Power station screen cabinet equipment temperature early warning method and device and computer equipment

Abstract

The application relates to a power station screen cabinet equipment temperature early warning method and device, computer equipment, a computer readable storage medium and a computer program product. The method comprises the following steps: collecting a dual-light image of a power station screen cabinet equipment to be analyzed according to a collection parameter corresponding to this time of collection; the dual-light image comprises an infrared thermal imaging image and a visible light image; the dual-light image is subjected to pretreatment, feature extraction treatment, spatial registration treatment and information fusion treatment to generate a fusion image; the fusion image is subjected to temperature field reconstruction treatment and hot spot detection treatment to obtain temperature field distribution information and hot spot area information; and the power station screen cabinet equipment is subjected to temperature analysis according to the temperature field distribution information and the hot spot area information to obtain temperature early warning information of the power station screen cabinet equipment. The method can improve the temperature early warning accuracy of the power station screen cabinet equipment.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of computer, in particular to a power station screen cabinet equipment temperature early warning method and device, computer equipment, computer readable storage medium and computer program product. BACKGROUND

[0002] The power station screen cabinet equipment is a key equipment in the power system, mainly used for control, protection and monitoring of the power system. These devices will generate a large amount of heat during operation, which may cause equipment failure, insulation aging and even fire accidents if the temperature is too high. Therefore, real-time temperature monitoring and early warning of the power station screen cabinet equipment has important practical significance and is an important means to ensure the safe and stable operation of the power system.

[0003] Traditional temperature monitoring methods rely on single temperature sensors or manual inspection, which have problems such as delayed response, insufficient detection accuracy, and difficulty in comprehensive coverage. Especially during the operation of the power station screen cabinet equipment, local overheating or failure may occur, and traditional monitoring methods often fail to effectively identify potential temperature abnormalities, resulting in poor temperature early warning accuracy of the power station screen cabinet equipment. SUMMARY

[0004] Therefore, it is necessary to provide a power station screen cabinet equipment temperature early warning method, device, computer equipment, computer readable storage medium and computer program product capable of improving the temperature early warning accuracy of the power station screen cabinet equipment to solve the above technical problems.

[0005] In a first aspect, the present application provides a power station screen cabinet equipment temperature early warning method, comprising:

[0006] According to the collection parameters corresponding to the current collection, a dual-light image of the power station screen cabinet equipment to be analyzed is collected and acquired; the dual-light image includes an infrared thermal imaging image and a visible light image;

[0007] The dual-light image is preprocessed, feature extraction processed, spatially registered and information fused to generate a fusion image;

[0008] The fusion image is temperature field reconstructed and hot spot detected to obtain temperature field distribution information and hot spot region information;

[0009] According to the temperature field distribution information and the hot spot region information, temperature analysis is performed on the power station screen cabinet equipment to obtain temperature early warning information of the power station screen cabinet equipment.

[0010] In one embodiment, before collecting and acquiring the dual-light image of the power station screen cabinet equipment to be analyzed according to the collection parameters corresponding to the current collection, the method further comprises:

[0011] In the case that the current acquisition is the first acquisition, the acquisition parameter corresponding to the lowest temperature anomaly level is taken as the acquisition parameter of the current acquisition.

[0012] In the case that the current acquisition is not the first acquisition, the acquisition parameter corresponding to the temperature anomaly level determined in the last acquisition is taken as the acquisition parameter of the current acquisition.

[0013] In one embodiment, the preprocessing, feature extraction, spatial registration and information fusion of the dual-light images to generate a fusion image include:

[0014] The dual-light images are preprocessed to obtain preprocessed dual-light images; the preprocessing at least includes quality assessment, image enhancement and region segmentation;

[0015] Multi-scale local features and region features are extracted from the preprocessed dual-light images to construct feature description information;

[0016] According to the feature description information, the preprocessed dual-light images are registered to obtain an affine transformation matrix of the infrared thermal imaging image relative to the visible light image;

[0017] According to the affine transformation matrix, the preprocessed dual-light images are fused to generate a fusion image.

[0018] In one embodiment, the registration of the preprocessed dual-light images according to the feature description information to obtain an affine transformation matrix of the infrared thermal imaging image relative to the visible light image includes:

[0019] According to the region segmentation result of the dual-light images, initial registration points are selected from the feature points corresponding to the feature description information; the region segmentation result is obtained through the region segmentation processing;

[0020] According to the feature description information, candidate registration points are selected from the initial registration points;

[0021] Using a random sample consensus model, the error registration points in the candidate registration points are removed to obtain target registration points;

[0022] According to the target registration points, the affine transformation matrix is calculated and determined.

[0023] In one embodiment, the fusion of the preprocessed dual-light images according to the affine transformation matrix to generate a fusion image includes:

[0024] According to the affine transformation matrix and the infrared thermal imaging image, a registered infrared thermal imaging image is obtained;

[0025] An edge map of the visible light image is obtained, and a temperature gradient map of the registered infrared thermal imaging image is obtained;

[0026] According to the edge map and the temperature gradient map, a fusion weight of the dual-light image is determined;

[0027] According to the fusion weight, the registered infrared thermal imaging image and the visible light image are fused to obtain a fused image.

[0028] In one of the embodiments, the temperature field reconstruction processing and the hotspot detection processing are performed on the fused image to obtain temperature field distribution information and hotspot area information, including:

[0029] According to the registered infrared thermal imaging image, initial temperature field distribution information is obtained;

[0030] According to the initial temperature field distribution information, the temperature gradient map, the fused image, a preset reference temperature, and a preset power station screen cabinet equipment area importance weight, the temperature field distribution information is obtained;

[0031] According to the temperature field distribution information and a region segmentation result of the dual-light image, a threshold judgment method is used to determine the hotspot area information; the region segmentation result is obtained through the region segmentation processing.

[0032] In one of the embodiments, according to the temperature field distribution information and the hotspot area information, temperature analysis is performed on the power station screen cabinet equipment to obtain temperature warning information of the power station screen cabinet equipment, including:

[0033] Historical temperature field distribution information and historical hotspot area information corresponding to historical collection are obtained;

[0034] According to the historical temperature field distribution information, the historical hotspot area information, the temperature field distribution information, and the hotspot area information, temperature size change trend prediction information and hotspot area expansion trend information are determined;

[0035] According to the temperature size change trend prediction information, the hotspot area expansion trend information, and the temperature field distribution information, a temperature abnormality level of the current collection is determined;

[0036] According to the temperature abnormality level of the current collection, the temperature size change trend prediction information, the hotspot area expansion trend information, and the temperature field distribution information, temperature warning information for the power station screen cabinet equipment is generated.

[0037] In a second aspect, the present application also provides a power station screen cabinet equipment temperature early warning device, comprising:

[0038] An image acquisition module is configured to acquire a dual-light image of the power station screen cabinet equipment to be analyzed according to the acquisition parameters corresponding to the current acquisition; the dual-light image comprises an infrared thermal imaging image and a visible light image.

[0039] An image fusion module is configured to perform preprocessing, feature extraction processing, spatial registration processing and information fusion processing on the dual-light image to generate a fusion image.

[0040] A temperature detection module is configured to perform temperature field reconstruction processing and hot spot detection processing on the fusion image to obtain temperature field distribution information and hot spot area information.

[0041] A temperature analysis module is configured to perform temperature analysis on the power station screen cabinet equipment according to the temperature field distribution information and the hot spot area information to obtain temperature early warning information of the power station screen cabinet equipment.

[0042] In a third aspect, the present application also provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the following steps when executing the computer program:

[0043] An image acquisition module is configured to acquire a dual-light image of the power station screen cabinet equipment to be analyzed according to the acquisition parameters corresponding to the current acquisition; the dual-light image comprises an infrared thermal imaging image and a visible light image.

[0044] An image fusion module is configured to perform preprocessing, feature extraction processing, spatial registration processing and information fusion processing on the dual-light image to generate a fusion image.

[0045] A temperature detection module is configured to perform temperature field reconstruction processing and hot spot detection processing on the fusion image to obtain temperature field distribution information and hot spot area information.

[0046] A temperature analysis module is configured to perform temperature analysis on the power station screen cabinet equipment according to the temperature field distribution information and the hot spot area information to obtain temperature early warning information of the power station screen cabinet equipment.

[0047] In a fourth aspect, the present application also provides a computer readable storage medium having a computer program stored thereon, wherein the computer program is executed by a processor to implement the following steps:

[0048] An image acquisition module is configured to acquire a dual-light image of the power station screen cabinet equipment to be analyzed according to the acquisition parameters corresponding to the current acquisition; the dual-light image comprises an infrared thermal imaging image and a visible light image.

[0049] An image fusion module is configured to perform preprocessing, feature extraction processing, spatial registration processing and information fusion processing on the dual-light image to generate a fusion image.

[0050] performing temperature field reconstruction processing and hot spot detection processing on the fusion image to obtain temperature field distribution information and hot spot region information;

[0051] performing temperature analysis on the power station panel cabinet equipment according to the temperature field distribution information and the hot spot region information to obtain temperature early warning information of the power station panel cabinet equipment.

[0052] In a fifth aspect, the present application also provides a computer program product comprising a computer program which, when executed by a processor, implements the following steps:

[0053] acquiring a dual-light image of the power station panel cabinet equipment to be analyzed according to the corresponding acquisition parameters of the present acquisition; the dual-light image comprises an infrared thermal imaging image and a visible light image;

[0054] performing preprocessing, feature extraction processing, spatial registration processing and information fusion processing on the dual-light image to generate a fusion image;

[0055] performing temperature field reconstruction processing and hot spot detection processing on the fusion image to obtain temperature field distribution information and hot spot region information;

[0056] performing temperature analysis on the power station panel cabinet equipment according to the temperature field distribution information and the hot spot region information to obtain temperature early warning information of the power station panel cabinet equipment.

[0057] The power station screen cabinet equipment temperature early warning method, device, computer equipment, computer readable storage medium and computer program product, first, according to the collection parameters corresponding to this collection, the double light images of the power station screen cabinet equipment to be analyzed are collected and acquired, wherein the double light images include infrared thermal imaging images and visible light images, the comprehensive perception of the equipment state is realized by combining the image information of the two kinds, the infrared image provides the temperature distribution of the equipment surface, helps to detect temperature abnormities, the visible light image provides the appearance structure and detail information of the equipment, which is helpful for equipment fault positioning, can overcome the problem of limited monitoring range of a single sensor, and ensures full coverage monitoring of the equipment state, in addition, dynamic adjustment of the collection parameters can optimize resource utilization and ensure efficient collection in key areas; then, the double light images are preprocessed, feature extraction processed, spatial registration processed and information fusion processed to generate a fusion image, the collected double light images are preprocessed to improve image quality, reduce noise and enhance contrast, multi-scale local features and regional features are obtained through feature extraction to ensure that the key information in the image is completely retained, the infrared image and the visible light image are accurately aligned using spatial registration technology to ensure the consistency of the data of the two, finally, the two kinds of image information are comprehensively processed through information fusion to generate a fusion image, which enhances the reliability and accuracy of the image data and lays a solid foundation for subsequent temperature field reconstruction and hot spot detection; then, the fusion image is subjected to temperature field reconstruction processing and hot spot detection processing to obtain temperature field distribution information and hot spot region information, which improves the detection ability of the system for temperature abnormities, especially the identification ability for potential risks such as local overheating, and provides key support for temperature early warning; finally, according to the temperature field distribution information and the hot spot region information, temperature analysis is performed on the power station screen cabinet equipment to obtain temperature early warning information of the power station screen cabinet equipment, the overall distribution of the temperature field and the characteristics (such as temperature value, position, size, etc.) of the hot spot region are used to accurately evaluate the temperature abnormality level of the equipment, and temperature early warning information is generated, the early warning information includes abnormal regions that may exist in the equipment and corresponding temperature states, which is helpful for maintenance personnel to take timely measures, thereby effectively reducing the equipment failure risk. In the above method, through the spatial registration technology of the infrared thermal imaging image and the visible light image, the temperature data and the equipment structure information are accurately aligned; through image preprocessing and information fusion, a high-quality fusion image is generated, which provides a reliable basis for temperature field reconstruction, hot spot detection accurately locates the high-temperature region through threshold judgment and region segmentation, quickly identifies potential abnormities, finally based on the temperature distribution and the hot spot information, the temperature abnormality level is accurately evaluated and early warning information is generated, realizing comprehensive monitoring and accurate early warning of the temperature state of the equipment. BRIEF DESCRIPTION OF DRAWINGS

[0058] In order to make the technical solutions in the embodiments of the present application or the related art clearer, the accompanying drawings needed to be used in the description of the embodiments of the present application or the related art will be briefly introduced. Obviously, the accompanying drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other related drawings can be obtained without creative effort based on these drawings.

[0059] Figure 1 A flowchart of a power station screen cabinet equipment temperature early warning method in an embodiment;

[0060] Figure 2 A flowchart of a fused image generation step in an embodiment;

[0061] Figure 3 A schematic diagram of a power station screen cabinet equipment temperature early warning system in an embodiment;

[0062] Figure 4 A structural block diagram of a power station screen cabinet equipment temperature early warning device in an embodiment;

[0063] Figure 5 An internal structure diagram of a computer device in an embodiment. DETAILED DESCRIPTION

[0064] In order to make the technical solutions in the embodiments of the present application or the related art clearer, the accompanying drawings needed to be used in the description of the embodiments of the present application or the related art will be briefly introduced. Obviously, the accompanying drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other related drawings can be obtained without creative effort based on these drawings.

[0065] In an embodiment, as shown in Figure 1 , a power station screen cabinet equipment temperature early warning method is provided, and the embodiment takes the method applied to a terminal as an example for illustration. It can be understood that the method can also be applied to a server, and can also be applied to a system including a terminal and a server, and is realized through the interaction of the terminal and the server. The terminal can be, but is not limited to, various personal computers, notebook computers, smart phones and tablet computers, etc. The server can be a stand-alone physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services. In the embodiment, the method includes the following steps:

[0066] Step S101: According to the acquisition parameters corresponding to the present acquisition, a dual-light image of the power station screen cabinet equipment to be analyzed is acquired.

[0067] The dual-light image includes an infrared thermal imaging image and a visible light image.

[0068] The collection parameters refer to the camera parameters set by the terminal during the collection process, including the temperature range of the infrared camera, the emissivity, and the exposure time and white balance parameters of the visible light camera. These parameters are adjusted and configured according to the operating environment and monitoring requirements of the power plant screen cabinet equipment.

[0069] Exemplarily, the terminal starts the infrared thermal imaging camera and the visible light camera simultaneously through the built-in dual-light camera module, and images the target equipment according to the set collection parameters. The terminal adjusts the collection parameters according to the real-time operating state of the equipment, for example, dynamically adjusts the emissivity of the infrared thermal imaging camera according to the highest temperature range of the equipment surface, to ensure the accuracy of temperature measurement. At the same time, the terminal adjusts the exposure time and white balance parameters of the visible light camera in real time according to the environmental light conditions, to obtain a clear visible light image. Finally, the terminal completes the synchronous collection of dual-light images and transmits them to the subsequent processing module.

[0070] In step S102, the dual-light images are preprocessed, feature extraction processed, spatially registered, and information fused to generate a fused image.

[0071] The preprocessing refers to noise removal, contrast enhancement, brightness adjustment, etc. of the collected dual-light images, to improve the image quality and provide clear and stable input data for subsequent analysis. The feature extraction processing is to extract key local features and regional features from the images, including texture, shape, edge, etc. information, which helps to accurately identify each component of the equipment and potential problem areas. The spatial registration processing refers to aligning the infrared thermal imaging image and the visible light image, so that their temperature information and structural information can be accurately matched, to avoid deviation between the images. The information fusion processing is to fuse the registered images, combining the temperature data of the infrared image and the structural information of the visible light image, to generate the final fused image, providing accurate data basis for subsequent temperature analysis and hot spot detection.

[0072] Exemplarily, the terminal first pre-processes the collected dual-light images, removes noise using Gaussian filtering or median filtering, and enhances image contrast through histogram equalization. Then, using image processing algorithms such as Scale-Invariant Feature Transform (SIFT) or Speeded-Up Robust Features (SURF) algorithm, the terminal extracts important feature points and region descriptors from the pre-processed images, which will serve as the basis for subsequent registration and matching. In the spatial registration stage, the terminal accurately aligns the infrared thermal imaging image and the visible light image through feature point matching algorithms such as Random Sample Consensus (RANSAC) algorithm, ensuring the accurate correspondence of the two images in space. Finally, the terminal merges the registered images through weighted fusion technology, generates a fused image using the temperature information in the infrared image and the structural information in the visible light image, which has higher temperature detection accuracy and structural details.

[0073] In step S103, temperature field reconstruction processing and hot spot detection processing are performed on the fused image to obtain temperature field distribution information and hot spot region information.

[0074] Among them, temperature field reconstruction refers to reconstructing the temperature field of the device surface based on the temperature data in the fused image through interpolation algorithm and heat conduction model. The purpose of temperature field reconstruction is to expand discrete temperature data points into a complete temperature distribution map, which can provide the temperature variation trend of each region of the device surface. Hot spot detection is to identify the region with abnormal temperature, i.e. the hot spot region whose temperature exceeds the preset safety range, by setting a temperature threshold or using regional statistical analysis. The hot spot region information includes the spatial position, area, temperature value and other characteristics of the hot spot, which is crucial for device maintenance and fault warning.

[0075] Exemplarily, the terminal performs interpolation processing (such as Bilinear Interpolation or Kriging Interpolation) on the temperature information of the fused image, smooths the temperature data points in the image, and generates a continuous temperature field distribution map. In addition, the terminal uses Temperature Gradient Analysis and Local Statistical Analysis to identify the region with excessively high temperature in the device, i.e. the hot spot region, by setting a temperature threshold or using a standard deviation-based anomaly detection method (such as Z-score method). Through these algorithms, the terminal can accurately identify and mark the potential fault region in the device, providing key data for the subsequent warning system.

[0076] Step S104, according to the temperature field distribution information and the hotspot area information, temperature analysis is performed on the power station screen cabinet equipment to obtain temperature early warning information of the power station screen cabinet equipment.

[0077] Among them, the temperature analysis refers to evaluating the overall temperature state of the power station screen cabinet equipment based on the temperature field distribution information and the hotspot area information. By analyzing the temperature values of each region on the surface of the equipment and the extension trend of the hotspots, it can be accurately judged whether there is a potential temperature abnormality or failure in the equipment. The temperature early warning information is generated through the analysis result, including the severity of the current temperature abnormality, the location of the abnormal area, the temperature value and the predicted temperature change trend, providing timely and accurate early warning data for the operation and maintenance personnel.

[0078] Illustratively, the terminal first performs global and local analysis on the temperature field distribution information to evaluate the overall temperature state of the equipment and determine whether there is overheating phenomenon on the surface of the equipment. In combination with the hotspot area information, the terminal identifies the local overheating area and calculates whether its temperature exceeds the preset safety threshold (such as the maximum safe temperature of the equipment). In addition, the terminal predicts whether the equipment temperature is likely to continue to rise or fail according to the change trend of the equipment temperature (by comparing the currently collected temperature data with the historical data) and the expansion of the hotspot area, and generates temperature early warning information. The early warning information includes the temperature abnormality level (such as mild, moderate and severe abnormality), the spatial position of the abnormal area and the temperature data, and provides operation suggestions (such as whether to shut down, cool or maintain).

[0079] In the power station screen cabinet equipment temperature early warning method, first, according to the corresponding collection parameters of this collection, the dual-light images of the power station screen cabinet equipment to be analyzed are collected and acquired, wherein the dual-light images include infrared thermal imaging images and visible light images, the comprehensive perception of the equipment state is realized by combining the image information of the two kinds, the infrared image provides the temperature distribution of the equipment surface, helps to detect temperature anomalies, the visible light image provides the appearance structure and detail information of the equipment, which is helpful for equipment fault positioning, can overcome the problem of limited monitoring range of a single sensor, and ensures full coverage monitoring of the equipment state, in addition, dynamic adjustment of the collection parameters can optimize resource utilization and ensure efficient collection in key areas; then, the dual-light images are preprocessed, feature extraction processed, spatial registration processed and information fusion processed to generate a fusion image, the collected dual-light images are preprocessed to improve image quality, reduce noise and enhance contrast, multi-scale local features and regional features are obtained through feature extraction to ensure that the key information in the image is completely retained, the infrared image and the visible light image are accurately aligned using spatial registration technology to ensure the consistency of the data of the two, finally, through information fusion, the two kinds of image information are comprehensively processed to generate a fusion image, which enhances the reliability and accuracy of the image data and lays a solid foundation for subsequent temperature field reconstruction and hot spot detection; then, the fusion image is processed for temperature field reconstruction and hot spot detection to obtain temperature field distribution information and hot spot area information, which improves the detection ability of the system for temperature anomalies, especially the identification ability for potential risks such as local overheating, and provides key support for temperature early warning; finally, according to the temperature field distribution information and the hot spot area information, the temperature of the power station screen cabinet equipment is analyzed to obtain the temperature early warning information of the power station screen cabinet equipment, the overall distribution of the temperature field and the characteristics (such as temperature value, position, size, etc.) of the hot spot area are used to accurately evaluate the temperature anomaly level of the equipment, and the temperature early warning information is generated, the early warning information includes the abnormal area that may exist in the equipment and the corresponding temperature state, which is helpful for maintenance personnel to take timely measures, thereby effectively reducing the equipment failure risk. In the above method, through the spatial registration technology of the infrared thermal imaging image and the visible light image, the temperature data and the equipment structure information are accurately aligned; through image preprocessing and information fusion, a high-quality fusion image is generated, which provides a reliable basis for temperature field reconstruction, and hot spot detection accurately locates the high-temperature area through threshold judgment and region segmentation, quickly identifies potential anomalies, finally based on the temperature distribution and hot spot information, accurately evaluates the temperature anomaly level and generates early warning information, realizes comprehensive monitoring and accurate early warning of the temperature state of the equipment.

[0080] In an exemplary embodiment, before the step S101 of collecting the dual-light images of the power station screen cabinet equipment according to the collection parameters corresponding to the current collection, the method further comprises: in the case that the current collection is the first collection, setting the collection parameters corresponding to the lowest temperature abnormality level as the collection parameters of the current collection; in the case that the current collection is not the first collection, setting the collection parameters corresponding to the temperature abnormality level determined in the last collection as the collection parameters of the current collection.

[0081] The temperature abnormality level is an index classified according to the severity of the temperature state of the equipment. The collection parameters include the setting parameters of the infrared camera and the visible light camera, such as the temperature range of the infrared image, the emissivity, the exposure time, the white balance, etc., and the setting of these parameters directly affects the quality of the collected images and the accuracy of the temperature data.

[0082] For example, the terminal first determines whether the current collection is the first collection. If it is the first collection, the temperature abnormality level is set to the lowest level (i.e. the "normal" or slight abnormality level), and the corresponding collection parameters are set according to the level, such as a lower temperature range of the infrared thermal imaging camera and a standard exposure setting, to ensure that the collection parameters are not too sensitive. For subsequent collections, the terminal adjusts the parameters of the current collection according to the temperature abnormality level determined in the last collection. If the last collection result shows that the temperature abnormality is moderate or severe abnormality, the terminal adjusts the collection parameters, such as increasing the temperature range and increasing the exposure time, in order to capture more detailed temperature information, thereby providing higher-precision data support for subsequent temperature analysis and abnormality detection.

[0083] Specifically, the highest temperature can be used to determine the temperature abnormality level, i.e. the highest temperature reaches which temperature range of the temperature abnormality level, and which temperature abnormality level is adopted. When T < T1, the normal temperature range, the base frequency f0 is adopted; when T1≤T < T2, the slight abnormality, 2 times the base frequency is adopted; when T2≤T < T3, the moderate abnormality, 4 times the base frequency is adopted; when T ≥ T3, the severe abnormality, 8 times the base frequency is adopted; wherein T1, T2, T3 are preset temperature thresholds; for example, T1=45℃, T2=65℃, T3=85℃, f0=5min / time, then the collection frequency is adjusted as follows: normal state, f=f0=5min / time; slight abnormality, f=2f0=2.5min / time; moderate abnormality, f=4f0=1.25min / time; severe abnormality, f=8f0=0.625min / time.

[0084] The parameter adjustment of the dual-light camera also determines the specific parameter value according to real-time feedback. The adjustable parameters of the infrared camera include the temperature measurement range and the emissivity; the adjustable parameters of the visible light camera include the exposure time and the ISO sensitivity.

[0085] Further, the image can be collected 3-5 groups of dual-light images, each group of data including an infrared thermal image, a visible light image, and a collection timestamp.

[0086] In this embodiment, by dynamically adjusting the collection parameters, the collection process can be intelligently optimized according to the real-time temperature state and abnormality level of the device, ensuring the accuracy and reliability of the temperature data, avoiding excessive data collection when the device temperature is normal, reducing unnecessary resource waste, and improving the sensitivity and detection accuracy of the data when the device temperature is abnormal.

[0087] In one exemplary embodiment, as shown in Figure 2 The above step S102 of pre-processing, feature extraction, spatial registration and information fusion of the dual-light image to generate the fusion image can also be implemented by the following steps:

[0088] Step S201, pre-processing the dual-light image to obtain a pre-processed dual-light image.

[0089] Step S202, extracting multi-scale local features and regional features from the pre-processed dual-light image to construct feature description information.

[0090] Step S203, according to the feature description information, performing registration processing on the pre-processed dual-light image to obtain an affine transformation matrix of the infrared thermal image relative to the visible light image.

[0091] Step S204, according to the affine transformation matrix, performing fusion processing on the pre-processed dual-light image to generate a fusion image.

[0092] The pre-processing at least includes quality evaluation processing, image enhancement processing and region segmentation processing.

[0093] The dual-light image pre-processing refers to necessary image processing on the collected infrared thermal image and visible light image to improve the image quality and remove possible noise. Quality evaluation is the process of evaluating the image quality to ensure that the temperature information in the image is clear and reliable, and common indicators such as signal-to-noise ratio and contrast are used for evaluation. Image enhancement includes contrast enhancement, brightness adjustment, edge sharpening, etc., aiming to highlight important information in the image and improve the effect of feature extraction. The region segmentation method (such as superpixel segmentation, edge-based segmentation algorithm, etc.) divides the image into multiple regions, which facilitates independent analysis of the temperature features of different regions in subsequent processing.

[0094] Exemplarily, the terminal performs quality evaluation on the collected dual-light images, uses algorithms such as peak signal-to-noise ratio and structural similarity to evaluate the quality of the images, and removes noise in the images through Gaussian filtering or median filtering. Then, histogram equalization is used to enhance the contrast of the images, so that the temperature information is more prominent, and superpixel segmentation is used to divide the images into regions, so as to extract device regions with different functions.

[0095] In the feature extraction stage, the terminal extracts multi-scale local features and regional features from the preprocessed images. Common feature extraction methods include SIFT and SURF, and regional features such as local binary pattern. These feature description information is used for subsequent image registration and fusion processing.

[0096] In the spatial registration processing stage, the terminal uses feature point matching algorithms (such as SIFT or SURF) to match the infrared thermal imaging images and visible light images, and calculates the affine transformation matrix between them. Through the application of RANSAC algorithm, the matching results are optimized, so as to obtain accurate registration results.

[0097] Finally, the terminal performs information fusion processing on the infrared thermal imaging images and visible light images according to the calculated affine transformation matrix. The information fusion method includes weighted average, pixel-level fusion, etc., to ensure that the final generated fusion image can simultaneously retain temperature data (infrared image) and structural data (visible light image), providing high-quality data support for subsequent temperature field reconstruction and hotspot detection.

[0098] Specifically, ① calculate the quality score of the dual-light images: .

[0099] Wherein, SNR(I) is the signal-to-noise ratio, E(I) is the edge sharpness, C(I) is the contrast, w1, w2, w3 are weight coefficients and w1+w2+w3=1. For example, w1=0.4, w2=0.3, w3=0.3, the signal-to-noise ratio SNR is calculated using the signal-noise separation method based on Gaussian filtering. The edge sharpness is based on the gradient amplitude statistics of the image. The contrast uses the Michelson contrast of the local region.

[0100] Determine whether the quality score Q(I) of the dual-light images is less than the threshold value Q thre , if yes, perform image enhancement processing and then execute region segmentation, if not, directly execute region segmentation. For example, Q thre can be set to 0.6.

[0101] ②Region segmentation is performed on the dual-light image, and the contour of the cabinet equipment is extracted based on the level set method; the image is divided into several sub-regions by using a multi-scale super-pixel segmentation algorithm; key functional regions are identified in combination with prior knowledge, and a region importance weight map W(x, y) = γ1·W temp (x,y)+γ2·W str (x,y) is constructed, where W temp is the temperature importance weight, W str is the structural importance weight, γ1 and γ2 are weight coefficients, and γ1+γ2=1. In combination with the structural design and working principle of the power station equipment, it is determined which regions are crucial to the reliability and stability of the equipment, such as the power module, control panel and heat dissipation area, etc., which are crucial to the normal operation of the equipment, and therefore the weights of these regions are higher. The energy function of the level set method is designed as: E(φ)=μ·E reg (φ)+ν·E area (φ)+λ·E edge (φ), where E reg is a regularization term, E area is a region term, E edge is an edge term, and μ, ν and λ are weight coefficients.

[0102] A physical constraint model is established based on the structural characteristics of the power station cabinet equipment, including the relative position relationship of the equipment components and the spatial distribution law of the key functional regions; the position relationship of the components of the equipment in the image provides spatial constraints for temperature field reconstruction. By modeling the positions of these components, the temperature distribution can be optimized and adjusted according to the known equipment structure. For example, the power module is usually located at the bottom of the equipment or near the heat dissipation area, while the control panel and sensor are usually located at the upper part or middle of the equipment.

[0103] ③The enhancement process of the visible light image includes histogram equalization and contrast enhancement; the enhancement process of the infrared thermal imaging image includes temperature normalization and denoising processing.

[0104] Histogram equalization includes: setting a contrast limit threshold α=3.0, the histogram exceeding the threshold is cropped and redistributed; the image is divided into 8×8 sub-blocks, and histogram equalization is performed on each sub-block, and the inter-block boundary is merged using bilinear interpolation. Contrast enhancement includes: using piecewise linear transformation, gray mapping: g(x)=α·f(x)+β, where α is the contrast gain, ranging from 1.2 to 1.8, β is the brightness adjustment parameter, ranging from -30 to 30, and the parameters are adaptively adjusted according to the statistical characteristics of the image. Temperature normalization is: In(x, y)=(I(x, y)-T min ) / (T max -T min ), T min is the minimum temperature value of the current image, and Tmax is the maximum temperature value of the current image. The denoising process is handled by a median filter.

[0105] ④ Extracting multi-scale local features of dual-light images, including: extracting SIFT feature points and their 128-dimensional descriptors for visible light images; extracting SURF feature points and their 64-dimensional descriptors for infrared images; extracting Harris corner features for both dual-light images;

[0106] Extracting regional features of dual-light images, including: extracting shape descriptors based on HOG operators; extracting temperature distribution pattern features for infrared images, including temperature gradient direction histograms and hotspot distribution features; extracting texture features based on LBP operators; specifically, the hotspot distribution features use threshold segmentation and connected region labeling, and form spatial distribution descriptions including hotspot density maps, hotspot distance statistics, and shape features through centroid calculation.

[0107] Constructing context-aware feature descriptors D: .

[0108] where, is the local feature description, which fuses SIFT / SURF descriptors and Harris corner features; is the neighborhood structure information, which includes HOG shape descriptors, temperature distribution pattern features, and LBP texture features, used to describe the structural characteristics of the region around the feature points; is the constraint term based on prior knowledge, which includes regional importance weights and physical constraint information.

[0109] The feature descriptor D realizes the organic combination of multi-scale and multi-modal features by fusing local feature descriptions (SIFT / SURF and Harris features), context structure information (HOG shape descriptors, temperature distribution features, and LBP texture features), and constraint terms based on prior knowledge (regional importance weights and physical constraint information). This not only ensures the accuracy of feature matching, but also fully utilizes regional structure information and domain expertise, improving the accuracy and reliability of power plant screen cabinet equipment image registration and anomaly detection.

[0110] In this embodiment, through preprocessing, feature extraction, spatial registration, and information fusion processing of dual-light images, the image data can be optimized and accurately aligned, eliminating errors and noise that may occur during image acquisition. By effectively fusing temperature data of infrared images and structural information of visible light images, the final fusion image provides accurate and comprehensive data support for subsequent temperature field reconstruction and hotspot detection, significantly improving the precision and accuracy of temperature monitoring. Ultimately, it improves the real-time temperature monitoring capability of power plant screen cabinet equipment, especially in identifying local temperature anomalies and potential fault areas, with high reliability.

[0111] In an exemplary embodiment, the step S203 performs registration processing on the preprocessed dual-light image according to the feature description information to obtain an affine transformation matrix of the infrared thermal imaging image relative to the visible light image, and further comprises: screening initial registration points from the feature points corresponding to the feature description information according to a region segmentation result of the dual-light image; the region segmentation result is obtained through region segmentation; screening candidate registration points from the initial registration points according to the feature description information; using a random sample consensus model to remove false registration points in the candidate registration points to obtain target registration points; and calculating and determining the affine transformation matrix according to the target registration points.

[0112] Exemplarily, the terminal first divides the infrared thermal imaging image and the visible light image into different regions through region segmentation of the dual-light image. In these regions, key feature points (such as corner points, edge points, etc.) are extracted, and initial registration points are screened according to the feature descriptors of each region. These initial registration points are used as the starting point of registration, and candidate registration points with higher matching degree are further screened out through a feature matching algorithm. Then, the terminal uses the RANSAC algorithm for optimization to remove false matching points. RANSAC randomly selects a subset from the candidate points to calculate a preliminary affine transformation matrix, and screens the optimal registration points by verifying the applicability of the transformation matrix to other points, finally selects the target registration points, and calculates the accurate affine transformation matrix. The affine transformation matrix contains transformation parameters such as translation, rotation, and scaling, and can accurately align the infrared thermal imaging image to the visible light image. Through the matrix, the terminal can accurately align the infrared thermal imaging image and the visible light image, ensuring that the subsequent image analysis is based on consistent spatial coordinates, thereby providing a reliable data basis for temperature field reconstruction and hot spot detection.

[0113] Specifically, the visible light image can provide clear edge and structure information, so the visible light image is selected as the reference image and the infrared thermal imaging image is selected as the image to be registered as the reference image for registration.

[0114] ①Coarse matching based on region segmentation result: the matching region is divided using the region feature graph G(V, E), where the vertex set V represents the segmented region and the edge set E represents the spatial adjacency relationship between regions. For example, region connectivity analysis is performed by common edge determination, i.e., whether two regions share boundary pixels, and an adjacency matrix is constructed according to the common edge determination; the priority search region is determined according to the region importance weight W(x, y); in the priority search region, the nearest neighbor search of the feature descriptor D is performed using a k-d tree to obtain an initial matching pair set M1; in the nearest neighbor search, k can take a value of 2, i.e., taking 2 nearest neighbors, the feature dimension of the k-d tree is the dimension of D, and the splitting is performed according to the maximum variance dimension, and the leaf node capacity is 10 feature points at most. In the search, the distance ratio threshold is set to 0.8, and the search radius is limited to 100 pixels.

[0115] ② Perform fine matching based on spatial constraint: calculate the Euclidean distance of feature descriptors as similarity measure; construct spatial constraint SC: .

[0116] wherein, denote two feature point matching pairs to be evaluated, each of which contains one feature point in the reference image and one feature point in the image to be registered, denotes the distance between the feature point pairs, denotes the angle between the line connecting the feature point pairs and the horizontal direction, , are weight coefficients; is distance normalization, i.e. ; is angle normalization, i.e. , , are the maximum distance threshold and the maximum angle threshold, respectively.

[0117] Filter the initial matching pair set M1 based on the spatial constraint SC to obtain the optimized matching pair set M2; for example, set the SC threshold to 0.8, i.e. keep the matching pairs with SC less than 0.8.

[0118] ③ Estimate the affine transformation matrix H using the RANSAC algorithm, and define the confidence evaluation function as follows:

[0119]

[0120] wherein, is the cosine similarity of the feature descriptor D; is the point pair compliance to the affine transformation matrix H; is the prior constraint based on the region importance weight; , , are weight coefficients and .

[0121] For example, the affine transformation matrix estimation process is: random sampling: select 3 matching points each time; calculate the affine matrix: 6 degrees of freedom; inlier statistics: calculate the projection error.

[0122] Filter the final matching pair set M3 according to the Score threshold, and use the final matching pair set M3 to calculate the affine transformation matrix to complete image registration. For example, the Score threshold can be set to 0.7, i.e. keep the matching pairs with Score>0.7 as the final matching pair set M3.

[0123] IV. Then, the parameters of the affine transformation matrix can be solved by using the least square method, and then the pixels of the infrared thermal imaging image are transformed to the pixel coordinate system of the visible light image according to the affine transformation matrix, and the image registration is completed.

[0124] For example, there are n pairs of matching points (xi, yi) and (x i ′, y i ′) (each pair of matching points contains the coordinates of the visible light image and the coordinates of the infrared thermal imaging image). For each pair of matching points, the following two equations can be obtained:

[0125]

[0126] Where a, b, c, d, e, f are the affine transformation parameters to be solved.

[0127] These equations are arranged in matrix form: A·θ=B.

[0128]

[0129] Since the equation set can be overdetermined (i.e. 2N>6), the least square method is needed to solve the parameter vector θ. The formula of the least square solution is: T is the matrix transpose.

[0130] Or it can also be solved by using QR decomposition (orthogonal triangular decomposition), singular value decomposition (SVD) and other methods.

[0131] Through the priority search mechanism guided by the region feature graph G(V, E) and the fine matching process based on the spatial constraint SC, combined with the region importance weight W(x, y) and the device physical characteristic constraint, multiple constraints and optimization of the registration process are realized; combined with the screening mechanism of the confidence evaluation function Score, both the registration accuracy and the calculation complexity are reduced, which is especially suitable for the scene of power station screen cabinet equipment with specific structural characteristics, and has strong robustness and practicality.

[0132] In this embodiment, high-precision image registration is realized by combining region segmentation, feature description information screening and RANSAC optimization algorithm. Through the screening of the initial registration points, the screening of the candidate registration points and the optimization of the RANSAC algorithm, the accurate alignment of the infrared thermal imaging image and the visible light image is ensured, and the registration error caused by different image shooting angles or other factors is eliminated. The finally generated affine transformation matrix provides an accurate spatial coordinate basis for subsequent temperature field reconstruction and hot spot detection, thereby significantly improving the accuracy and reliability of temperature monitoring.

[0133] In an exemplary embodiment, the step S204 generates the fused image by fusing the preprocessed dual-light images according to the affine transformation matrix, further comprising: obtaining a registered infrared thermal imaging image according to the affine transformation matrix and the infrared thermal imaging image; obtaining an edge map of the visible light image and a temperature gradient map of the registered infrared thermal imaging image; determining a fusion weight of the dual-light images according to the edge map and the temperature gradient map; and fusing the registered infrared thermal imaging image and the visible light image according to the fusion weight to obtain the fused image.

[0134] Exemplarily, the terminal first aligns the infrared thermal imaging image using the affine transformation matrix, and accurately registers it to the spatial coordinate system of the visible light image. Then, edge information is extracted by applying edge detection to the visible light image, and the edge map is generated by the edge information in the visible light image, which can highlight the obvious structure boundaries and contours of the device surface. The edge information plays an important role in the fusion process, and is usually used to determine which areas need high weight. The temperature gradient of the infrared image is calculated to generate the temperature gradient map. Then, the terminal calculates the weight of each image area by a weighted fusion algorithm, and the hotspot area and the edge area are given a higher weight. The edge area has a higher structural importance, and the area with a larger temperature gradient indicates temperature anomaly, ensuring that the key information of these areas is fully reflected in the final fused image. Finally, based on these fusion weights, the terminal fuses the infrared thermal imaging image and the visible light image to generate a high-quality fused image. Common fusion methods include weighted averaging method, that is, adjusting the contribution proportion of the infrared image and the visible light image according to the importance (weight) of the area, so that the final fused image can retain both structural information (from the visible light image) and temperature information (from the infrared image). For example, the temperature information of the hotspot area may depend more on the infrared image, while the edge area may depend more on the visible light image. The image combines temperature information and structural information, providing accurate data basis for subsequent temperature field reconstruction and hotspot detection.

[0135] Specifically, the registered infrared thermal imaging image IR and the visible light image VIS are normalized to unify the image value range to the interval [0, 1], unify the numerical range of the two images, and eliminate the dimensional difference.

[0136] The Sobel edge map Ed(x, y) of the visible light image is calculated, and the Sobel operator calculation is as follows:

[0137] Horizontal direction gradient: ;

[0138] Vertical direction gradient: ;

[0139] Edge intensity: The edge map reflects the structural information and contour features in the visible light image.

[0140] The temperature gradient map Tg(x, y) of the infrared thermal image is extracted, the horizontal temperature gradient: ; the vertical temperature gradient: ; the temperature gradient amplitude: ; and the temperature gradient map reflects the degree of temperature change in the infrared image.

[0141] An adaptive weight Z(x, y) is generated , μ1 and μ2 are adjustment coefficients and μ1 + μ2 = 1; when setting, μ1 < μ2 can be allowed, the high edge response area is biased to retain visible light information, and the high temperature gradient area is biased to retain infrared information.

[0142] The fusion image is generated: .

[0143] wherein F(x, y) is the fusion image, IR(x, y) is the normalized infrared thermal image, and VIS(x, y) is the normalized visible light image. By using the fusion formula, more visible light details are retained in the edge significant area, and more infrared information is retained in the area with large temperature gradient, thereby realizing adaptive fusion of structural information and temperature information.

[0144] The dual-light image fusion algorithm realizes complementary advantages of structural details of the visible light image and temperature information of the infrared image by introducing the adaptive weight Z(x, y) to organically combine the Sobel edge map Ed(x, y) and the temperature gradient map Tg(x, y); by the adjustable weight coefficients μ1 and μ2, the system can flexibly adjust the fusion proportion of the structural information and the temperature information according to actual application requirements, thereby ensuring not only the visual quality of the fusion image but also the accurate transmission of the temperature information.

[0145] In the embodiment, by accurate registration of the affine transformation matrix and weighted fusion based on the edge map and the temperature gradient map, a precise fusion image can be generated, which not only retains the temperature distribution information of the power station screen cabinet equipment, but also accurately presents the structural details of the equipment surface. The weighted fusion method ensures that the information of hot spot areas, edge areas and other key areas is preferentially retained, thereby improving the comprehensive information amount and accuracy of the image, providing a reliable basis for subsequent temperature field reconstruction, hot spot detection and temperature warning, and significantly improving the accuracy of equipment monitoring and fault prediction.

[0146] In an exemplary embodiment, the step S103 of performing temperature field reconstruction processing and hot spot detection processing on the fusion image to obtain temperature field distribution information and hot spot region information further comprises: obtaining initial temperature field distribution information according to the registered infrared thermal imaging image; obtaining temperature field distribution information according to the initial temperature field distribution information, the temperature gradient map, the fusion image, a preset reference temperature, and a preset power station screen cabinet equipment region importance weight; determining a hot spot region by using a threshold judgment method according to the temperature field distribution information and a region segmentation result of the dual-light image to obtain hot spot region information; and the region segmentation result is obtained through region segmentation processing.

[0147] Illustratively, the terminal first obtains preliminary temperature field distribution information from the registered infrared thermal imaging image, which shows the temperature changes of each region of the equipment surface. The infrared thermal imaging image provides temperature information of each region of the equipment surface, and after preprocessing and registration, a preliminary temperature field distribution map can be obtained, which presents the rough distribution of the temperature of the equipment surface. Then, combined with the temperature gradient map and the fusion image, an optimized temperature field distribution map is obtained by weighted calculation based on the preset reference temperature and the region importance weight. Considering the different functional areas of the equipment and the importance of each region, the key regions (such as power supply, control panel, etc.) are given higher weights to ensure that they receive more attention in the temperature field analysis. The preset reference temperature provides a standard reference temperature value for the equipment, which is used to compare the actual temperature of the equipment surface to identify temperature abnormal regions. According to the temperature field distribution and the segmentation of the equipment regions, the terminal identifies the regions with excessively high temperature (i.e., the hot spot region) by using the threshold judgment method, and outputs the hot spot region information including the hot spot position, temperature value, and size.

[0148] Specifically, ① temperature field reconstruction is performed based on the fusion image F(x, y), and the reconstruction function is: .

[0149] Wherein, T(x, y) is the reconstructed temperature field distribution function, representing the actual temperature value at point (x, y) in the screen cabinet equipment plane coordinate system; T base is the reference temperature value, which can be specifically set as the ambient temperature; F(x, y) is the fusion image; W(x, y) is the region importance weight; T g (x, y) is the temperature gradient map; div represents the divergence operator; K(x, y) is the correction coefficient, wherein K0 is the reference correction coefficient, which is obtained according to camera calibration, G(x, y) is the normalized temperature gradient amplitude, L(x, y) is the local temperature uniformity, and β1, β2 are weight coefficients.

[0150] Further, the divergence operator acts on the temperature gradient map T g (x, y), and the calculation formula is: where T g (x) and T g (y) are the temperature gradient components in x and y directions respectively, which is used to describe the diffusion characteristics of temperature field in space, reflecting the degree of heat convergence or divergence.

[0151] The calculation of normalized temperature gradient amplitude G(x, y): first calculate the partial derivatives of temperature in x and y directions: T x =∂T / ∂x, T y =∂T / ∂y, calculate the gradient amplitude: , normalized processing: , G(x, y) represents the degree of temperature change, the larger the value, the faster the temperature change.

[0152] The calculation of local temperature uniformity L(x, y): in the local neighborhood N(x, y) of point (x, y), calculate the local standard deviation: , where μ is the local mean; local uniformity: ; L(x, y) describes the smoothness of temperature distribution, the larger the value, the more uniform the temperature distribution.

[0153] The reconstruction function realizes the mapping conversion from image gray value to actual temperature value by combining the reference temperature, fusing image information, regional weight and temperature gradient characteristics, while considering the spatial distribution characteristics of temperature field.

[0154] ②Based on T(x, y), calculate the local statistical characteristics, use threshold method to determine the hot spot area, extract the feature parameters of hot spot area, including the highest temperature, area and shape characteristics, establish the hot spot tracking table, record the position and temperature change.

[0155] The determination standard of hot spot area is: .

[0156] Where, H(x, y) represents the hot spot area; N(x, y) represents the local neighborhood; μ is the local mean; σ is the local standard deviation; κ is the detection coefficient; T thre is the preset temperature threshold.

[0157] Further, in the local neighborhood determination, taking the point (x, y) as the center, combining the region segmentation result described above, considering the physical structure characteristics of the equipment, N(x, y)={(i, j) | (i, j) belongs to the same superpixel region and |i-x|≤r and |j-y|≤r}, r is the search radius, for example, the value is 5-7 pixels, the same superpixel region refers to the result of the multi-scale superpixel segmentation described above, and it is required that (i, j) and (x, y) are in the same functional region. In the calculation, the superpixel region where the point (x, y) is located is determined, a (2r+1)×(2r+1) window is taken in the region, the temperature mean μ and the standard deviation σ in the window are calculated, and it is judged whether T(x, y) exceeds the threshold value.

[0158] In the embodiment, by combining the temperature field reconstruction, the temperature gradient map, the region segmentation, and the hot spot detection method, the accurate monitoring of the temperature state of the power station screen cabinet equipment can be realized. The temperature field distribution information is optimized through comprehensive analysis of multiple factors, especially the attention to the key regions and the high temperature regions, which significantly improves the identification ability of the temperature anomaly. Through the threshold value judgment of the hot spot detection method, the high temperature region in the equipment can be quickly located, and the early warning information can be provided in time, so that the equipment failure or the catastrophic event can be effectively avoided.

[0159] In an exemplary embodiment, the step S104 performs temperature analysis on the power station screen cabinet equipment according to the temperature field distribution information and the hot spot region information to obtain temperature early warning information of the power station screen cabinet equipment, and further includes: obtaining historical temperature field distribution information and historical hot spot region information corresponding to historical collection; determining temperature size change trend prediction information and hot spot region expansion trend information according to the historical temperature field distribution information, the historical hot spot region information, the temperature field distribution information, and the hot spot region information; determining a temperature anomaly level of the current collection according to the temperature size change trend prediction information, the hot spot region expansion trend information, and the temperature field distribution information; and generating temperature early warning information for the power station screen cabinet equipment according to the temperature anomaly level of the current collection, the temperature size change trend prediction information, the hot spot region expansion trend information, and the temperature field distribution information.

[0160] Exemplarily, the terminal analyzes the temperature change trend and the expansion trend of the hotspot area by acquiring historical temperature field distribution and historical data of historical hotspot area, and combining the current temperature field distribution information. The temperature size change trend prediction information is based on the comparison of historical temperature data and current temperature field distribution to predict whether the temperature will rise or fall in the future period of time. The hotspot area expansion trend information is predicted by comparing the temperature of the current hotspot area and the change trend of the historical hotspot area to predict whether the hotspot area will further expand. Then, according to these data, the terminal will calculate the temperature anomaly level and generate corresponding warning information. For example, if the current temperature change trend shows an upward trend and the hotspot area continues to expand, the terminal may issue a "severe anomaly" warning and suggest further cooling or equipment inspection measures. The temperature anomaly level is comprehensively evaluated according to the current collected temperature field distribution and historical temperature field distribution data, combined with temperature size change trend prediction information and hotspot area expansion trend information. By analyzing the speed of temperature change and the expansion of the hotspot area, the temperature anomaly level of this collection can be accurately judged, such as normal, mild anomaly, moderate anomaly, severe anomaly, etc. Finally, according to the determined temperature anomaly level and the temperature change trend, the hotspot area expansion trend and the temperature field distribution information, the temperature warning information is generated. The temperature warning information will clearly indicate the abnormal degree of the current temperature of the equipment, and provide targeted warning response suggestions, such as whether to shut down, whether to cool or other maintenance measures.

[0161] Specifically, the temperature field T(x, y) output by the temperature monitoring module and the hotspot detection result are time-sequenced, a monitoring data sequence {T(x, y, t)} is established, the data is spatially weighted in combination with the area importance weight W(x, y) to generate a regional temperature feature vector, including the maximum temperature, the average temperature, the temperature standard deviation and the hotspot area; then, the sliding time window method is used to calculate the temperature change rate and acceleration, the exponential smoothing prediction temperature trend is used, and the trend evaluation index is established: .

[0162] Where, ΔT / Δt represents the change rate of temperature with time, Δ 2 T / Δt 2 represents the change rate of temperature change rate, P(t) is the temperature prediction deviation, ξ1, ξ2, ξ3 are weight coefficients and ξ1+ξ2+ξ3=1.

[0163] Further, , , the temperature prediction deviation P(t) is the temperature trend predicted by exponential smoothing, , T pre (t) is the exponential smoothing prediction value based on historical data, T act (t) is the actual measurement value.

[0164] Next, based on the multi-level temperature threshold and the trend evaluation index, an abnormality discrimination model is constructed, the discrimination basis includes the temperature overrun degree, the overrun duration, the change trend and the spatial distribution characteristics, and the temperature abnormality level is output. The temperature overrun degree is the comparison between the actual temperature and the preset threshold T1, T2 and T3, the overrun duration is the duration of the abnormal state, the change trend is the numerical value and change direction of the trend evaluation index R(t), and the spatial distribution characteristics include the shape characteristics and expansion trend of the hot spot region. The abnormality recognition unit not only considers the static temperature threshold judgment, but also integrates the dynamic trend analysis result, and through the comprehensive evaluation of these characteristics, the accurate recognition and level division of the device abnormal state are realized, which provides a reliable basis for the early warning decision.

[0165] According to the abnormality recognition result, the early warning information is generated, including the temperature abnormality level, the abnormal region position, the highest temperature value and the temperature change trend, and the early warning information is pushed to the system management module.

[0166] For example, the temperature abnormality level is the same as the level corresponding to the image acquisition module acquisition control unit, which is specifically set as: normal: T < T1; mild abnormality: T1≤T < T2; moderate abnormality: T2≤T < T3; severe abnormality: T≥T3.

[0167] According to the abnormality recognition result, the early warning report containing the temperature abnormality level (normal, mild abnormality, moderate abnormality, severe abnormality), the accurate position coordinates of the abnormal region, the current highest temperature value and the temperature change trend based on the trend evaluation index R(t) and other key information is generated.

[0168] In this embodiment, by combining historical data and real-time temperature data, the temperature change trend of the power station screen cabinet equipment and the expansion of the hot spot region can be accurately predicted, and the accuracy of temperature abnormality detection and early warning is significantly improved. Based on the temperature field distribution information, the historical data and the trend prediction, the temperature abnormality level is generated, and the temperature early warning information is generated in time, which helps the operation and maintenance personnel to quickly identify the potential fault risk of the equipment and take corresponding measures. Not only the early warning ability of equipment failure is improved, but also the damage and accidents caused by temperature abnormality are effectively avoided.

[0169] In an exemplary embodiment, as shown in Figure 3 The present application provides a power station screen cabinet equipment temperature early warning system, which comprises:

[0170] An image acquisition module 301 is used to acquire the dual-light images of the screen cabinet equipment in real time, i.e. infrared thermal imaging images and visible light images;

[0171] An image processing module 302 is used to pre-process, feature extract, spatially register and information fuse the acquired dual-light images to generate a fused image.

[0172] a temperature monitoring module 303 configured to perform temperature field reconstruction and hotspot detection on the fused image;

[0173] an intelligent analysis module 304 configured to perform trend analysis on the output data of the temperature monitoring module, execute abnormality identification, and generate early warning information;

[0174] a system management module 305 configured to implement monitoring data storage, abnormal event recording, early warning information pushing, and remote access control, and provide a human-computer interaction interface.

[0175] The image acquisition module 301 comprises: an infrared image acquisition unit configured to control infrared camera parameters and acquire infrared thermal imaging images; a visible light image acquisition unit configured to control visible light camera parameters and acquire visible light images; and an acquisition control unit configured to receive real-time feedback of the intelligent analysis module 304 through the system management module 305, and realize acquisition triggering and parameter adjustment of the dual-light camera.

[0176] The image processing module 302 comprises: a preprocessing unit configured to perform quality assessment, image enhancement, and region segmentation on the acquired infrared thermal imaging images and visible light images; a feature extraction unit configured to extract multi-scale local features and regional features from the preprocessed dual-light images, and construct feature descriptors; an image registration unit configured to perform registration on the dual-light images using a hierarchical registration strategy; and an information fusion unit configured to perform fusion processing on the registered dual-light images, and generate a fused image.

[0177] The temperature monitoring module 303 comprises: a temperature field reconstruction unit configured to perform temperature field reconstruction based on the fused image; and a hotspot detection unit configured to calculate local statistical features based on the temperature field, perform hotspot region determination using a threshold method, and extract feature parameters of the hotspot region.

[0178] The intelligent analysis module 304 comprises: a data processing unit configured to perform time series arrangement on the temperature field and the hotspot detection result output by the temperature monitoring module, establish a monitoring data sequence, perform spatial weighting on the data in combination with regional importance weights, generate a regional temperature feature vector including the maximum temperature, average temperature, temperature standard deviation, and hotspot area; a trend analysis unit configured to calculate temperature change rate and acceleration using a sliding time window method, predict temperature trend using exponential smoothing, and establish a trend evaluation index; an abnormality identification unit configured to construct an abnormality discrimination model based on multi-level temperature thresholds and the trend evaluation index, determine the output temperature abnormality level based on the temperature overrun degree, overrun duration, change trend, and spatial distribution characteristics; and an early warning generation unit configured to generate early warning information including the temperature abnormality level, abnormal region position, maximum temperature value, and temperature change trend based on the abnormality identification result, and push the early warning information to the system management module 305.

[0179] The system management module 305 serves as the control center of the entire early warning system, responsible for the storage management of monitoring data, the recording and tracking of abnormal events, and the timely pushing of early warning information, while providing a human-computer interaction interface. The module receives real-time feedback from the intelligent analysis module 304, automatically adjusts the collection frequency of the collection control unit (corresponding to the reference frequency f0, 2 times the reference frequency, 4 times the reference frequency, and 8 times the reference frequency, respectively) according to the temperature abnormality level (T < T1 is the normal temperature range, T1 ≤ T < T2 is mild abnormality, T2 ≤ T < T3 is moderate abnormality, and T ≥ T3 is severe abnormality), and adjusts the parameter configuration of the dual-light camera accordingly. In addition, the module also provides remote access control function, allowing operation and maintenance personnel to check equipment status, adjust system parameters and respond to abnormal early warning at any time, realizing intelligent management and control of the entire early warning system.

[0180] In this embodiment, (1) through adaptive collection control, multi-feature fusion registration, region importance weighted information fusion, and multi-level early warning mechanism based on trend analysis, the accuracy of power plant screen cabinet equipment temperature monitoring is improved, timely early warning of temperature abnormalities is realized, and a strong guarantee is provided for the safe operation of the power system; (2) an adaptive collection frequency adjustment mechanism based on temperature abnormality level is adopted, the collection frequency is dynamically adjusted according to the equipment temperature state, the system resource utilization is optimized while ensuring the monitoring effect, the system operation efficiency is improved, and the system response speed to abnormal conditions is enhanced; (3) by constructing a feature descriptor containing local features, context information and prior constraints, combined with a hierarchical registration strategy, the accuracy and robustness of dual-light image registration are improved, effectively solving the problem of image registration difficulty in complex environments; (4) the adaptive fusion method based on region importance weight and temperature gradient information makes full use of the complementary characteristics of dual-light images, improves the quality of fused images, and provides a reliable data basis for subsequent temperature monitoring; (5) the temperature field reconstruction model considering local statistical features is adopted, combined with the hotspot detection method based on region importance weight, the accuracy of temperature monitoring is improved, and the temperature abnormal area can be effectively identified and located; (6) the multi-level early warning mechanism based on time series data analysis realizes early warning of temperature abnormalities through trend evaluation indicators and multi-dimensional discriminant models, improves the accuracy and timeliness of system early warning, and provides effective decision basis for equipment maintenance.

[0181] It should be understood that although the steps in the flowcharts involved in the embodiments described above are shown in sequence according to the arrows, the steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified herein, the execution of the steps is not strictly limited in sequence, and the steps can be executed in other orders. Moreover, at least some of the steps in the flowcharts involved in the embodiments described above can include multiple steps or multiple stages, which are not necessarily executed at the same time, but can be executed at different times, and the execution order of the steps or stages is not necessarily sequential, but can be alternately or alternately executed with at least part of other steps or steps or stages in other steps.

[0182] Based on the same inventive concept, the embodiments of the present application also provide a power station screen cabinet equipment temperature early warning device for implementing the power station screen cabinet equipment temperature early warning method described above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme described in the above method, so the specific limitations in one or more power station screen cabinet equipment temperature early warning device embodiments provided below can refer to the limitations of the power station screen cabinet equipment temperature early warning method described above, which will not be repeated here.

[0183] In one exemplary embodiment, as shown in Figure 4 A power station screen cabinet equipment temperature early warning device is provided, comprising: an image acquisition module 401, an image fusion module 402, a temperature detection module 403, and a temperature analysis module 404, wherein:

[0184] The image acquisition module 401 is configured to acquire a dual-light image of the power station screen cabinet equipment to be analyzed according to the corresponding acquisition parameters of this acquisition; the dual-light image includes an infrared thermal imaging image and a visible light image.

[0185] The image fusion module 402 is configured to perform preprocessing, feature extraction processing, spatial registration processing, and information fusion processing on the dual-light image to generate a fused image.

[0186] The temperature detection module 403 is configured to perform temperature field reconstruction processing and hot spot detection processing on the fused image to obtain temperature field distribution information and hot spot region information.

[0187] The temperature analysis module 404 is configured to perform temperature analysis on the power station screen cabinet equipment according to the temperature field distribution information and the hot spot region information to obtain temperature early warning information of the power station screen cabinet equipment.

[0188] In one of the embodiments, the power station screen cabinet equipment temperature early warning device further comprises a collection parameter determination module, which is configured to, in the case of the current collection being the first collection, take the collection parameter corresponding to the lowest temperature abnormality level as the collection parameter of the current collection; and in the case of the current collection not being the first collection, take the collection parameter corresponding to the temperature abnormality level determined in the last collection as the collection parameter of the current collection.

[0189] In one of the embodiments, the image fusion module 402 is further configured to pre-process the dual-light images to obtain pre-processed dual-light images; the pre-processing at least includes quality evaluation processing, image enhancement processing and region segmentation processing; extract multi-scale local features and region features from the pre-processed dual-light images to construct feature description information; perform registration processing on the pre-processed dual-light images according to the feature description information to obtain an affine transformation matrix of the infrared thermal imaging image relative to the visible light image; and perform fusion processing on the pre-processed dual-light images according to the affine transformation matrix to generate a fusion image.

[0190] In one of the embodiments, the image fusion module 402 is further configured to, according to the region segmentation result of the dual-light images, screen out initial registration points from the feature points corresponding to the feature description information; the region segmentation result is obtained through the region segmentation processing; according to the feature description information, screen out candidate registration points from the initial registration points; use a random sample consensus model to eliminate false registration points in the candidate registration points to obtain target registration points; and calculate and determine the affine transformation matrix according to the target registration points.

[0191] In one of the embodiments, the image fusion module 402 is further configured to, according to the affine transformation matrix and the infrared thermal imaging image, obtain a registered infrared thermal imaging image; obtain an edge map of the visible light image and a temperature gradient map of the registered infrared thermal imaging image; determine a fusion weight of the dual-light images according to the edge map and the temperature gradient map; and perform fusion processing on the registered infrared thermal imaging image and the visible light image according to the fusion weight to obtain a fusion image.

[0192] In one of the embodiments, the temperature detection module 403 is further configured to, according to the registered infrared thermal imaging image, obtain initial temperature field distribution information; according to the initial temperature field distribution information, the temperature gradient map, the fusion image, a preset reference temperature and a preset power station screen cabinet equipment region importance weight, obtain temperature field distribution information; and according to the temperature field distribution information and a region segmentation result of the dual-light images, determine hot spot region information by using a threshold judgment method; the region segmentation result is obtained through the region segmentation processing.

[0193] In one of the embodiments, the temperature analysis module 404 is further configured to acquire historical temperature field distribution information and historical hotspot area information corresponding to historical acquisition; determine temperature size change trend prediction information and hotspot area expansion trend information according to the historical temperature field distribution information, the historical hotspot area information, the temperature field distribution information and the hotspot area information; determine a temperature anomaly level of the current acquisition according to the temperature size change trend prediction information, the hotspot area expansion trend information and the temperature field distribution information; and generate temperature early warning information for the power station screen cabinet equipment according to the temperature anomaly level of the current acquisition, the temperature size change trend prediction information, the hotspot area expansion trend information and the temperature field distribution information.

[0194] The modules in the power station screen cabinet equipment temperature early warning device can be implemented by software, hardware or a combination thereof. The modules can be embedded in or independent of the processor in the computer device in hardware form, or stored in the memory in the computer device in software form, so as to be called and executed by the processor to perform the operations corresponding to the modules.

[0195] In one exemplary embodiment, a computer device, which can be a terminal, is provided, and an internal structure diagram of the computer device can be as shown in FIG. 1. Figure 5 The computer device includes a processor, a memory, an input / output interface, a communication interface, a display unit and an input device. The processor, the memory and the input / output interface are connected through a system bus, and the communication interface, the display unit and the input device are connected to the system bus through the input / output interface. The processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operating system and the computer program in the non-volatile storage medium to run. The input / output interface of the computer device is configured to exchange information between the processor and external devices. The communication interface of the computer device is configured to perform wired or wireless communication with external terminals. The wireless communication can be achieved through WIFI, mobile cellular network, near field communication (NFC) or other technologies. The computer program is executed by the processor to implement a power station screen cabinet equipment temperature early warning method. The display unit of the computer device is configured to form a visually visible picture, which can be a display screen, a projection device or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen. The input device of the computer device can be a touch layer overlaid on the display screen, or a key, trackball or touchpad arranged on the shell of the computer device, or an external keyboard, touchpad or mouse, etc.

[0196] Those skilled in the art can understand that Figure 5The structure shown in the figure is only a block diagram of part of the structure related to the scheme of the present application, and does not constitute a limitation on the computer device to which the scheme of the present application is applied. The specific computer device can include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0197] In an embodiment, a computer device is also provided, including a memory and a processor, the memory storing a computer program, and the processor implementing the steps in the above method embodiments when executing the computer program.

[0198] In an embodiment, a computer readable storage medium is provided, storing a computer program, and the computer program implementing the steps in the above method embodiments when executed by a processor.

[0199] In an embodiment, a computer program product is provided, including a computer program, and the computer program implementing the steps in the above method embodiments when executed by a processor.

[0200] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in the present application are all information and data authorized by the user or authorized by all parties, and the collection, use and processing of related data need to comply with relevant regulations.

[0201] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer readable storage medium, and when executed, can include the processes of the above-mentioned embodiment methods. Any reference to memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile memory and volatile memory. The non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical storage, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. The volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration but not limitation, the RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc. The database involved in the embodiments provided in the present application can include at least one of a relational database and a non-relational database. The non-relational database can include a distributed database based on a block chain, etc., without being limited thereto. The processor involved in the embodiments provided in the present application can be a general-purpose processor, a central processing unit, a graphics processing unit, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, an artificial intelligence (AI) processor, etc., without being limited thereto.

[0202] The technical features of the above embodiments can be combined in any manner. To make the description concise, all possible combinations of the technical features in the above embodiments are not described, but as long as the combinations of the technical features do not exist contradictions, they should be considered as the scope of the present application.

[0203] The above-described embodiments are merely illustrative of several embodiments of the present application, and the description is relatively specific and detailed, but should not be understood as a limitation on the scope of the patent. It should be noted that for those skilled in the art, without departing from the concept of the present application, a number of modifications and improvements can be made, which are all within the scope of the present application. Therefore, the scope of protection of the present application should be subject to the appended claims.

Claims

1. A power station cubicle equipment temperature early warning method, characterized in that, The method comprises: In the case of the current collection being the first collection, the collection parameter corresponding to the lowest temperature anomaly level is taken as the collection parameter of the current collection; in the case of the current collection not being the first collection, the collection parameter corresponding to the temperature anomaly level determined in the last collection is taken as the collection parameter of the current collection; According to the collection parameter corresponding to the current collection, a dual-optical image of the power station screen cabinet equipment to be analyzed is collected and acquired; the dual-optical image comprises an infrared thermal imaging image and a visible light image; The dual-optical image is preprocessed to obtain a preprocessed dual-optical image; the preprocessing at least comprises quality evaluation processing, image enhancement processing and region segmentation processing; multi-scale local features and region features are extracted from the preprocessed dual-optical image to construct feature description information; according to the region segmentation result of the dual-optical image, initial registration points are screened out from feature points corresponding to the feature description information; the region segmentation result is obtained through the region segmentation processing; according to the feature description information, candidate registration points are screened out from the initial registration points; an iterative closest point model is used to eliminate false registration points in the candidate registration points to obtain target registration points; according to the target registration points, an affine transformation matrix of the infrared thermal imaging image relative to the visible light image is calculated and determined; according to the affine transformation matrix and the infrared thermal imaging image, a registered infrared thermal imaging image is obtained; through edge information in the visible light image, an edge map of the visible light image is obtained, and temperature gradient calculation is performed on the registered infrared thermal imaging image to obtain a temperature gradient map of the registered infrared thermal imaging image; according to the edge map and the temperature gradient map, a fusion weight of the dual-optical image is determined; according to the fusion weight, the registered infrared thermal imaging image and the visible light image are fused to obtain a fusion image; The fusion image is subjected to temperature field reconstruction processing and hot spot detection processing to obtain temperature field distribution information and hot spot region information; the temperature field distribution information is determined according to initial temperature field distribution information, a temperature gradient map of the registered infrared thermal imaging image, a fusion image, a preset reference temperature and a preset power station screen cabinet equipment region importance weight; the initial temperature field distribution information is obtained according to the registered infrared thermal imaging image; According to the temperature field distribution information and the hot spot region information, temperature analysis is performed on the power station screen cabinet equipment to obtain temperature early warning information of the power station screen cabinet equipment.

2. The method of claim 1, wherein, The temperature field reconstruction processing and the hot spot detection processing are performed on the fusion image to obtain the temperature field distribution information and the hot spot region information, which comprises: According to the registered infrared thermal imaging image, initial temperature field distribution information is obtained; According to the initial temperature field distribution information, the temperature gradient map, the fusion image, a preset reference temperature and a preset power station screen cabinet equipment region importance weight, the temperature field distribution information is obtained; According to the temperature field distribution information and the region segmentation result of the dual-light image, a hot spot region information is determined by using a threshold judgment method; the region segmentation result is obtained by the region segmentation processing.

3. The method of claim 1, wherein, According to the temperature field distribution information and the hot spot region information, temperature analysis is performed on the power station screen cabinet equipment to obtain temperature warning information of the power station screen cabinet equipment, including: Obtain historical temperature field distribution information and historical hot spot region information corresponding to historical collection; According to the historical temperature field distribution information, the historical hot spot region information, the temperature field distribution information and the hot spot region information, determine temperature size change trend prediction information and hot spot region expansion trend information; According to the temperature size change trend prediction information, the hot spot region expansion trend information and the temperature field distribution information, determine the temperature abnormality level of the current collection; According to the temperature abnormality level of the current collection, the temperature size change trend prediction information, the hot spot region expansion trend information and the temperature field distribution information, generate temperature warning information for the power station screen cabinet equipment.

4. The method of claim 1, wherein, The hot spot region information includes the spatial position, area and temperature value of the hot spot.

5. The method of claim 1, wherein, The temperature warning information includes the severity of the current temperature abnormality, the position of the abnormal region, the temperature value and the predicted temperature change trend.

6. A power plant cubicle equipment temperature early warning device, characterized in that, The device includes: The collection parameter determination module is configured to, in a case where the current collection is the first collection, determine the collection parameter corresponding to the lowest temperature abnormality level as the collection parameter of the current collection; in a case where the current collection is not the first collection, determine the collection parameter corresponding to the temperature abnormality level determined in the last collection as the collection parameter of the current collection; The image collection module is configured to collect and obtain a dual-light image of the power station screen cabinet equipment to be analyzed according to the collection parameter corresponding to the current collection; the dual-light image includes an infrared thermal imaging image and a visible light image; The image fusion module is configured to: pre-process the dual-light image to obtain a pre-processed dual-light image; the pre-processing at least includes quality evaluation processing, image enhancement processing, and region segmentation processing; extract multi-scale local features and region features from the pre-processed dual-light image to construct feature description information; according to a region segmentation result of the dual-light image, screen out initial registration points from feature points corresponding to the feature description information; the region segmentation result is obtained through the region segmentation processing; according to the feature description information, screen out candidate registration points from the initial registration points; use a random sample consensus model to eliminate false registration points in the candidate registration points to obtain target registration points; according to the target registration points, calculate and determine an affine transformation matrix of the infrared thermal imaging image relative to the visible light image; according to the affine transformation matrix and the infrared thermal imaging image, obtain a registered infrared thermal imaging image; through edge information in the visible light image, obtain an edge map of the visible light image, and perform temperature gradient calculation on the registered infrared thermal imaging image to obtain a temperature gradient map of the registered infrared thermal imaging image; according to the edge map and the temperature gradient map, determine a fusion weight of the dual-light image; according to the fusion weight, perform fusion processing on the registered infrared thermal imaging image and the visible light image to obtain a fusion image. The temperature detection module is configured to: perform temperature field reconstruction processing and hot spot detection processing on the fusion image to obtain temperature field distribution information and hot spot region information; the temperature field distribution information is determined according to initial temperature field distribution information, a temperature gradient map of a registered infrared thermal imaging image, a fusion image, a preset reference temperature, and a preset power station cabinet equipment region importance weight; the initial temperature field distribution information is obtained according to the registered infrared thermal imaging image; The temperature analysis module is configured to: perform temperature analysis on the power station cabinet equipment according to the temperature field distribution information and the hot spot region information to obtain temperature early warning information of the power station cabinet equipment.

7. The apparatus of claim 6, wherein, The temperature detection module is further configured to: obtain initial temperature field distribution information according to the registered infrared thermal imaging image; and obtain the temperature field distribution information according to the initial temperature field distribution information, the temperature gradient map, the fusion image, a preset reference temperature, and a preset power station cabinet equipment region importance weight. According to the temperature field distribution information and the region segmentation result of the dual-light image, a threshold judgment method is used to determine the hot spot region information. The region segmentation result is obtained through the region segmentation processing.

8. The apparatus of claim 6, wherein, The temperature analysis module is further configured to acquire historical temperature field distribution information and historical hotspot area information corresponding to historical collection; determine temperature size change trend prediction information and hotspot area expansion trend information according to the historical temperature field distribution information, the historical hotspot area information, the temperature field distribution information and the hotspot area information; determine a temperature anomaly level of the current collection according to the temperature size change trend prediction information, the hotspot area expansion trend information and the temperature field distribution information; and generate temperature early warning information for the power station screen cabinet equipment according to the temperature anomaly level of the current collection, the temperature size change trend prediction information, the hotspot area expansion trend information and the temperature field distribution information. 9.A computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the computer device is configured to perform the method according to any one of claims 1-8 when the computer program is executed by the processor. The computer program, when executed by the processor, implements the steps of the method of any one of claims 1 to 5.

10. A computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program, when executed by the processor, implements the steps of the method of any one of claims 1 to 5.

Images