A Fault Diagnosis Method Based on Infrared Images

Abstract

The present application relates to the field of infrared detection technology, and aims to provide a fault diagnosis method based on infrared images, the method comprising: obtaining a device image pre-stored at the current viewing angle; obtaining an infrared image at the current viewing angle; identifying overlapping halo areas in the infrared image; when overlapping halo areas exist, comparing the overlapping halo areas with the device image to obtain temperature images of multiple target devices; judging whether the target devices have thermal faults based on the infrared images of the multiple target devices. By identifying overlapping halo areas, the heating device blocked by the front device can be identified, and combined with the device temperature, it can be judged whether the blocked heating device has a heating fault, thereby improving the accuracy of detecting the fault location of the power equipment.

Description

Technical Field

[0001] The present application relates to the field of infrared detection technology, and in particular to a fault diagnosis method based on infrared images. Background Art

[0002] Thermal failure is the most common failure in the operation of current power equipment. Currently, most of the power equipment in operation is tested by thermal detection to determine whether the power equipment has a thermal failure.

[0003] Most of the current thermal detection methods are infrared image detection, that is, infrared images of power equipment are obtained through infrared devices to determine whether the power equipment has a thermal fault. In the general infrared image judgment process, an infrared image of the power equipment at a time point is first obtained, and after a preset time, an infrared image of the power equipment is obtained again. By comparing and judging the color depth of the power equipment in the two infrared images, it is determined whether the power equipment has a thermal fault; or, the infrared image of the power equipment is directly obtained, and the temperature of the power equipment is determined by the infrared image. When the device temperature is greater than the preset temperature threshold, it is determined that the power equipment has a thermal fault.

[0004] However, the two existing infrared image detection methods have inaccurate judgments when applied to specific power equipment. In some large-scale power equipment, the power components are densely arranged and the number of heating components is large. It is difficult to accurately judge the high-temperature power components through infrared images, and it is also impossible to accurately indicate faults through infrared images. Summary of the invention

[0005] The purpose of the present application is to provide a fault diagnosis method based on infrared images, which can improve the accuracy of detecting fault locations of power equipment.

[0006] The above application objectives of this application are achieved through the following technical solutions:

[0007] A fault diagnosis method based on infrared images, the method comprising:

[0008] Get the device image stored in the current perspective;

[0009] Get the infrared image of the current viewing angle;

[0010] Identify overlapping halo areas in infrared images;

[0011] When there are overlapping halo areas, the overlapping halo areas are compared with the device image to obtain temperature images of multiple target devices;

[0012] Determining whether a thermal failure occurs in the target device according to the infrared images of the plurality of target devices.

[0013] In a preferred embodiment, the method of identifying overlapping halo regions in an infrared image comprises:

[0014] Identify a front-side component in the infrared image, wherein the front-side component is an electrical component where a heat source is located in the infrared image;

[0015] Identifying an incomplete image around the infrared image of the front-side device itself;

[0016] If the incomplete image exists, the image area in the grayscale image corresponding to the incomplete image is identified as an overlapping halo area.

[0017] In a preferred embodiment, the method for identifying the front-side device in the infrared image is:

[0018] Performing grayscale processing on the infrared image of the current viewing angle to obtain a grayscale image;

[0019] Identify a high heat peak point in the grayscale image, where the high heat peak point is a pixel point whose grayscale value is smaller than the pixel points within a preset range around the high heat peak point;

[0020] The power device corresponding to the high thermal peak point is identified as a front-side device.

[0021] In a preferred embodiment, the method for identifying the incomplete image around the infrared image of the front device itself is:

[0022] Binarizing the grayscale image to obtain a binary image;

[0023] Constructing a front grayscale area corresponding to the front device in the binary image according to the binary image;

[0024] When the edge line of the front grayscale region is destroyed, if the image region destroying the front grayscale region is not other front grayscale regions, the image region destroying the front grayscale region is identified as the incomplete image.

[0025] In a preferred embodiment, the binarization threshold of the binarization process is:

[0026] Kx=Kc+K; K is a constant, Kc is an edge threshold of the edge grayscale of the grayscale region of the current front-side device, and Kx is the binarization threshold.

[0027] In a preferred embodiment, the method of comparing the overlapping halo area with the device image to obtain temperature images of multiple target devices includes:

[0028] Identifying the power device corresponding to the overlapping halo area according to the shape data of the overlapping halo area and the positional relationship with the infrared image of the front device;

[0029] The power device corresponding to the overlapping halo area and the power device corresponding to the front-side device are output as the target device.

[0030] In a preferred embodiment, the method for determining whether the target device has a thermal failure based on the infrared images of the plurality of target devices includes:

[0031] Calculating the center temperature of the target device according to the overlapping halo area and the shape of the target device corresponding to the overlapping halo area;

[0032] When the center temperature is greater than a preset temperature, it is determined that a thermal failure occurs in the target device.

[0033] In a preferred embodiment, the method for determining whether a thermal failure occurs in the target device based on the infrared images of the plurality of target devices further comprises:

[0034] Calculating the device temperature corresponding to the high thermal peak point of the front-side device;

[0035] When the device temperature is greater than a preset temperature, it is determined that a thermal failure occurs in the target device.

[0036] In a preferred embodiment, the center temperature is the temperature of the geometric center of the target device, or the temperature of a pre-marked hot spot of the target device.

[0037] In a preferred embodiment, the method further comprises:

[0038] Obtaining the target device center temperature or the growth value of the device temperature within a preset time according to the above;

[0039] If the growth value is greater than the preset growth value of the target device, a prompt message indicating that the target period is an emergency device is output.

[0040] To summarize, the present application includes the following beneficial technical effects: by identifying overlapping halo areas, the heating device blocked by the front device can be identified, and combined with the device temperature, it can be judged whether the blocked heating device has a heating fault, thereby improving the accuracy of detecting the fault location of the power equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 It is a flowchart of the fault diagnosis method in the embodiment of the present application.

[0042] Figure 2 It is a schematic diagram of the process of identifying overlapping halo areas in infrared images in an embodiment of the present application.

[0043] Figure 3 It is a schematic diagram of the process of identifying the front-side device in the infrared image in the embodiment of the present application.

[0044] Figure 4 It is a schematic diagram of the process of identifying an incomplete image in an embodiment of the present application. DETAILED DESCRIPTION

[0045] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0046] In addition, the term "and / or" in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article, unless otherwise specified, generally means that the associated objects before and after are in an "or" relationship.

[0047] The embodiments of the present application are further described in detail below in conjunction with the drawings in the specification.

[0048] An embodiment of the present application provides a fault diagnosis method based on infrared images, wherein the infrared images are obtained by taking infrared devices installed on power equipment, and the infrared devices are connected to an equipment monitoring system. The main process of the method is described as follows.

[0049] like Figure 1 As shown:

[0050] Step 100: Obtain the device image pre-stored at the current viewing angle.

[0051] The device image corresponds to an image obtained by photographing the power equipment at the installation position of the infrared device after the infrared device is installed. The device image is pre-stored in the system and can be called in real time by the infrared device. The device image contains the device information of the power devices displayed by itself, such as resistor R1, inductor L1, transformer T1, central controller U1, etc., and after the power equipment is installed, the models of the power devices used are fixed, and the device information also includes the shape information and location information of the power devices.

[0052] Furthermore, the device image is a line image, and the line image contains the contour lines of each power device in the image obtained by taking a photo. The method of indicating the name and contour lines of the structure in the image in the photographed image is a mature technology in the prior art, and this application will not be described in detail.

[0053] Furthermore, a pre-marked hot point is marked for each contour line in the line image, and the pre-marked hot point is the hot point of the power device. For example, after an overcurrent occurs in a resistor device, the geometric center of the resistor heats up first, and the geometric center of the contour line of the resistor device is marked as the pre-marked hot point. For another example, when a circuit breaker is overloaded, the coil of the circuit breaker heats up first, and the contour line of the circuit breaker corresponding to its own coil is marked as the pre-marked hot point.

[0054] For an electric device that is prone to thermal failure, the first part to generate heat when a thermal failure occurs is a part of its own area. The pre-marked hot spots are manually entered, and the pre-marked hot spots are based on the points corresponding to the parts of its own area.

[0055] Step 200: Acquire an infrared image of the current viewing angle.

[0056] The infrared image is an image of the device acquired in real time by the infrared device during the operation of the power device.

[0057] Step 300: Identify overlapping halo regions in an infrared image.

[0058] The overlapping halo area is used to indicate the power devices in the infrared image that do not visually indicate their own heating state in the infrared image. Since there are many power devices in large power equipment and the spatial positions of the power devices are complex, it is easy for multiple power devices to overlap when taking pictures of the power equipment from a fixed perspective. In this case, it is difficult to directly and accurately determine which specific heating power devices are included by directly using information such as color depth and brightness in the infrared image. Therefore, the identification of overlapping halo areas is used to assist in the accurate identification of power devices.

[0059] Reference Figure 2 , the method of identifying overlapping halo areas in infrared images is as follows:

[0060] Step 310: Identify the front side device in the infrared image, where the front side device is an electric device with a heat point in the infrared image. The heat point is a point that characterizes the heat position of the electric device, and the temperature of the heat point is higher than the temperature of the rest of the positions of the electric device. The heat point can be a specific point in the infrared image, or a partial area corresponding to the electric device in the infrared image. In the embodiment of the present application, the heat point is a specific point in the infrared image.

[0061] Reference Figure 3 , the method for identifying the front side device in the infrared image is:

[0062] Step 311: grayscale processing is performed on the infrared image of the current viewing angle to obtain a grayscale image.

[0063] Step 312: Identify high thermal peak points in the grayscale image, where the high thermal peak points are pixels whose grayscale values ​​are less than or equal to those of pixels within a preset range around the high thermal peak points.

[0064] Specifically, first overlap the grayscale image with the line image, select the corresponding power device area through the contour lines in the line image, then select the point with the smallest grayscale value in the power device area as the pre-selected point (in the grayscale image obtained by grayscale processing, the higher the corresponding temperature, the smaller the grayscale value after grayscale processing). Finally, select a circular area with the pre-selected point as the center and the preset distance as the radius, and judge whether the grayscale values ​​of the remaining points in the circular area are greater than or equal to the grayscale value of the pre-selected point. If so, the pre-selected point is the high thermal peak point of the corresponding power device. The preset distance is a fixed value set manually in advance. It should be noted that the preset distance configured for each power device has different fixed values ​​set according to its own geometric size, installation position, installation angle, device model, etc.

[0065] Step 313: Identify the power device corresponding to the high thermal peak point as a front-side device.

[0066] Step 320: Identify the incomplete image around the infrared image of the front-side device itself.

[0067] After the front-side device is identified, the closed area formed by the heat of the front-side device can be delineated. Since the heat of power devices is mostly the heat of the power devices as a whole, such as when an inductor heats up, the coils that make up the inductor will heat up, that is, multiple high-heat peaks in the grayscale image of the power device can be identified, and only delineating a circular area as the closed area based on the high-heat peaks will result in errors. Figure 4 , the method for identifying the incomplete image in this application is as follows:

[0068] Step 321: binarize the grayscale image to obtain a binary image.

[0069] The binarization threshold of the binarization process is Kx=Kc+K; K is a constant and greater than 0, Kc is the edge threshold of the grayscale of the grayscale area edge of the current front-side device, and Kx is the binarization threshold. After selecting the front-side device, the grayscale image is overlapped with the line image to obtain the grayscale value corresponding to the edge line of the front-side device, the point with the lowest grayscale value in the edge line is selected as the standard edge point, and the grayscale value corresponding to the standard edge point is set as the edge threshold. Among them, the value of K is set manually and is positively correlated with the warning temperature of the current front-side device.

[0070] Step 322: constructing a front grayscale area corresponding to the front device in the binary image according to the binary image.

[0071] After obtaining the binary image, the area consisting of points containing the high heat peak point and having grayscale values ​​of 0 is selected as the preprocessing area, and the preprocessing area is patrolled and positioned to obtain the edge line of the preprocessing area. The edge line of the preprocessing area is divided into a plurality of arc segments and straight line segments, and the arc segments are bent toward the high heat peak point or away from the high heat peak point. The method of dividing the edge line of the preprocessing area into arc segments and straight line segments is a mature solution in the prior art and will not be described in detail in this application.

[0072] A patrol circle is constructed with the high thermal peak points of the front-side device as the center, and the radius of the patrol circle is gradually expanded from zero. When the number of tangents to the edge line in the preprocessing area is the largest, arc segments and straight line segments tangent to the patrol circle are selected and connected by smooth curves to obtain the front-side grayscale area.

[0073] It should be noted that a front-side device has multiple high thermal peaks. When it is necessary to construct a patrol circle, first connect the edges of the area composed of multiple high thermal peaks, and select high thermal peaks on the edge connection line at equal intervals to construct the patrol circle. The smaller the spacing set for selecting high thermal peaks, the more accurate the front-side grayscale area is.

[0074] Step 323: When the edge line of the front grayscale area is destroyed, if the image area that destroys the front grayscale area is not other front grayscale areas, the image area that destroys the front grayscale area is identified as the incomplete image. It should be noted that each front area corresponds to a front grayscale area, and the corresponding binarization thresholds in the process of obtaining the front grayscale area are different. The edge line of the current grayscale area destroyed by other front grayscale areas can be determined by superimposing different front grayscale areas. The heating boundary line of each heating area in the binary image is identified. When the heating boundary line intersects with the edge line of the front grayscale area, it is determined that the image area where the heating boundary line is located destroys the edge line of the front grayscale area. If the position of the heating boundary destroying the current front grayscale area is different from the position of the edge line of the current front grayscale area destroyed by other front grayscale areas, the image area of ​​the destroyed current front grayscale area is determined to be an incomplete image.

[0075] Step 330: If the incomplete image exists, the image area in the grayscale image corresponding to the incomplete image is identified as an overlapping halo area.

[0076] Step 400: Compare the overlapping halo area with the device image to obtain temperature images of multiple target devices.

[0077] The power device corresponding to the overlapping halo area is identified based on the shape data of the overlapping halo area and the positional relationship with the infrared image of the front-side device; and the power device corresponding to the overlapping halo area and the power device corresponding to the front-side device are output as the target device.

[0078] Step 500: Determine whether the target device has a thermal failure based on the infrared images of the multiple target devices. When the grayscale image is generated, the grayscale value of the pixel in the grayscale image is proportional to the temperature of the corresponding position, that is, the larger the grayscale value, the higher the temperature of the corresponding position of the device.

[0079] First, the device temperature corresponding to the high thermal peak point of the front-side device is calculated. When the device temperature is greater than a preset temperature, it is determined that a thermal failure occurs in the target device, that is, a thermal failure occurs in the front-side device.

[0080] Then, the center temperature of the target device is calculated based on the overlapping halo area and the shape of the target device corresponding to the overlapping halo area. The edge line of the target device in the grayscale image is constructed, and the edge line of the target device is overlapped with the overlapping halo area. The temperature of the pre-marked hot point is calculated based on the pre-marked hot point of the target device and the temperature gradient change from the overlapping halo area to the pre-marked hot point to obtain the center temperature. That is, when the power device generates heat, a temperature gradient change is formed from the pre-marked point to the edge of the power device. The greater the distance from the pre-marked hot point, the lower the temperature value. The grayscale value corresponding to the temperature of the pre-marked hot point is calculated by the relative position of the overlapping halo area and the pre-marked hot point. In another embodiment, the center temperature is the temperature of the geometric center of the target device, and the center temperature is obtained by calculating the temperature of the geometric center based on the geometric center of the target device and the temperature gradient change from the overlapping halo area to the geometric center.

[0081] In the technical solution of the present application, the heating device blocked by the front device can be identified by identifying the overlapping halo area, and the device temperature can be combined to determine whether the blocked heating device has a heating fault, thereby improving the accuracy of detecting the fault location of the power equipment.

Claims

1. A fault diagnosis method based on infrared images, characterized in that: The method comprises: Get the device image stored in the current view; Get the infrared image of the current viewing angle; Identifying overlapping halo areas in the infrared image; the overlapping halo areas are used to indicate that the power components in the infrared image do not visually indicate their own heating state in the infrared image; When there are overlapping halo areas, the overlapping halo areas are compared with the device image to obtain temperature images of multiple target devices; determining whether a thermal failure occurs in the target device according to the infrared images of the plurality of target devices; Methods for identifying overlapping halo regions in infrared images include: Identify a front-side component in the infrared image, wherein the front-side component is an electrical component where a heat source is located in the infrared image; Identifying an incomplete image around the infrared image of the front-side device itself; If the incomplete image exists, the image area in the grayscale image corresponding to the incomplete image is identified as an overlapping halo area; The method for identifying the front side device in the infrared image is: Performing grayscale processing on the infrared image of the current viewing angle to obtain a grayscale image; Identify a high heat peak point in the grayscale image, where the high heat peak point is a pixel point whose grayscale value is smaller than the pixel points within a preset range around the high heat peak point; Identify the power device corresponding to the high thermal peak point as a front-side device; The method of comparing the overlapping halo area with the device image to obtain temperature images of multiple target devices includes: Identifying the power device corresponding to the overlapping halo area according to the shape data of the overlapping halo area and the positional relationship with the infrared image of the front device; The power device corresponding to the overlapping halo area and the power device corresponding to the front-side device are output as the target device.

2. A fault diagnosis method based on infrared images according to claim 1, characterized in that: The method for identifying the incomplete image around the infrared image of the front device itself is: Binarizing the grayscale image to obtain a binary image; Constructing a front grayscale area corresponding to the front device in the binary image according to the binary image; When the edge line of the front grayscale region is destroyed, if the image region destroying the front grayscale region is not other front grayscale regions, the image region destroying the front grayscale region is identified as the incomplete image.

3. A fault diagnosis method based on infrared images according to claim 2, characterized in that: The binarization threshold of the binarization process is: Kx=Kc+K; K is a constant, Kc is an edge threshold of the edge grayscale of the grayscale region of the current front-side device, and Kx is the binarization threshold.

4. A fault diagnosis method based on infrared images according to claim 3, characterized in that: The method for determining whether a thermal failure occurs in the target device based on the infrared images of the multiple target devices includes: Calculating the center temperature of the target device according to the overlapping halo area and the shape of the target device corresponding to the overlapping halo area; When the center temperature is greater than a preset temperature, it is determined that a thermal failure occurs in the target device.

5. The fault diagnosis method based on infrared images according to claim 4 is characterized in that: The method for determining whether a thermal failure occurs in the target device based on the infrared images of the plurality of target devices also includes: Calculating the device temperature corresponding to the high thermal peak point of the front-side device; When the device temperature is greater than a preset temperature, it is determined that a thermal failure occurs in the target device.

6. A fault diagnosis method based on infrared images according to claim 5, characterized in that: The center temperature is the temperature of the geometric center of the target device, or the temperature of a pre-marked hot spot of the target device.

7. The fault diagnosis method based on infrared images according to claim 5 is characterized in that: The method further comprises: Obtaining the target device center temperature or the growth value of the device temperature within a preset time according to the above; If the growth value is greater than a preset growth value of the target device, a prompt message indicating that the target device is an emergency device is output.

Images