Systems for monitoring switchgear
By combining visual images with infrared images, combining images are generated to provide temperature information and feature recognition capabilities, the installation difficulties and identification problems of infrared cameras in unknown electrical assets are solved, and the fine temperature monitoring and component recognition of switch equipment is realized, improving the improvement of the improvement of the Fault detection capability.
Patent Information
- Application Number
- CN202110119590.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Priority Date
- 2020-01-29
- Filing Date
- 2021-01-28
- Publication Date
- 2024-12-03
- Estimated Expiration
- 2041-01-28
AI Technical Summary
Existing infrared cameras are difficult to install in unknown electrical assets, with low resolution and low sensitivity making it difficult to detect internal components in cold states of switching equipment, identifying individual components and objects, and identifying types, topology and compartments.
Using a system combining infrared cameras and vision cameras, pixels in the visual image are mapped into infrared images through processing units, and combined images are generated to provide temperature information and feature recognition capabilities, leveraging the high resolution of the visual camera and the temperature information of the infrared camera. , realize fine temperature monitoring and component identification of switching equipment.
Improves the temperature monitoring accuracy and component recognition capabilities in the cold state of the switching equipment, can clearly identify object boundaries in low-resolution infrared images, supports the recognition of the type, topology and compartment of the switching equipment, and enhances fault detection capabilities. .
Smart Images

Figure CN113191383B_ABST
Abstract
Description
Technical Field
[0001] The invention relates to a system for monitoring switchgear. Background Art
[0002] Infrared (IR) thermal imaging technology is widely used for non-contact temperature sensing and provides a convenient solution for permanent installation of online temperature monitoring of electrical assets such as switchgear. Temperature extraction of thermal images from IR cameras supports the detection of critical hot spots to prevent overheating of switchgear.
[0003] Today, the use of standalone IR cameras presents the following disadvantages to existing and new solutions for electrical assets:
[0004] [1] Particularly in unknown electrical assets, for commissioning and retrofitting purposes, the correct installation of IR cameras is rather cumbersome. Thermal images are not informative if the IR camera fully captures all relevant components of the electrical asset. This is especially true in the cold state of the switchgear if cost-effective IR cameras with low spatial resolution and low sensitivity are used. In this case, internal components can only be reliably detected after the switchgear has been heated up.
[0005] [2] It is difficult to identify individual components of an electrical asset and possibly other objects (such as forgotten tools) from thermal IR images alone. Identification of components and objects is critical to detecting the type of fault and which phases are affected.
[0006] [3] For switchgear, it is difficult to identify the type, topology, and compartments from thermal IR images alone.
[0007] [4] Low-resolution cameras make [1] to [3] even more difficult. In thermal images, the contour lines representing the boundary lines of the object being measured become less clear.
[0008] There is a need to address these issues. Summary of the invention
[0009] Therefore, it would be advantageous to have an improved system for monitoring switchgear.
[0010] The objects of the invention are solved by the subject-matter of the independent claims, wherein further embodiments are incorporated in the dependent claims.It should be noted that the system is described for switchgear, but is applicable in other electrical systems that may be affected by overheated components.
[0011] In one aspect, a system for monitoring a switchgear is provided, the system comprising:
[0012] - Infrared camera;
[0013] -Vision camera;
[0014] - a processing unit; and
[0015] - output unit;
[0016] The infrared camera is configured to acquire an infrared image of the switchgear. The visual camera is configured to acquire a visual image of the switchgear. The processing unit is configured to map a plurality of pixels in the visual image to a plurality of pixels in the infrared image. The plurality of pixels in the visual image include a plurality of pixel subsets, wherein each subset includes at least one pixel. Each pixel subset in the visual image is mapped to a corresponding pixel in the infrared image, and each pixel subset in the visual image and the corresponding pixel in the infrared image are associated with image data of a same portion of the switchgear. The processing unit is configured to determine a combined image, the combined image including utilization of the plurality of pixels in the visual image mapped to the plurality of pixels in the infrared image. The output unit is configured to output the combined image.
[0017] In other words, a temperature monitoring system for switchgear uses an infrared image and a visual image, wherein the infrared image may be of lower resolution than the visual image, wherein the fields of view of the cameras overlap because they image at least some of the same objects. A combined image is provided from the two images, the combined image providing feature recognition capabilities associated with the visual image with temperature information from the infrared image.
[0018] In an example, the infrared camera and the visual camera are housed within the same housing.
[0019] In an example, the housing is configured to be mounted to or removed from the switchgear.
[0020] In an example, the lens of the infrared camera is located adjacent to the lens of the visual camera.
[0021] In an example, the processing unit is configured to map a plurality of pixels in the visual image to a plurality of pixels in the infrared image using a field of view of the visual camera and a field of view of the infrared camera.
[0022] In an example, the field of view of the visual camera at least partially overlaps the field of view of the infrared camera.
[0023] In an example, the processing unit is configured to map a plurality of pixels in the visual image to a plurality of pixels in the infrared image using a range of a line of sight of the visual camera and a line of sight of the infrared camera.
[0024] Here, line of sight can refer to the field of view of the camera, so pixel mapping can be performed based on the spatial area imaged by the camera knowing the field of view of the camera. However, line of sight can also mean the centerline of the camera's optical axis - in other words, the direction the camera's optical axis points. Then, knowing this information allows pixel mapping, where camera specification information (such as the angular range of imaging) can be used.
[0025] Pixel mapping may involve assigning pixels of the visual image to pixels of the IR image, or vice versa, where a mapping table may be used. Such a mapping table may be used to give the temperature of each pixel of the visual image in the area overlapping with the IR camera image, and then be generally applicable to the selected type of camera.
[0026] Pixel mapping may also involve assigning colors representing the temperature from the thermal image to the pixels of the visual image, where again a mapping table may be used.
[0027] In an example, the processing unit is configured to map a plurality of pixels in the visual image to a plurality of pixels in the infrared image using a distance between a detector of the visual camera and a detector of the infrared camera.
[0028] Thus, knowing the lateral distance between the visual camera and the infrared camera, such as the lateral distance between the centers of their detectors, the overlap of their images can then be determined. For example, the cameras may be offset from each other but pointed in parallel directions, but by knowing the angular fields of view of the two cameras, it can be determined which pixels of the visual image correspond to which pixels of the infrared image in terms of imaging the same portion of the object. Similarly, if the two cameras are laterally offset but at an angle to each other, then again by knowing the angular fields of view of the cameras, the pixels are related to each other. Even more, if the two cameras are offset from each other, with one slightly in front of the other, then the lateral distance and forward distance can be utilized along with triangulation and angular fields of view to relate the pixels to each other.
[0029] In an example, the processing unit is configured to map a plurality of pixels in the visual image to a plurality of pixels in the infrared image using a mapping table.
[0030] In an example, the processing unit is configured to detect at least a portion of the switching device, the detecting comprising processing the combined image by the processing unit using an object recognition algorithm.
[0031] In an example, the processing unit is configured to detect one or more objects that are not part of the switchgear, the detecting comprising processing the combined image by the processing unit using an object recognition algorithm.
[0032] In an example, the detecting of the one or more portions of the switching device includes utilizing, by the processing unit, input from a user identifying at least one characteristic of the switching device.
[0033] In an example, the at least one characteristic of the switchgear comprises a topology and / or a compartment and / or a phase of the switchgear.
[0034] In an example, the processing unit is configured to assign at least one temperature to at least one detected portion of the switchgear, the at least one detected portion comprising utilization of an infrared image.
[0035] In an example, the resolution of the infrared camera is smaller than the resolution of the visual camera.
[0036] The above aspects and examples will be apparent from and elucidated with reference to the embodiments described hereinafter. BRIEF DESCRIPTION OF THE DRAWINGS
[0037] Exemplary embodiments will be described below with reference to the following drawings:
[0038] Figure 1 An example of a system for monitoring switchgear is shown;
[0039] Figure 2 An example of an image mapping workflow is shown;
[0040] Figure 3 An example of application of an image mapping workflow is shown;
[0041] Figure 4 An example of an excerpt from a mapping table is shown;
[0042] Figure 5 An example of combined image data is shown;
[0043] Figure 6 shows an example of a thermal infrared image at a) and a corresponding visual image at b) where an object detected in the visual image mapped onto the combined infrared and visual image enables infrared image data of the object to be extracted; and
[0044] Figure 7 An example of temperature feature extraction is shown. DETAILED DESCRIPTION
[0045] Figures 1 to 7A system for monitoring switchgear is provided. According to an example, the system includes: an infrared camera, a visual camera, a processing unit, and an output unit. The infrared camera is configured to acquire an infrared image of the switchgear. The visual camera is configured to acquire a visual image of the switchgear. The processing unit is configured to map a plurality of pixels in the visual image to a plurality of pixels in the infrared image. The plurality of pixels in the visual image include a plurality of pixel subsets, each subset including at least one pixel. Each pixel subset in the visual image is mapped to a corresponding pixel in the infrared image, and each pixel subset in the visual image and the corresponding pixel in the infrared image are associated with image data of a same portion of the switchgear. The processing unit is configured to determine a combined image, the combined image including utilization of the plurality of pixels in the visual image mapped to the plurality of pixels in the infrared image. The output unit is configured to output the combined image.
[0046] According to an example, the infrared camera and the visual camera are housed in the same housing.
[0047] According to an example, the housing is configured to be mounted on or removed from the switchgear.
[0048] According to an example, the lens of the infrared camera is located adjacent to the lens of the visual camera.
[0049] According to an example, the processing unit is configured to map a plurality of pixels in the visual image to a plurality of pixels in the infrared image using a field of view of the visual camera and a field of view of the infrared camera.
[0050] According to an example, the field of view of the visual camera at least partially overlaps the field of view of the infrared camera.
[0051] According to an example, the processing unit is configured to map a plurality of pixels in the visual image to a plurality of pixels in the infrared image using a range of a line of sight of the visual camera and a line of sight of the infrared camera.
[0052] According to an example, the processing unit is configured to map a plurality of pixels in the visual image to a plurality of pixels in the infrared image using a distance between a detector of the visual camera and a detector of the infrared camera.
[0053] According to an example, the processing unit is configured to map a plurality of pixels in the visual image to a plurality of pixels in the infrared image using a mapping table.
[0054] According to an example, the processing unit is configured to detect at least a portion of the switching device, the detecting comprising processing the combined image by the processing unit using an object recognition algorithm.
[0055] According to an example, the processing unit is configured to detect one or more objects that are not part of the switchgear, the detecting comprising processing the combined image by the processing unit using an object recognition algorithm.
[0056] According to an example, the detecting of the one or more parts of the switching device includes utilizing, by the processing unit, an input from a user identifying at least one characteristic of the switching device.
[0057] According to an example, the at least one characteristic of the switchgear comprises a topology and / or a compartment and / or a phase of the switchgear.
[0058] According to an example, the processing unit is configured to assign at least one temperature to at least one detected portion of the switchgear, the at least one detected portion comprising utilization of the infrared image.
[0059] According to an example, the resolution of the infrared camera is smaller than the resolution of the visual camera.
[0060] Thus, switchgear or other electrical equipment may be installed in association with the above-described system.
[0061] Therefore, the system is a dual camera system with an IR camera and a visual camera with a mapping process for temperature monitoring of electrical assets. The dual camera system includes a specially designed housing in which the IR camera and the visual camera are mounted. The lenses of the two cameras should be kept at a minimum distance from each other to maximize the overlap of the IR image and the visual image. Taking into account this distance and the field of view of the two cameras and what they observe in the calibration step, a mapping process is derived to obtain an enhanced thermal image that represents a fine overlap of heat on the visual image. The enhanced thermal image enables components to be identified in electrical assets and the temperatures of these areas of interest can be shown. The temperature characteristics of these areas of interest can be easily extracted for condition monitoring, such as maximum / minimum temperature, average temperature, and ambient temperature.
[0062] Continuing to refer to the drawings, the system for monitoring switchgear is described in further detail with respect to specific embodiments.
[0063] Figure 1 An example design of a dual camera housing with an infrared camera and a visual camera mounted therein is shown, wherein Figure 1 A front view is shown. The dual camera housing is specifically designed to house two cameras and ensure a fixed minimum distance / offset between the two lenses. Another benefit of the housing is the flexible fixing in the electrical asset. Figure 1 A possible implementation of a dual camera housing manufactured by 3D printing is shown. As mentioned above, the design focuses on maintaining a minimum distance between the two lenses of the infrared camera and the visual camera.
[0064] Figure 2: An example of an image mapping method is shown: at a) a separate visual image and a thermal image showing a heated black rectangular object are shown. At b) the overlap of the visual image and the thermal image is shown by precisely defining the origin of two local coordinate systems (x, y) in the middle of the images. At c) the transformation to a common coordinate system (X, Y) is shown, resulting in an enhanced thermal image as a fine superposition of the two images. The development of the mapping method provides a fine superposition of the two images, resulting in an enhanced thermal image. A mapping table is generated to assign pixels of the visual image to pixels of the IR image or vice versa. The mapping table gives the temperature of each pixel of the visual image in the area overlapping with the IR camera image and is generally applicable to the selected type of camera.
[0065] Figure 3 : An illustrative representation of detecting corresponding pixels in two images by applying an image mapping method is shown.
[0066] Figure 4 : An extract of a mapping table is shown. It should be noted that visual images usually have a higher resolution than thermal images. Therefore, multiple pixels of the visual image are assigned to the same pixel of the IR image.
[0067] therefore, Figures 2 to 4 Actually represents the calibration step, where Figure 2 We outline a mapping method developed to provide a detailed overlay of heat on visual images, where the local coordinate systems of the two images are transformed to a common coordinate system taking into account the field of view and camera offset. Figure 3 Shown by Figure 4 An exemplary mapping table shown in is a schematic diagram of how visual image pixels are assigned to IR image pixels and vice versa.
[0068] Figure 5 An example of an enhanced thermal image is shown, where a black rectangular object of interest has approximately the same temperature as the environment: at a) the black object is not fully captured by the thermal IR image, while at b) the black object is fully captured by the thermal IR image. Figure 5 It is outlined how enhanced thermal images can support placing the camera in the right position during commissioning so that all relevant components are fully captured. In particular, if the electrical asset is not in operation, the temperature of all components is approximately the same. Therefore, if the temperature of all components or phases is to be monitored, thermal images do not provide any additional information. Even the commissioning of unknown electrical assets is therefore easily handled, as enhanced thermal images provide information if all relevant components are fully captured. This is also a benefit for retrofit applications.
[0069] Figure 6An example of definition of regions of interest is shown: at a) an enhanced thermal image with three different objects of interest is shown. At b) the detection of pixels capturing objects of interest in the visual image (left) and determining the corresponding IR image pixels in the thermal image (right) by a mapping table is shown.
[0070] The user can identify the type, topology and compartments of the switchgear from the enhanced thermal image. Alternatively, object recognition algorithms can be used for automatic identification. Automatic identification of type, topology and compartments as well as components utilizes object recognition algorithms using image processing methods. Regions of interest are defined by the user and / or automatically by key components or phases, which can be any geometric 2D objects in the visual image, such as Figure 6 The corresponding IR pixels are derived from the mapping table to obtain the temperature information of the area of interest.
[0071] Figure 7 An example of temperature feature extraction is shown. The maximum temperature is determined for three defined areas of interest. Alternatively, another temperature feature can be extracted, such as the minimum temperature and the average temperature, but if the corresponding areas are defined in the visual image, the ambient temperature can also be extracted. Therefore, temperature features of the areas of interest, such as the maximum / minimum temperature, the average temperature and the ambient temperature can be easily extracted. It should be noted that the system can use a low spatial resolution IR camera (e.g. 32x32 pixels) in combination with a higher spatial resolution visual camera (e.g. USB camera with 480x640 resolution), both of which are cost-effective.
Claims
1. A switchgear monitoring system, the system comprising: - Infrared camera; -Vision camera; - processing unit; as well as - output unit; wherein the infrared camera is configured to acquire an infrared image of the switchgear; wherein the visual camera is configured to acquire a visual image of the switchgear; wherein the resolution of the infrared camera is smaller than the resolution of the visual camera; wherein the processing unit is configured to utilize an object recognition algorithm to identify at least one feature of the switchgear in the visual image of the switchgear; wherein the processing unit is configured to map a plurality of pixels in the visual image of the at least one identified feature of the switch device to a corresponding plurality of pixels in the infrared image, wherein the processing unit is configured to: map the plurality of pixels in the visual image of the at least one identified feature of the switch device to the corresponding plurality of pixels in the infrared image using a field of view of the visual camera and a field of view of the infrared camera and a mapping table, wherein the field of view of the visual camera at least partially overlaps with the field of view of the infrared camera, wherein the plurality of pixels in the visual image of the at least one identified feature of the switch device comprises a plurality of subsets of pixels, each subset comprising a plurality of pixels, and wherein each subset of pixels in the visual image and the corresponding pixel in the infrared image are associated with image data of a same portion of the switch device in the identified region of interest; wherein the processing unit is configured to determine a combined image comprising utilization of the plurality of pixels in the visual image mapped to the corresponding plurality of pixels in the infrared image; and Wherein, the output unit is configured to output the combined image.
2. The system of claim 1, wherein the infrared camera and the visual camera are disposed in the same housing.
3. The system of claim 2, wherein the housing is configured to be mounted to and removed from the switchgear.
4. The system of any one of claims 1 to 3, wherein the lens of the infrared camera is located near the lens of the visual camera.
5. The system according to any one of claims 1 to 3, wherein the processing unit is configured to map the multiple pixels in the visual image to the multiple pixels in the infrared image using the range of the line of sight of the visual camera and the line of sight of the infrared camera.
6. The system according to any one of claims 1 to 3, wherein the processing unit is configured to map the multiple pixels in the visual image to the multiple pixels in the infrared image using a distance between a detector of the visual camera and a detector of the infrared camera. 7 . The system according to claim 1 , wherein the processing unit is configured to map the plurality of pixels in the visual image to the plurality of pixels in the infrared image using a mapping table.
8. The system according to any one of claims 1 to 3, wherein the processing unit is configured to detect at least a portion of the switching device, the detecting comprising processing the combined image by the processing unit using an object recognition algorithm.
9. The system of claim 8, wherein the processing unit is configured to detect one or more objects that are not part of the switchgear, the detecting comprising processing the combined image by the processing unit using the object recognition algorithm.
10. The system of claim 8, wherein the detecting of the one or more portions of the switching device comprises utilizing, by the processing unit, input from a user identifying at least one characteristic of the switching device.
11. The system of claim 10, wherein the at least one characteristic of the switchgear comprises a topology and / or a compartment and / or a phase of the switchgear.
12. The system of claim 8, wherein the processing unit is configured to assign at least one temperature to at least one detected portion of the switchgear, the at least one detected portion comprising a utilization of the infrared image.
