system,
A camera unit for photographing a thermal image; And
After converting the thermal image into an edge-only edge image, the converted edge image is compared with a template storing an edge of a region of interest (ROI) to track the ROI, And a controller for measuring the temperature of the tracked area of interest from the thermal image to diagnose the failure.

The method according to claim 1,
The control unit
Measuring the temperature of the tracked area of interest from the thermal image of the tracked area of interest,
And diagnoses the failure as a range of the temperature change according to the surrounding environment of the accumulated and accumulated interest area within a preset time.

The method according to claim 1,
The control unit
Converting the thermal image into an edge image, and moving the lens of the camera unit so as to maximize the number of white pixels obtained from the edge image.

The method of claim 3,
The control unit
Until the number of white pixels of the edge image becomes maximum,
After moving the lens of the camera unit by a set distance in either one of forward and backward directions,
If the number of white pixels of the edge image at the post-movement position is equal to or greater than the number of pixels of the edge image at the pre-movement position,
And controls the lens to move in a direction opposite to the one direction if the number of white pixels of the edge image at the position after the movement is less than the number of pixels of the edge image at the position before the movement.

5. The method of claim 4,
The control unit
If the number of white pixels of the edge image at the post-movement position is equal to or greater than the number of pixels of the edge image at the pre-
The moving distance is changed in inverse proportion to the rate of change of the number of white pixels of the edge image at the post-moving position with respect to the number of pixels of the edge image at the pre-movement position with respect to the set moving distance, Is continuously moved.

The method according to claim 1,
The control unit
If the correlation between the transformed edge image and a template storing an edge of the ROI is greater than or equal to a predetermined value, it is determined that matching is performed and tracking is performed.

In the fault diagnosis method,
Capturing a thermal image;
Comparing the converted edge image with a template storing an edge of a region of interest (ROI), and tracking the ROI, after converting the thermal image into an edge-only edge image; And
And measuring the temperature of the tracked area of interest from a thermal image of the tracked ROI to diagnose the failure.

8. The method of claim 7,
The step of diagnosing
Measuring the temperature of the tracked area of interest from the thermal image of the tracked area of interest,
Wherein diagnosis of a failure is made when a range of temperature change according to the surrounding environment of the accumulated and accumulated interest area is out of a predetermined time.

8. The method of claim 7,
Before the tracking step,
Further comprising the step of converting the thermal image into an edge image and moving the lens of the camera unit so as to maximize the number of white pixels obtained from the edge image to perform focusing.

10. The method of claim 9,
The step of performing focusing
Until the number of white pixels of the edge image becomes maximum,
After moving the lens of the camera unit by a set distance in either one of forward and backward directions,
If the number of white pixels of the edge image at the post-movement position is equal to or greater than the number of pixels of the edge image at the pre-movement position,
And moving the lens in a direction opposite to the one direction when the number of white pixels of the edge image at the position after the movement is less than the number of pixels of the edge image at the position before the movement.

11. The method of claim 10,
The step of performing focusing
If the number of white pixels of the edge image at the post-movement position is equal to or greater than the number of pixels of the edge image at the pre-
The moving distance is changed in inverse proportion to the rate of change of the number of white pixels of the edge image at the post-moving position with respect to the number of pixels of the edge image at the pre-movement position with respect to the set moving distance, Is continuously moved.

8. The method of claim 7,
The step of tracking
And if the correlation between the transformed edge image and a template storing an edge of the ROI is greater than or equal to a predetermined value, it is determined that matching is performed and tracking is performed.