Electric power fault diagnosis and early warning system based on infrared spectrum

Abstract

The invention discloses a power failure diagnosis and early warning system based on an infrared spectrum, which relates to the related field of image data processing, and comprises the following steps: performing infrared scanning monitoring on a power equipment area, and determining an infrared scanning image; and carrying out filtering noise reduction and gray processing on the infrared scanning image to determine an electric power infrared image. And a corner window determined based on a pixel structure tensor is used for determining local area image positioning, and a power fault diagnosis module is supervised and trained through transfer learning. And performing structure tensor calculation on the electric power infrared image pixel by pixel through the electric power fault diagnosis module, positioning a corner window image, performing fault feature judgment, and determining a fault diagnosis list. And traversing the fault diagnosis list, performing fault marking in the power infrared image, and performing display early warning of the power fault on a terminal interface. The technical problems that in the prior art, a power fault detection system cannot accurately recognize and prioritize faults in real time, so that maintenance response is not timely, and the system operation risk is increased are solved.

Description

Technical Field

[0001] This application relates to the field of image data processing, and particularly to a power failure diagnosis and early warning system based on infrared spectra. Background Art

[0002] With the continuous expansion of the scale of the power system and the increase in the complexity of power equipment, ensuring the reliable operation of power equipment has become particularly important. Traditional power failure detection methods mainly rely on manual inspections and basic monitoring technologies, which have the disadvantages of untimely detection, low accuracy, and inability to provide real-time early warnings. Therefore, there is an urgent need for an intelligent early warning system that can monitor and diagnose power equipment failures in real time and accurately, and effectively prioritize and handle failures.

[0003] Therefore, in the prior art, the power failure detection system cannot identify and prioritize failures in real time and accurately, resulting in technical problems such as untimely maintenance response and increased system operation risks. Summary of the Invention

[0004] This application provides a power failure diagnosis and early warning system based on infrared spectra, which solves the technical problems in the prior art that the power failure detection system cannot identify and prioritize failures in real time and accurately, resulting in untimely maintenance response and increased system operation risks. It achieves the technical effects of being able to detect and classify power failures in real time and accurately, and through fault feature clustering and priority ranking, realizing an efficient maintenance response, and improving the operation reliability and safety of the power system.

[0005] This application provides a power failure diagnosis and early warning system based on infrared spectra. The system includes: a scanned image acquisition module for performing infrared scanning monitoring on the power equipment area to determine an infrared scanned image; an infrared image acquisition module for performing filtering, noise reduction, and grayscale processing on the infrared scanned image based on a power data interface to determine a power infrared image; a fault analysis module for determining local image positioning with a corner window determined based on a pixel structure tensor, and supervising and training a power failure diagnosis module through transfer learning; a power failure diagnosis module for transmitting the power infrared image back to the power failure diagnosis module, calculating the structure tensor pixel by pixel, positioning the corner window image, and performing fault feature determination to determine a fault diagnosis sheet; a display and early warning module for traversing the fault diagnosis sheet, marking faults in the power infrared image, and displaying and warning of power failures on a terminal interface.

[0006] In a possible implementation, the fault analysis module is further configured to: for the power equipment area, determine the critical temperature value under standard power operation, generate an infrared reference map; establish a mapping between the infrared reference map and the power equipment area, and based on the mapping relationship, determine the power fault characteristics of the mapped power structure; based on the infrared reference map and the power fault characteristics, perform sample-driven training to generate the power fault diagnosis module.

[0007] In a possible implementation, the fault analysis module is further configured to: determine a diagnostic large model through sample supervised training based on power fault records; introduce the functional relationship between the gray value and the temperature value, perform transfer calls on the diagnostic large model, add a corner reconstruction layer and perform supervised training until convergence to determine the power fault diagnosis module.

[0008] In a possible implementation, the power fault diagnosis module is further configured to: transmit the power infrared image to the power fault diagnosis module, perform corner window determination based on the pixel structure tensor, and reconstruct and determine the corner window image; identify the corner window image, and based on the functional relationship between the gray value and the temperature value as a reference, perform temperature value over-limit determination based on the infrared reference map to locate the abnormal temperature area; for the abnormal temperature area, make a power fault decision based on the mapping relationship and integrate and output the fault diagnosis form.

[0009] In a possible implementation, the power fault diagnosis module is further configured to: traverse the power infrared image to identify a first pixel point, where the first pixel point is any pixel point in the power infrared image; for the first pixel point, combine the neighborhood pixel points, identify and determine a first pixel value matrix; perform corner window determination on the first pixel value matrix, if it is a corner window, retain the first pixel value matrix, if it is not a corner window, screen out the first pixel value matrix.

[0010] In a possible implementation, the power fault diagnosis module is further configured to: determine a preset gradient value, where the preset gradient value is the gradient threshold between pixels; traverse the first pixel value matrix, calculate the neighborhood pixel gradient feature as the structure tensor of the first pixel point, where the structure tensor includes 4 items and is used to quantify the gradient change feature of the pixel point, and the neighborhood pixel gradient feature includes a gradient coefficient and a gradient direction; based on the preset gradient value, perform corner window determination by determining the structure tensor; where, if at least one item of the structure tensor is greater than the preset gradient value, use the first pixel value matrix as the corner window.

[0011] In a possible implementation, the power failure diagnosis module is further configured to: for the power infrared image, determine N corner window candidates pixel by pixel, and determine the corner window image by discretely stitching the N corner window candidates based on the relative distribution of the image.

[0012] In a possible implementation, the display and warning module is further configured to: traverse the fault diagnosis list, determine the fault characteristics based on the fault level and the fault diffusion characteristics, where the fault characteristics include a fault coefficient; perform a homologous clustering process on the fault characteristics to determine M fault clusters, where the faults within a cluster are homologous; traverse the M fault clusters, generate M groups of fault warning messages, where the warning types of each group of fault warning messages are the same, and determine the power operation and maintenance priority based on the fault coefficient; based on the M groups of fault warning messages, perform a display and warning of the power failure on the terminal interface.

[0013] One or more technical solutions provided in this application have at least the following technical effects or advantages: The power failure diagnosis and warning system based on infrared spectra provided in this application includes: a scanned image acquisition module for performing infrared scanning and monitoring on a power equipment area to determine an infrared scanned image; an infrared image acquisition module for filtering, denoising, and gray-level processing on the infrared scanned image based on a power data interface to determine a power infrared image; a fault analysis module for determining local image positioning by using a corner window determined based on a pixel structure tensor, and supervising and training a power failure diagnosis module through transfer learning; a power failure diagnosis module for transmitting the power infrared image back to the power failure diagnosis module, calculating the structure tensor pixel by pixel, positioning the corner window image, and determining the fault characteristics to obtain a fault diagnosis list; a display and warning module for traversing the fault diagnosis list, marking the faults in the power infrared image, and performing a display and warning of the power failure on the terminal interface. This solves the technical problem in the prior art that the power failure detection system cannot identify and prioritize faults in real time and accurately, resulting in untimely maintenance responses and increased system operation risks. It realizes the technical effect of being able to detect and classify power failures in real time and accurately, and through fault characteristic clustering and priority ranking, achieving an efficient maintenance response and improving the operation reliability and safety of the power system. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings of the embodiments of the present invention will be briefly introduced below. Flowcharts are used in this application to illustrate the operations performed by the system according to the embodiments of the present application. It should be understood that the operations before or below do not necessarily need to be executed precisely in sequence. On the contrary, according to the need, they can be executed in reverse order or simultaneously. At the same time, other operations can also be added to these processes, or one or several operations can be removed from these processes.

[0015] Figure 1 This is a schematic structural diagram of the power fault diagnosis and early warning system based on infrared spectra provided by the embodiments of the present application.

[0016] Figure 2 This is a schematic flowchart of the power fault diagnosis module in the power fault diagnosis and early warning system based on infrared spectra of the present application for integrating and outputting the fault diagnosis form.

[0017] Explanation of reference numerals: The scanning image acquisition module 11, the infrared image acquisition module 12, the fault analysis module 13, the power fault diagnosis module 14, and the display and early warning module 15. Detailed implementation manners

[0018] The above description is only an overview of the technical solution of the present application. In order to be able to understand the technical means of the present application more clearly, it can be implemented according to the content of the description. And in order to make the above and other purposes, features and advantages of the present application more obvious and understandable, the following specifically gives the detailed implementation manners of the present application.

[0019] In order to make the purpose, technical solution and advantages of the present application clearer, the present application will be further described in detail below with reference to the accompanying drawings. The described embodiments should not be regarded as limitations of the present application. All other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present application.

[0020] In the following description, "some embodiments" are involved, which describe a subset of all possible embodiments. However, it can be understood that "some embodiments" can be the same subset or different subsets of all possible embodiments, and can be combined with each other without conflict. The terms "first\second" involved are only used to distinguish similar objects and do not represent a specific order for the objects. The terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, system, product or server including a series of steps or units does not necessarily have to be limited to those clearly listed steps or units, but may include other steps or modules not clearly listed or inherent to these processes, systems, products or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the technical field to which the present application belongs. The terms used herein are only for the purpose of describing the embodiments of the present application.

[0021] The embodiments of the present application provide a power fault diagnosis and early warning system based on infrared spectra, as Figure 1 shown, the system includes: The scanning image acquisition module 11 is used to perform infrared scanning monitoring on the power equipment area to determine an infrared scanning image.

[0022] An infrared image acquisition module 12, configured to perform filtering, noise reduction, and grayscale processing on the infrared scanned image based on a power data interface to determine a power infrared image.

[0023] A fault analysis module 13, configured to perform local image positioning determination with a corner window determined based on a pixel structure tensor, and supervise and train a power fault diagnosis module through transfer learning.

[0024] Specifically, a scanned image acquisition module 11 is configured to perform thermal imaging scanning and monitoring on an operation area of a power device through an infrared imaging device to determine an infrared scanned image. The infrared imaging device can detect the temperature distribution on the surface of the device. By capturing a thermal image, abnormal heating areas can be identified, and these areas may indicate potential power faults. Subsequently, through the infrared image acquisition module 12, based on the power data interface, the infrared scanned image is received, and the infrared scanned image is subjected to filtering, noise reduction, and grayscale processing to convert the color infrared image into a grayscale image to simplify the subsequent image analysis process. The grayscale image only contains brightness information, which helps to highlight the differences in temperature distribution, thereby determining the power infrared image. Further, through the fault analysis module 13, local image positioning determination is performed with a corner window determined based on a pixel structure tensor, and a power fault diagnosis module is supervised and trained through transfer learning. The power fault diagnosis module performs fault diagnosis on the corresponding power structure under the abnormal temperature mapping relationship in combination with the mapping relationship between the temperature area and the power device area for the abnormal temperature area.

[0025] Furthermore, the fault analysis module 13 is further configured to: determine a critical temperature value under standard power operation for the power device area, and generate an infrared reference map; establish a mapping between the infrared reference map and the power device area, and determine power fault characteristics of the mapped power structure based on the mapping relationship; perform sample-driven training based on the infrared reference map and the power fault characteristics to generate the power fault diagnosis module.

[0026] The supervised training power fault diagnosis module includes: for the power equipment area, determining the critical temperature value of the heating area of the power equipment under the standard power operation state, where the critical temperature value is the highest temperature allowed in the heating area of the power equipment under the normal operation state of the power equipment. According to the power equipment and the critical temperature value of the heating area of the power equipment, infrared reference maps of each area are generated, and the critical temperatures of different areas reflect different fault characteristics of the equipment. Taking a transformer as an example, by obtaining the critical temperature values of key parts such as the transformer winding, oil tank, and terminal connection, and constructing the corresponding infrared reference maps. When the temperature at a certain place in the transformer winding exceeds the preset critical value, there may be a fault such as winding insulation aging or short circuit at the corresponding position. Further, a mapping relationship between the infrared reference map and the power equipment area is established, and according to the mapping relationship between the power equipment area and the infrared reference map, the power fault characteristics of the mapped power structure are determined. The power fault characteristics of the power structure refer to the critical temperature characteristics shown in the infrared reference map of the heating area of the power equipment. Finally, based on the infrared reference map and the power fault characteristics, sample-driven training is carried out to generate the power fault diagnosis module.

[0027] Further, the fault analysis module 13 is also used to: determine the diagnostic large model through sample supervised training based on power fault records; introduce the functional relationship between the gray value and the temperature value, perform migration calls on the diagnostic large model, add a corner reconstruction layer and perform supervised training until convergence to determine the power fault diagnosis module.

[0028] Generate the power fault diagnosis module, including: through sample supervised training based on power fault records, using fault record samples with labels, training the model through machine learning algorithms to enable it to identify and classify different types of power faults, and obtaining a diagnostic large model. The samples of the power fault records include infrared spectra of various fault samples, as well as corresponding fault category identifiers and abnormal temperature region identifiers. Subsequently, introduce the functional relationship between the gray value and the temperature value, where the functional relationship is the mathematical relationship between the gray value and the actual temperature. Through the functional relationship, the gray information in the image can be accurately converted into temperature information, enhancing the model's ability to identify temperature anomalies. Based on the functional relationship, perform transfer invocation on the diagnostic large model. Taking the pre-trained diagnostic large model as the basis, apply the transfer learning method, add a corner reconstruction layer, and perform supervised training until convergence to determine the power fault diagnosis module. The corner reconstruction layer first performs corner window screening analysis, obtains the heat generation area of the power equipment for image segmentation and reconstruction, and performs gray conversion based on the image reconstruction result of the obtained heat generation area. Further, through the functional relationship between the gray value and the temperature value of the corner window image, perform over-limit determination based on the critical temperature, and obtain the area exceeding the critical temperature. Obtain the infrared abnormal temperature area according to the area exceeding the critical temperature, and combine the mapping relationship between the infrared abnormal temperature area and the power equipment area to determine the corresponding fault diagnosis result. Through the power fault diagnosis module, it is used to obtain the infrared abnormal temperature area according to the power infrared image, and combine the mapping relationship between the infrared abnormal temperature area and the power equipment area to perform fault diagnosis on the corresponding power structure under the abnormal temperature mapping relationship.

[0029] The power fault diagnosis module 14 is used to transmit the power infrared image back to the power fault diagnosis module, perform structure tensor calculation pixel by pixel, locate the corner window image and perform fault feature determination to determine the fault diagnosis list; the display warning module 15 is used to traverse the fault diagnosis list, perform fault marking in the power infrared image, and perform display warning of the power fault on the terminal interface.

[0030] The power failure diagnosis module 14 is used to transmit the power infrared image back to the power failure diagnosis module, calculate the structure tensor for each pixel of the power infrared image, determine the corner window according to the calculation result, and perform fault feature determination to determine the fault diagnosis form. Finally, through the display and warning module 15, traverse the fault diagnosis form, mark the fault in the power infrared image, and display and warn of the power failure on the terminal interface. This solves the technical problem in the prior art that the power failure detection system cannot identify and prioritize faults in real time and accurately, resulting in untimely maintenance response and increased system operation risks. It achieves the technical effect of being able to detect and classify power failures in real time and accurately, and through fault feature clustering and priority sorting, achieve an efficient maintenance response, and improve the operation reliability and safety of the power system.

[0031] Further, as Figure 2 shown, the power failure diagnosis module 14 is further used to: transmit the power infrared image to the power failure diagnosis module, perform corner window determination based on the pixel structure tensor, and reconstruct and determine the corner window image; identify the corner window image, and based on the functional relationship between the gray value and the temperature value, perform temperature limit determination based on the infrared reference map to locate the abnormal temperature area; for the abnormal temperature area, make a power failure decision based on the mapping relationship, and integrate and output the fault diagnosis form.

[0032] The determination of the fault diagnosis form includes: transmitting the power infrared image to the power failure diagnosis module, performing corner window determination based on the pixel structure tensor, obtaining the gradient change feature of each pixel point in its neighborhood, obtaining the corner window, and reconstructing and determining the corner window image. Further, identify the corner window image, and based on the functional relationship between the gray value and the temperature value, perform temperature limit determination of the corner window image based on the infrared reference map to obtain the abnormal temperature area where the temperature value exceeds the limit. Finally, for the abnormal temperature area, obtain the fault type corresponding to the mapping relationship based on the mapping relationship, complete the power failure decision, and integrate and output the fault diagnosis form. The fault diagnosis form contains specific fault categories and abnormal temperature areas.

[0033] Further, the power failure diagnosis module 14 is further used to: traverse the power infrared image, identify the first pixel point, where the first pixel point is any pixel point in the power infrared image; for the first pixel point, combine the neighborhood pixel points, identify and determine the first pixel value matrix; perform corner window determination on the first pixel value matrix, if it is a corner window, retain the first pixel value matrix, if it is not a corner window, screen out the first pixel value matrix.

[0034] Performing corner window determination based on the pixel structure tensor includes: traversing the power infrared image to identify a first pixel point, where the first pixel point is any pixel point within the power infrared image. Subsequently, for the first pixel point, obtain the neighboring pixel points of the first pixel point to obtain a first pixel value matrix. Perform corner window determination on the first pixel value matrix. If it is a corner window, it indicates that the corresponding first pixel value may be an edge pixel value, and then retain the first pixel value matrix. If it is not a corner window, the corresponding first pixel value is not an edge pixel value, and the first pixel value matrix is screened out.

[0035] Furthermore, the power fault diagnosis module 14 is further configured to: determine a preset gradient value, where the preset gradient value is the gradient threshold between pixels; traverse the first pixel value matrix, calculate the gradient features of neighboring pixels as the structure tensor of the first pixel point, where the structure tensor includes 4 items and is used to quantify the gradient change features of the pixel point, and the neighboring pixel gradient features include a gradient coefficient and a gradient direction; based on the preset gradient value, perform corner window determination by determining the structure tensor; where if at least one item of the structure tensor is greater than the preset gradient value, the first pixel value matrix is used as the corner window.

[0036] Performing corner window determination on the first pixel value matrix includes: determining a preset gradient value, where the preset gradient value is the gradient threshold between pixels. When the gradient coefficient is greater than this preset gradient value, the difference between adjacent pixels is relatively large, and the corresponding pixel may be an edge pixel. Traverse the first pixel value matrix, calculate the gradient features of the first pixel value and neighboring pixels as the structure tensor of the first pixel point. Where the structure tensor includes 4 items and is used to quantify the gradient change features of the pixel point, and the neighboring pixel gradient features include a gradient coefficient and a gradient direction. Based on the preset gradient value, by determining whether the structure tensor is greater than the preset gradient value, obtain the first pixel points greater than the preset gradient value to complete the corner window determination. Where if at least one item of the structure tensor is greater than the preset gradient value, the first pixel value matrix is used as the corner window.

[0037] Furthermore, the power fault diagnosis module 14 is further configured to: perform corner determination for each pixel point of the power infrared image to determine N corner windows; based on the relative distribution of the image, perform discrete stitching on the N corner windows to determine the corner window image.

[0038] Reconstruct and determine the corner window image, including: based on the acquired power infrared image, perform corner window determination on the pixel points in the power infrared image, obtain a corner window where at least one item of the structure tensor is greater than the preset gradient value, then the corresponding pixel point is the corner window, obtain all corner windows that meet the requirements, and determine N corner windows. After completing the corner detection of the power infrared image, these scattered corner windows need to be further processed to form a "corner window image" for subsequent fault diagnosis analysis or model input. According to the relative positions of the corner windows in the image, that is, the spatial positions of each corner window in the infrared image coordinate system and the distance relationship between them. Treat each corner window of the N corner windows as a discrete small piece of image, and splice the corner windows in order from left to right, from top to bottom, and when the distance between corner windows is less than the preset distance threshold, then splice the corresponding corner windows in order. Thus, determine the spliced corner window image. These corner window images only retain the information around the most recognizable key points in the image, reducing the interference of a large number of irrelevant regions.

[0039] Further, the display warning module 15 is further configured to: traverse the fault diagnosis list, determine fault features based on the fault level and fault diffusion characteristics, where the fault features include fault coefficients; perform homologous clustering processing on the fault features to determine M fault clusters, where the faults within the clusters are homologous faults; traverse the M fault clusters, generate M groups of fault warning information, where the warning types of each group of fault warning information are the same, determine the power operation and maintenance priority based on the fault coefficients; and perform display warning of power faults on the terminal interface based on the M groups of fault warning information.

[0040] Traverse the fault diagnosis list and check each fault record in the fault diagnosis list one by one. Assume that the fault diagnosis list contains three faults: overheating of transformer connection terminals, poor contact of switchgear, and abnormal temperature of distribution panel. Determine the fault characteristics according to the fault level and fault spread characteristics. Each of the fault types corresponds to a preset specific fault level and fault spread characteristic parameter, that is, the influence level on other devices. Add the two to obtain the fault characteristics. Exemplarily: Overheating of transformer connection terminals may cause equipment damage, with a level of 3. Poor contact of switchgear may cause a short circuit, with a level of 2. Further, group the homologous faults together through a clustering algorithm to form groups. The faults in each cluster have high similarity in characteristics. Determine M fault clusters, where M is a positive integer. Among them, the faults within the cluster are homologous faults, and the homologous faults are faults with the same fault characteristics, including the same fault location and fault coefficient. Further, traverse the M fault clusters to generate M groups of fault warning information. The warning types of each group of fault warning information are the same. Determine the power operation and maintenance priority based on the fault coefficient. Homologous faults use the same warning method, such as icon flashing at different frequencies, etc., which is convenient for identification and can directly perform fault determination. Based on the M groups of fault warning information, perform a display warning of power faults on the terminal interface.

[0041] In the embodiment of the present application, there is a scanning image acquisition module for performing infrared scanning monitoring on the power equipment area to determine an infrared scanning image; an infrared image acquisition module for performing filtering, noise reduction, and grayscale processing on the infrared scanning image based on a power data interface to determine a power infrared image; a fault analysis module for performing local image positioning determination with a corner window determined based on pixel structure tensors, and through transfer learning, supervising and training a power fault diagnosis module; a power fault diagnosis module for transmitting the power infrared image back to the power fault diagnosis module, calculating the structure tensor pixel by pixel, positioning the corner window image, and performing fault characteristic determination to determine a fault diagnosis list; a display warning module for traversing the fault diagnosis list, marking faults in the power infrared image, and performing a display warning of power faults on the terminal interface. This solves the technical problem in the prior art that the power fault detection system cannot identify and prioritize faults in real time and accurately, resulting in untimely maintenance responses and increased system operation risks. It achieves the technical effect of being able to detect and classify power faults in real time and accurately, and through fault characteristic clustering and priority ranking, realizing an efficient maintenance response, and improving the operation reliability and safety of the power system.

[0042] The above specific embodiments do not constitute a limitation on the protection scope of the present application. Those skilled in the art should understand that various modifications, combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principle of the present application shall be included within the protection scope of the present application. In some cases, the actions or steps recited in the present application can be executed in a different order from that in the embodiments and still achieve the desired results. Additionally, the processes depicted in the drawings do not necessarily require the specific order or sequential order shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

Claims

1. The power fault diagnosis and early warning system based on infrared spectrum is characterized by: The system comprises: A scanning image acquisition module is used to perform infrared scanning monitoring on the power equipment area and determine the infrared scanning image; An infrared image acquisition module, used to perform filtering, noise reduction and grayscale processing on the infrared scanning image based on the power data interface to determine a power infrared image; The fault analysis module is used to determine the local image positioning based on the corner point window determined by the pixel structure tensor, and supervise the training of the power fault diagnosis module through transfer learning; The power fault diagnosis module is used to transmit the power infrared image back to the power fault diagnosis module, perform structural tensor calculation pixel by pixel, locate the corner point window image and perform fault feature determination, and determine the fault diagnosis list; The display warning module is used to traverse the fault diagnosis list, mark the fault in the power infrared image, and display the power fault warning on the terminal interface.

2. The power fault diagnosis and early warning system based on infrared spectrum according to claim 1, characterized in that: The fault analysis module is also used for: For the power equipment area, determining the critical temperature value under standard power operation and generating an infrared reference spectrum; Establishing a mapping between the infrared reference spectrum and the power equipment area, and determining the power fault characteristics of the mapped power structure based on the mapping relationship; Based on the infrared reference spectrum and the power fault characteristics, sample drive training is performed to generate the power fault diagnosis module.

3. The power fault diagnosis and early warning system based on infrared spectrum as claimed in claim 2 is characterized in that: The fault analysis module is also used for: Determine the diagnostic model by conducting supervised training based on samples of power fault records; The functional relationship between grayscale value and temperature value is introduced, the large diagnosis model is migrated and called, a corner point reconstruction layer is added and supervised training is performed until convergence, and the power fault diagnosis module is determined.

4. The power fault diagnosis and early warning system based on infrared spectrum according to claim 1, characterized in that: The power fault diagnosis module is also used for: The power infrared image is transmitted to the power fault diagnosis module, and a corner point window determination is performed based on a pixel structure tensor to reconstruct and determine the corner point window image; Identify the corner point window image, and based on the functional relationship between the gray value and the temperature value, perform temperature value limit crossing judgment based on the infrared reference spectrum to locate the abnormal temperature area; For the abnormal temperature area, a power fault decision is made based on the mapping relationship, and the fault diagnosis list is integrated and output.

5. The power fault diagnosis and early warning system based on infrared spectrum as claimed in claim 4, characterized in that: The power fault diagnosis module is also used for: Traversing the power infrared image, identifying a first pixel point, wherein the first pixel point is any pixel point in the power infrared image; For the first pixel point, combine neighboring pixel points to identify and determine a first pixel value matrix; A corner point window is determined for the first pixel value matrix. If it is a corner point window, the first pixel value matrix is ​​retained; if it is not a corner point window, the first pixel value matrix is ​​screened out.

6. The power fault diagnosis and early warning system based on infrared spectrum according to claim 5, characterized in that: The power fault diagnosis module is also used for: Determine a preset gradient value, wherein the preset gradient value is a gradient threshold between pixels; Traversing the first pixel value matrix, calculating neighborhood pixel gradient features as the structure tensor of the first pixel point, wherein the structure tensor includes 4 items for quantifying the gradient change features of the pixel point, and the neighborhood pixel gradient features include a gradient coefficient and a gradient direction; Based on the preset gradient value, determining the structure tensor to determine the corner window; If at least one item of the structure tensor is greater than the preset gradient value, the first pixel value matrix is ​​used as the corner point window.

7. The power fault diagnosis and early warning system based on infrared spectrum according to claim 6, characterized in that: The power fault diagnosis module is also used for: For the electric power infrared image, corner point determination is performed pixel by pixel to determine N corner point windows; Based on the relative distribution of the image, the N corner point windows are discretely spliced ​​to determine the corner point window image.

8. The infrared spectrum-based power fault diagnosis and early warning system according to claim 1, characterized in that: The power fault diagnosis module is also used for: Traversing the fault diagnosis list, and determining fault characteristics based on the fault level and fault diffusion characteristics, wherein the fault characteristics include a fault coefficient; Perform homologous clustering processing on the fault features to determine M fault clusters, wherein the faults in the clusters are homologous; Traversing the M fault clusters, generating M groups of fault warning information, wherein the warning type of each group of fault warning information is consistent, and determining the power operation and maintenance priority based on the fault coefficient; Based on the M groups of fault warning information, a display warning of power failure is performed on the terminal interface.

Images