Arrester intelligent diagnosis system and method based on infrared spectrum

Abstract

The invention provides an intelligent lightning arrester diagnosis system and method based on an infrared spectrum, and particularly relates to the technical field of electrical equipment live detection and diagnosis, and the system integrates data acquisition, analysis, storage and display functions. The system automatically collects basic ledger and infrared spectrogram data of the lightning arrester through the data acquisition unit, carries out deep analysis by using an infrared spectrogram analysis service module in the data analysis unit, accurately identifies faults and generates results. And the data storage unit adopts an atlas file storage FTP module, so that the analysis result is efficiently and safely stored, and tracing and management are facilitated. The application display unit visually displays key information of the lightning arrester equipment through a macroscopic display module, the key information comprises an early warning state, defect properties, commissioning time, manufacturers and the like, comprehensive and instant equipment state overview is provided for operation and maintenance personnel, the problems of low accuracy and low efficiency of defect diagnosis of the lightning arrester equipment are solved, and the fault diagnosis efficiency of the lightning arrester equipment is improved. Therefore, the accuracy and efficiency of lightning arrester equipment defect diagnosis are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of live detection and diagnosis of electrical equipment, and in particular to an intelligent diagnosis system and method for lightning arresters based on infrared spectra. Background Art

[0002] During the operation of the power system, the online monitoring data of the arrester is affected by the environment and electromagnetic interference. In addition, since the abnormal heating of the arrester belongs to voltage-type heating, it is not easy to detect, resulting in low detection efficiency and a large workload in the subsequent spectrum processing. This paper studies the integration, storage, retrieval and spectrum processing framework of the arrester infrared spectrum, and realizes the key technologies of automatic recognition of the arrester infrared spectrum and automatic fault diagnosis. It is also particularly important to study the real-time monitoring of the arrester valve status based on the discharge recording monitoring of the arrester and the zero-value and low-value discharge characteristics of the valve.

[0003] There are many types of lightning arresters at present. The distribution of lightning arresters in power systems is too wide. The performance of different types of lightning arresters varies. Under the influence of lightning, surge, dirt, corrosion and other factors, lightning arresters are easily damaged. Their safe and stable operation directly affects the safety of the power grid. Therefore, it is necessary to monitor the status of lightning arresters. The main problems are as follows: 1. The accuracy of infrared diagnosis is insufficient. Manual observation of infrared spectra is prone to misjudgment, misjudgment, and missed judgment, and it is time-consuming and difficult. In particular, voltage-induced thermal defects have the characteristics of small relative temperature difference, strong concealment, few typical cases, and great harm, which require precise infrared detection and analysis; 2. The value of data has not been fully activated. Infrared spectrum data are scattered and stored randomly. The scattered spectrum resources have not been centralized and shared, which is not conducive to the formation of knowledge precipitation and typical libraries, resulting in insufficient depth and breadth of application of image features, similar comparisons, and archival analysis and judgment methods; 3. There is a lack of standardized infrared diagnosis system. Both manual and airborne equipment infrared inspections lack systematic and standardized constraints. The quality and efficiency of manual inspections are highly dependent on the attitude and experience of operation and maintenance personnel. The effectiveness of infrared spectra has not been fully utilized and cannot support intelligent diagnosis and analysis applications.

[0004] In summary, how to solve the problem of low accuracy and efficiency in defect diagnosis of lightning arrester equipment is a difficult problem that needs to be solved urgently in this technical field. Summary of the invention

[0005] The main purpose of the present invention is to provide an intelligent diagnosis system and method for lightning arresters based on infrared spectra to solve the problems of low accuracy and efficiency in defect diagnosis of lightning arrester equipment, thereby improving the accuracy and efficiency of equipment defect diagnosis, reducing the difficulty and analysis intensity of infrared photography, and building a unified system of infrared analysis and diagnosis, effectively reducing the burden on front-line staff, improving work efficiency, and ensuring the quality of equipment operation and inspection.

[0006] In order to achieve the above-mentioned objectives, the present invention provides an intelligent diagnosis system and method for lightning arresters based on infrared spectra, thereby improving the accuracy and efficiency of defect diagnosis of lightning arrester equipment, reducing the difficulty and analysis intensity of infrared photography, and building a unified system of infrared analysis and diagnosis, thereby effectively reducing the burden on front-line staff, improving work efficiency, and ensuring the quality of equipment operation and inspection.

[0007] In a first aspect, the present invention provides an intelligent diagnosis system for lightning arresters based on infrared spectrum, the system comprising:

[0008] A data acquisition unit, the data acquisition unit is used to collect basic ledger data of lightning arresters and infrared spectrum data of lightning arresters;

[0009] A data analysis unit, the data analysis unit comprising an infrared spectrum analysis service module, the infrared spectrum analysis service module being used to analyze the basic ledger data of the lightning arrester and the infrared spectrum data of the lightning arrester and obtain a fault analysis result;

[0010] A data storage unit, the data storage unit includes a graph file storage FTP module, the graph file storage FTP module is used to store the fault analysis result in the form of a graph file;

[0011] An application display unit, the display unit includes a macro display module, the macro display module is used to display the fault analysis result, the fault analysis result is used to indicate the warning status information, defect nature information, commissioning time information and manufacturer information;

[0012] Wherein, the data acquisition unit, the data analysis unit, the data storage unit and the application display unit are connected via a service bus.

[0013] Optionally, the data acquisition unit includes:

[0014] A basic ledger data collection module, which is used to collect basic ledger data of lightning arresters;

[0015] The lightning arrester infrared spectrum data acquisition module is used to collect the lightning arrester infrared spectrum data.

[0016] Optionally, the infrared spectrum analysis service module includes:

[0017] An analysis module, the analysis module is used to analyze the infrared spectrum data of the lightning arrester to obtain a temperature matrix of the lightning arrester;

[0018] A filtering module, the filtering module is used to filter out the temperature data of the low temperature area in the temperature matrix to obtain the temperature data after filtering out the low temperature area;

[0019] A LUT mapping processing module, wherein the LUT mapping processing module processes the temperature data after filtering out the low temperature area through LUT mapping to obtain a grayscale image;

[0020] An identification module is used to identify the arrester and the arrester sub-components in the grayscale image through analysis by a recognition algorithm to obtain temperature information of the arrester sub-components.

[0021] Optionally, the infrared spectrum analysis service module further includes:

[0022] A denoising module, wherein the denoising module removes noise from the temperature information of the arrester subcomponent by a denoising algorithm to obtain a denoised subcomponent temperature information;

[0023] The analysis acquisition module is used to obtain the arrester thermal fault specification and obtain the fault analysis result by analyzing the casing regional temperature or analyzing the casing central axis temperature curve according to the arrester thermal fault specification and the de-noised sub-component temperature information.

[0024] Optionally, the data analysis unit further includes:

[0025] A data management service module, the data management service module is used to manage the basic ledger data of the lightning arrester and the fault analysis results;

[0026] A macro display service module, wherein the macro display service module is used to display the basic ledger data of the lightning arrester and the fault analysis results.

[0027] Optionally, the data storage unit further includes:

[0028] The structured data storage MySql module is used to store the collected arrester basic ledger data and the fault analysis results in a structured form.

[0029] Optionally, the application display unit further includes:

[0030] Arrester information management module, the arrester information management module is used to manage the arrester basic ledger data and the arrester infrared spectrum data;

[0031] The infrared spectrum analysis management module is used to analyze the infrared spectrum data of the arrester and confirm, display and export the analysis results.

[0032] In a second aspect, the present invention provides an intelligent diagnosis method for a lightning arrester based on infrared spectrum, which is applied to the intelligent diagnosis system for a lightning arrester based on infrared spectrum described in the first aspect, comprising:

[0033] Collect basic data of lightning arresters and infrared spectrum data of lightning arresters;

[0034] Analyze the basic ledger data of the lightning arrester and the infrared spectrum data of the lightning arrester and obtain the fault analysis result;

[0035] Storing the fault analysis results in the form of a graph file;

[0036] Display fault analysis results, which are used to indicate warning status information, defect nature information, commissioning time information and manufacturer information.

[0037] Optionally, analyzing the basic ledger data of the lightning arrester and the infrared spectrum data of the lightning arrester and obtaining a fault analysis result includes:

[0038] Analyze infrared spectrum data of the lightning arrester to obtain a temperature matrix of the lightning arrester;

[0039] Filtering out the temperature data of the low temperature region in the temperature matrix to obtain the temperature data after filtering out the low temperature region;

[0040] Processing the temperature data after filtering out the low temperature area through LUT mapping to obtain a grayscale image;

[0041] According to the grayscale image, the arrester and the arrester sub-components are identified by a recognition algorithm analysis to obtain temperature information of the arrester sub-components, wherein the arrester sub-components are used to indicate the arrester joint, the switch contact, the insulator, the bushing and the threaded area;

[0042] Remove the noise in the temperature information of the arrester subcomponent by a denoising algorithm to obtain the denoised subcomponent temperature information;

[0043] Obtain the arrester thermal fault specification, and obtain the fault analysis result by analyzing the casing regional temperature or analyzing the casing centerline temperature curve according to the arrester thermal fault specification, the arrester basic ledger data and the de-noised sub-component temperature information.

[0044] Optionally, the identifying the arrester and the arrester sub-components by analyzing the grayscale image through a recognition algorithm to obtain temperature information of the arrester sub-components includes:

[0045] Processing the grayscale image through a YOLACT++ model to identify the lightning arrester;

[0046] Processing the arrester through a FastInst model to identify subcomponents of the arrester;

[0047] The temperature spectrum of the arrester sub-component is analyzed to obtain temperature information of the arrester sub-component.

[0048] The intelligent diagnosis system and method of lightning arrester based on infrared spectrum provided by the present application is intended to solve the problem of low accuracy and efficiency of defect diagnosis of lightning arrester equipment. The system integrates four core units of data acquisition, intelligent analysis, data storage and intuitive display, and works efficiently and collaboratively through the service bus. The data acquisition unit comprehensively collects the basic ledger and infrared spectrum data of the lightning arrester, providing a solid foundation for subsequent analysis. The data analysis unit uses the infrared spectrum analysis service module, combined with the basic ledger, to accurately diagnose faults and generate results. The data storage unit ensures the safe storage and convenient management of the analysis results through the FTP module and the structured database MySql. The application display unit takes the macro display module as the core, intuitively presents the key information and early warning status of the lightning arrester equipment, and facilitates the operation and maintenance personnel to quickly grasp the equipment status. In addition, the present invention also provides an intelligent diagnosis method for lightning arrester based on infrared spectrum, which effectively filters out interference, extracts key temperature information, and accurately analyzes and locates faults according to the thermal fault specification of the lightning arrester through the fine processing of the infrared spectrum and the application of intelligent recognition algorithm. This method not only reduces the difficulty of infrared photography and analysis, but also builds a unified infrared analysis and diagnosis system, which greatly reduces the burden on front-line staff and improves work efficiency and equipment operation and maintenance quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] The drawings constituting a part of the present application are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0050] Figure 1 A schematic diagram of an intelligent diagnosis system for lightning arresters based on infrared spectra provided in this application;

[0051] Figure 2 A schematic flow chart of the intelligent diagnosis method for lightning arresters based on infrared spectra provided in this application.

[0052] The above drawings have shown clear embodiments of the present application, which will be described in more detail later. These drawings and text descriptions are not intended to limit the scope of the present application in any way, but to illustrate the concept of the present application to those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION

[0053] In order to make the purpose, technical solutions and advantages of this application clearer, the technical solutions in this application will be clearly and completely described below in conjunction with the drawings in this application. Obviously, the described embodiments are part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0054] The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of the present invention and the above drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the numbers used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in sequences other than those illustrated or described herein.

[0055] In the present invention, words such as "exemplary" or "for example" are used to indicate examples, illustrations or descriptions. Any embodiment or design described as "exemplary" or "for example" in this application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "exemplary" or "for example" is intended to present related concepts in a specific way.

[0056] During the operation of the power system, the online monitoring data of the arrester is affected by the environment and electromagnetic interference. In addition, since the abnormal heating of the arrester belongs to voltage-type heating, it is not easy to detect, resulting in low detection efficiency and a large workload in the subsequent spectrum processing. This paper studies the integration, storage, retrieval and spectrum processing framework of the arrester infrared spectrum, and realizes the key technologies of automatic recognition of the arrester infrared spectrum and automatic fault diagnosis. It is also particularly important to study the real-time monitoring of the arrester valve status based on the discharge recording monitoring of the arrester and the zero-value and low-value discharge characteristics of the valve.

[0057] There are many types of lightning arresters at present. The distribution of lightning arresters in power systems is too wide. The performance of different types of lightning arresters varies. Under the influence of lightning, surge, dirt, corrosion and other factors, lightning arresters are easily damaged. Their safe and stable operation directly affects the safety of the power grid. Therefore, it is necessary to monitor the status of lightning arresters. The main problems are as follows: 1. The accuracy of infrared diagnosis is insufficient. Manual observation of infrared spectra is prone to misjudgment, misjudgment, and missed judgment, and it is time-consuming and difficult. In particular, voltage-induced thermal defects have the characteristics of small relative temperature difference, strong concealment, few typical cases, and great harm, which require precise infrared detection and analysis; 2. The value of data has not been fully activated. Infrared spectrum data are scattered and stored randomly. The scattered spectrum resources have not been centralized and shared, which is not conducive to the formation of knowledge precipitation and typical libraries, resulting in insufficient depth and breadth of application of image features, similar comparisons, and archival analysis and judgment methods; 3. There is a lack of standardized infrared diagnosis system. Both manual and airborne equipment infrared inspections lack systematic and standardized constraints. The quality and efficiency of manual inspections are highly dependent on the attitude and experience of operation and maintenance personnel. The effectiveness of infrared spectra has not been fully utilized and cannot support intelligent diagnosis and analysis applications.

[0058] In summary, how to solve the problem of low accuracy and efficiency in defect diagnosis of lightning arrester equipment is a difficult problem that needs to be solved urgently in this technical field.

[0059] The application provides an intelligent diagnosis system and method for lightning arresters based on infrared spectra. The system automatically collects basic records and infrared spectra data of lightning arresters, uses advanced algorithms to conduct in-depth analysis and quickly locate faults, and realizes efficient storage and intuitive display of data. The service bus connects each unit to ensure seamless connection and efficient operation of the data processing process, aiming to improve the accuracy and efficiency of equipment defect diagnosis, reduce the manual burden, and ensure the reliable operation and efficient operation and maintenance of power equipment.

[0060] The technical solution of the present application and how the technical solution of the present application solves the above-mentioned technical problems are described in detail below with specific embodiments. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of the present application will be described below in conjunction with the accompanying drawings.

[0061] Figure 1 The schematic diagram of the intelligent diagnosis system for lightning arresters based on infrared spectrum provided in this application is as follows: Figure 1 As shown, the intelligent diagnosis system for lightning arresters based on infrared spectrum provided in this embodiment includes:

[0062] A data acquisition unit, the data acquisition unit is used to collect basic ledger data of lightning arresters and infrared spectrum data of lightning arresters;

[0063] A data analysis unit, the data analysis unit comprising an infrared spectrum analysis service module, the infrared spectrum analysis service module being used to analyze the basic ledger data of the lightning arrester and the infrared spectrum data of the lightning arrester and obtain a fault analysis result;

[0064] A data storage unit, the data storage unit includes a graph file storage FTP module, the graph file storage FTP module is used to store the fault analysis result in the form of a graph file;

[0065] An application display unit, the display unit includes a macro display module, the macro display module is used to display the fault analysis result, the fault analysis result is used to indicate the warning status information, defect nature information, commissioning time information and manufacturer information;

[0066] Wherein, the data acquisition unit, the data analysis unit, the data storage unit and the application display unit are connected via a service bus.

[0067] Furthermore, the data acquisition unit comprises:

[0068] A basic ledger data collection module, which is used to collect basic ledger data of lightning arresters;

[0069] The lightning arrester infrared spectrum data acquisition module is used to collect the lightning arrester infrared spectrum data.

[0070] Furthermore, the infrared spectrum analysis service module includes:

[0071] An analysis module, the analysis module is used to analyze the infrared spectrum data of the lightning arrester to obtain a temperature matrix of the lightning arrester;

[0072] A filtering module, the filtering module is used to filter out the temperature data of the low temperature area in the temperature matrix to obtain the temperature data after filtering out the low temperature area;

[0073] A LUT mapping processing module, wherein the LUT mapping processing module processes the temperature data after filtering out the low temperature area through LUT mapping to obtain a grayscale image;

[0074] An identification module is used to identify the arrester and the arrester sub-components in the grayscale image through analysis by a recognition algorithm to obtain temperature information of the arrester sub-components.

[0075] Furthermore, the infrared spectrum analysis service module also includes:

[0076] A denoising module, wherein the denoising module removes noise from the temperature information of the arrester subcomponent by a denoising algorithm to obtain a denoised subcomponent temperature information;

[0077] The analysis acquisition module is used to obtain the arrester thermal fault specification and obtain the fault analysis result by analyzing the casing regional temperature or analyzing the casing central axis temperature curve according to the arrester thermal fault specification and the de-noised sub-component temperature information.

[0078] Furthermore, the data analysis unit also includes:

[0079] A data management service module, the data management service module is used to manage the basic ledger data of the lightning arrester and the fault analysis results;

[0080] A macro display service module, wherein the macro display service module is used to display the basic ledger data of the lightning arrester and the fault analysis results.

[0081] Optionally, the data storage unit further includes:

[0082] The structured data storage MySql module is used to store the collected basic ledger data of the lightning arrester and the fault analysis results in a structured form, so as to associate the collected basic ledger data of the lightning arrester with the fault analysis results for easy query.

[0083] Furthermore, the application display unit also includes:

[0084] Arrester information management module, the arrester information management module is used to manage the arrester basic ledger data and the arrester infrared spectrum data;

[0085] The infrared spectrum analysis management module is used to analyze the infrared spectrum data of the arrester and confirm, display and export the analysis results.

[0086] It is understandable that this system uses image processing, deep learning and other artificial intelligence technologies to realize intelligent recognition and defect diagnosis of infrared spectra of lightning arresters based on "DL / T664-Application Specification of Infrared Diagnosis of Live Equipment", reduce the misjudgment, wrong judgment and missed judgment of manual processing, realize batch, automatic, intelligent diagnosis, and automatically generate diagnostic reports, greatly improve the accuracy of infrared analysis and diagnosis, and save 60% of diagnosis time. Through the archiving, processing, analysis, and tracking of infrared spectra, the scattered spectra resources are centralized and shared; the use of artificial intelligence technology improves the application depth and breadth of image feature method, similar comparison method, and archive analysis and judgment method, which helps to improve fault diagnosis, especially the accuracy of voltage-induced heating defects. Realize the standardized storage and management of infrared spectra data of lightning arresters, intelligent analysis and diagnosis, and comparison of analysis results, build a standard system that adapts to the professional and technical characteristics of infrared detection of lightning arresters, promote the level of infrared diagnosis and analysis, and deepen the application effect, thereby improving the standardization and leanness of infrared detection of lightning arresters and reducing the occurrence of lightning arrester failures.

[0087] The intelligent diagnosis of lightning arresters based on infrared spectrum provided in this embodiment has three general principles:

[0088] Principle 1 Application Positioning: This system realizes the standardized storage and management, automatic identification, and diagnosis report of lightning arrester infrared spectra, providing good lightning arrester infrared spectrum intelligent diagnosis services for various detection, testing, and inspection work, reducing safety hazards and ensuring the safe and stable operation of the power system.

[0089] Principle 2 Application Deployment: The infrared spectrum intelligent diagnosis system for lightning arresters is independently deployed in the Inner Mongolia Ultra-High Voltage Power Supply Company to provide service support for infrared spectrum analysis of lightning arresters.

[0090] Principle 3 Functional Division:

[0091] Lightning arrester infrared spectrum management

[0092] ① Arrester basic ledger management; ② Arrester infrared spectrum management; ③ Arrester analysis data management; ④ Macro statistical analysis

[0093] Infrared spectrum analysis and feature recognition

[0094] Automatic analysis of infrared spectra is achieved through noise reduction processing of infrared spectra, and manual marking and analysis are supported. Accurate recognition of thermal field characteristics of detection units in infrared spectra is achieved based on deep learning algorithms.

[0095] Infrared spectrum intelligent diagnosis service: perform intelligent diagnosis on infrared spectrum and provide diagnosis report (defect type, diagnosis basis, defect phenomenon, disposal suggestions).

[0096] Figure 2 The flowchart of the intelligent diagnosis method of lightning arrester based on infrared spectrum provided by the present application is described in detail, such as Figure 2 As shown, the intelligent diagnosis method of lightning arrester based on infrared spectrum provided in this embodiment includes:

[0097] S201: Collect basic inventory data of lightning arresters and infrared spectrum data of lightning arresters.

[0098] The specific implementation of collecting basic ledger data of lightning arresters and infrared spectrum data of lightning arresters includes the following steps:

[0099] Step 1: Preparation

[0100] Determine the collection scope and objectives: First, according to the power system management requirements and the distribution of lightning arresters, clarify the list of target lightning arresters for this data collection, including their location, model, commissioning time and other basic information.

[0101] Configure acquisition equipment: Select high-precision and high-stability data acquisition terminals, including but not limited to infrared thermal imagers, RFID readers, data loggers, etc., and ensure that all equipment has been calibrated to the standard state. For infrared thermal imagers, the appropriate temperature measurement range, image resolution and infrared correction parameters must be set in advance.

[0102] Establish a communication network: Build a stable and reliable communication network to ensure that the acquisition equipment can transmit data to the data center in real time or on a scheduled basis. Wireless communication (such as Wi-Fi, 4G / 5G) or wired communication (such as optical fiber) can be used, and can be flexibly selected according to the on-site environment.

[0103] Step 2: Collect basic data of lightning arresters

[0104] Collection method 1: RFID tag reading: Install a unique RFID tag for each arrester, and store the basic information of the arrester (such as equipment number, model, manufacturer, commissioning date, etc.) in the tag. Use the RFID reader to scan the tags on the arrester one by one, automatically read and record this information to the data collection terminal.

[0105] Manual review and supplement: For some information that cannot be directly obtained through RFID (such as historical maintenance records, special operating conditions, etc.), it is necessary to manually input it by on-site personnel or check paper files to ensure the integrity of the basic ledger data.

[0106] Collection method 2: Import the data into the system for management and maintenance according to the Excel template provided by the customer.

[0107] Step 3: Collect infrared spectrum data of lightning arrester

[0108] Environmental preparation: Before collecting infrared spectra, ensure that the ambient temperature is stable and there are no interference factors such as direct sunlight and strong wind to reduce the impact of the external environment on the infrared temperature measurement results.

[0109] Equipment deployment and debugging: Fix the infrared thermal imager at a suitable observation position, adjust the focal length, viewing angle and infrared parameters to ensure that the arrester as a whole and key parts (such as porcelain sleeves, connection parts, etc.) are clearly visible. Perform a pre-scan to check whether the image quality and temperature measurement accuracy meet the requirements.

[0110] Data acquisition: Start the data acquisition function of the infrared thermal imager, conduct a comprehensive scan of the target arrester, and generate an infrared spectrum. According to needs, different time intervals can be set for continuous monitoring to capture the dynamic process of temperature changes.

[0111] Data storage and transmission: The collected infrared spectrum data is saved to the data acquisition terminal and uploaded to the data center server in real time or at a fixed time through the pre-established communication network. At the same time, additional information such as acquisition time and environmental conditions is recorded for subsequent analysis.

[0112] It can be understood that through the above steps, the efficient and accurate collection of basic ledger data and infrared spectrum data of lightning arresters is achieved, laying a solid foundation for the subsequent data processing, fault diagnosis and performance evaluation of the lightning arrester status monitoring system. This method has the advantages of high degree of automation, comprehensive and accurate data, and strong real-time performance, which can significantly improve the intelligent level of power system operation and maintenance management.

[0113] S202: Analyze the basic inventory data of the lightning arrester and the infrared spectrum data of the lightning arrester to obtain a fault analysis result.

[0114] The analysis of the basic ledger data of the lightning arrester and the infrared spectrum data of the lightning arrester to obtain the fault analysis result includes the following steps:

[0115] Step 1: Analyze the infrared spectrum data of the lightning arrester to obtain the temperature matrix of the lightning arrester.

[0116] Specifically, the specific implementation of parsing the infrared spectrum data of the lightning arrester to obtain the temperature matrix of the lightning arrester includes the following steps:

[0117] Step 101: Data Preprocessing

[0118] First, the collected infrared spectrum data of the lightning arrester is preprocessed. This step mainly includes the following sub-steps:

[0119] Noise filtering: Use filtering algorithms (such as median filtering, Gaussian filtering, etc.) to remove random noise in the image and improve image quality.

[0120] Grayscale: If the acquired infrared spectrum is in color, it needs to be converted to a grayscale image for subsequent processing. Grayscale can be achieved by calculating the brightness value of each pixel.

[0121] Image enhancement: Enhance the contrast of the image through methods such as histogram equalization to make the temperature difference more obvious.

[0122] Step 102: Image Segmentation

[0123] The preprocessed image may contain multiple parts of the arrester and its surroundings. In order to obtain the temperature information of the arrester alone, image segmentation is required to separate the arrester from the background. Image segmentation can be performed using a threshold-based method, an edge detection-based method, or a region-based method.

[0124] Step 103: Temperature Matrix Construction

[0125] After completing the image segmentation, the next step is to construct the temperature matrix of the lightning arrester. The specific steps are as follows:

[0126] Pixel coordinate mapping: Establish the mapping relationship between the image pixel coordinates and the actual position coordinates of the arrester. This usually needs to be calculated based on the actual size of the arrester and the scale in the image.

[0127] Temperature value extraction: traverse the segmented arrester image and read the gray value of each pixel (or the temperature value after some conversion) as the element of the corresponding position in the temperature matrix.

[0128] Matrix construction: According to the shape and size of the arrester, a two-dimensional array (i.e., temperature matrix) is constructed to store these temperature values. The rows and columns of the matrix correspond to the pixels of the arrester in the horizontal and vertical directions in the image, respectively.

[0129] Step 104: Matrix verification and adjustment

[0130] After the temperature matrix is ​​constructed, it needs to be verified and adjusted to ensure its accuracy and rationality. Verification includes but is not limited to:

[0131] Temperature range check: Verify whether the temperature values ​​in the matrix are within a reasonable range, such as whether they exceed the temperature measurement range of the infrared thermal imaging device.

[0132] Continuity check: Check whether the temperature values ​​of adjacent elements in the matrix change continuously to avoid sudden temperature changes caused by image segmentation errors or noise interference.

[0133] Outlier processing: For abnormal temperature values ​​in the matrix (such as extremely high or extremely low temperature points), they need to be processed according to the specific situation, such as replacing them with the average value of neighboring points.

[0134] It can be understood that step 1 can be used to analyze the infrared spectrum data of the arrester and construct a temperature matrix reflecting the temperature distribution of the arrester. This matrix provides an important data basis for subsequent fault analysis and helps to improve the accuracy and efficiency of diagnosis.

[0135] Step 2: Filter out the temperature data of the low temperature area in the temperature matrix to obtain the temperature data after filtering out the low temperature area.

[0136] Furthermore, the temperature data of the low temperature area may be filtered out by a histogram bimodal method to obtain the temperature data after filtering out the low temperature area.

[0137] The specific implementation process of filtering out the temperature data of the low temperature area by the histogram bimodal method to obtain the temperature data after filtering out the low temperature area includes the following steps:

[0138] Step 201: Set low temperature threshold

[0139] First, according to the normal operating temperature range of the arrester and the ambient temperature, a reasonable low temperature threshold is set. This threshold should be lower than the lowest temperature of the arrester during normal operation to ensure that the low temperature area not related to the fault can be effectively filtered out.

[0140] Step 202: Traverse the temperature matrix

[0141] Traverse each element in the temperature matrix (i.e., the temperature value of each pixel) and compare it with the set low temperature threshold.

[0142] Step 203: Filter out low temperature data

[0143] For elements whose temperature is lower than the threshold, they are marked as invalid or their values ​​are set to specific values ​​(such as 0 or NaN, indicating non-number) so that these low temperature data are ignored in subsequent processing. At the same time, the coordinate information of these low temperature data can be recorded or retained for subsequent analysis needs.

[0144] Histogram bimodal filtering:

[0145] Step 204: Constructing a temperature histogram

[0146] The temperature matrix after filtering out the low temperature area is statistically analyzed to construct a temperature histogram. The horizontal axis of the histogram represents the temperature value, and the vertical axis represents the number of pixels corresponding to the temperature value.

[0147] Step 205: Identify double peaks

[0148] Analyze the histogram to see if there is a clear bimodal structure. A bimodal structure usually indicates that there are two main distribution areas in the temperature data, one corresponding to normal or lower temperature areas and the other corresponding to possible high temperature anomalies.

[0149] Step 206: Determine segmentation threshold

[0150] If there are two peaks in the histogram, a segmentation threshold is determined by calculating the lowest point between the two peaks or using other suitable methods. This threshold is used to divide the temperature data into a normal area and a potential fault area.

[0151] Step 207: Filtering Temperature Data

[0152] According to the determined segmentation threshold, the temperature matrix is ​​traversed again, and the temperature data above the threshold is regarded as the data of the potential fault area and retained or further analyzed; the temperature data below or equal to the threshold is regarded as the data of the normal area and can be selectively ignored or processed in other ways.

[0153] It can be understood that the low temperature area in the temperature matrix is ​​successfully filtered out through step 2, and the temperature data is further filtered through the histogram bimodal method to highlight the possible high temperature abnormal area on the arrester. This method improves the accuracy and efficiency of fault analysis and provides strong support for subsequent fault diagnosis and early warning.

[0154] Step 3: Processing the temperature data after filtering out the low temperature area through LUT mapping to obtain a grayscale image;

[0155] The specific implementation of processing the temperature data after filtering out the low temperature area by LUT mapping to obtain a grayscale image includes the following steps:

[0156] Step 301: Prepare temperature data

[0157] Make sure that the low temperature area has been filtered out according to the previous steps, and get the processed temperature matrix. Each element in the matrix represents the temperature value at the corresponding position.

[0158] Step 302: Define grayscale range

[0159] Grayscale images are usually represented by gray levels from 0 (black) to 255 (white). Depending on actual needs, this range can be adjusted to better reflect temperature differences. For example, 0-255 is selected to represent the range from the lowest valid temperature to the highest temperature.

[0160] Step 303: Create a Lookup Table (LUT)

[0161] Create a lookup table based on the range of temperature data and the target grayscale range. Each entry in the lookup table corresponds to a temperature value and is mapped to a grayscale level. The mapping can be linear or nonlinear, depending on how to better show the temperature difference.

[0162] Linear Mapping: Simply linearly map the temperature range to the grayscale range. For example, if the lowest temperature is T min , the maximum temperature is T max , then the gray level G corresponding to temperature T can be calculated by the following formula:

[0163]

[0164] Wherein, if the gray level G is not an integer, it is rounded down.

[0165] Non-linear mapping: For some application scenarios, it may be necessary to distinguish high-temperature areas more carefully, while the low-temperature areas should be relatively rough. In this case, non-linear mapping methods such as logarithmic mapping or segmented mapping can be used.

[0166] The following is the LUT mapping process:

[0167] Step 304: Apply the lookup table

[0168] Traverse each element in the temperature matrix and use the lookup table to map its temperature value to the corresponding gray level. For each temperature value T in the matrix, there is a corresponding gray level G in the lookup table.

[0169] Step 305: Constructing a grayscale image

[0170] The mapped grayscale data is constructed into a grayscale image according to the dimensions of the original temperature matrix. The grayscale value of each pixel in the grayscale image is the result of mapping the temperature value at the corresponding position.

[0171] 4. Result Verification

[0172] Step 306: Verify the grayscale image

[0173] Check the generated grayscale image to ensure that the temperature distribution and abnormal areas are correctly reflected. This can be verified by comparing the original temperature data with the grayscale image, or by analyzing the histogram of the grayscale image using image processing software.

[0174] It can be understood that step 3 successfully converts the temperature data after filtering out the low temperature area into a grayscale image by using the lookup table mapping method. This method not only improves the intuitiveness of temperature data visualization, but also facilitates subsequent temperature distribution analysis and anomaly detection. By selecting a suitable mapping method, the display effect of the grayscale image can be further optimized to meet different application requirements.

[0175] Step 4: Based on the grayscale image, identify the arrester and the arrester sub-components through recognition algorithm analysis to obtain temperature information of the arrester sub-components.

[0176] Among them, the lightning arrester subcomponents are used to indicate the lightning arrester joints, knife contacts, insulators, bushings and threaded areas. In the power system, temperature monitoring of lightning arresters and their subcomponents (such as joints, knife contacts, insulators, bushings and threaded areas) is crucial to prevent equipment failures. This embodiment describes in detail how to use grayscale images, combined with advanced image recognition algorithms (YOLACT++ and FastInst), to achieve automatic identification of lightning arresters and their subcomponents, and further parse their temperature information.

[0177] Further, the step of identifying the arrester and the arrester sub-components by analyzing the grayscale image through a recognition algorithm to obtain temperature information of the arrester sub-components includes:

[0178] Processing the grayscale image through a YOLACT++ model to identify the lightning arrester;

[0179] Processing the arrester through a FastInst model to identify subcomponents of the arrester;

[0180] The temperature spectrum of the arrester sub-component is analyzed to obtain temperature information of the arrester sub-component.

[0181] The specific implementation includes the following steps:

[0182] The first is the grayscale image preprocessing process:

[0183] Step 401: Grayscale image acquisition

[0184] First, make sure you have obtained a grayscale image of the arrester and its surroundings as described in the previous steps. The image should clearly reflect the outlines of the components and the temperature differences.

[0185] Step 402: Image enhancement (optional)

[0186] In order to improve the recognition accuracy, the grayscale image can be preprocessed, such as contrast enhancement, noise suppression, etc., to improve the image quality.

[0187] The second is the arrester identification process:

[0188] Step 403: Apply the YOLACT++ model

[0189] Model loading: Load the pre-trained YOLACT++ model. YOLACT++ is an efficient instance segmentation model that can simultaneously predict the category and pixel-level mask of the target.

[0190] Image input: Input the preprocessed grayscale image into the YOLACT++ model.

[0191] Result output: The model outputs the category label, bounding box and corresponding pixel-level mask of the lightning arrester.

[0192] Then comes the arrester subcomponent identification process:

[0193] Step 404: FastInst model application

[0194] Model selection and loading: For the fine recognition of the arrester subcomponents, select the FastInst model. FastInst is a model designed for fast instance segmentation and is suitable for fine structure recognition in complex backgrounds.

[0195] Region cropping: According to the arrester mask output by the YOLACT++ model, the arrester area is cropped from the original grayscale image.

[0196] Subcomponent identification: The cropped arrester image is input into the FastInst model, which outputs the categories, bounding boxes, and masks of the arrester subcomponents (connectors, switch contacts, insulators, bushings, and threaded areas).

[0197] Then the temperature information is extracted:

[0198] Step 405: Analyze subcomponent temperature spectrum

[0199] Temperature mapping: The grayscale value in the subcomponent mask area is converted into a temperature value by utilizing the mapping relationship between the grayscale level in the grayscale image and the temperature (such as the linear or nonlinear mapping described above).

[0200] Temperature information integration: For each identified arrester subcomponent, calculate the average or maximum value (select according to actual needs) of the temperature of all pixels in its area as the temperature information of the subcomponent.

[0201] Result recording: The category, location (bounding box coordinates) and temperature information of each arrester subcomponent are recorded in the database for subsequent analysis and monitoring.

[0202] Finally, verification and optimization

[0203] Step 406: Result verification

[0204] Manual review: Randomly select some recognition results for manual review to verify the recognition accuracy and the accuracy of temperature information.

[0205] Performance evaluation: Based on the review results, the performance of YOLACT++ and FastInst models in specific application scenarios is evaluated, including recognition speed, accuracy, and robustness.

[0206] Step 407: Model optimization (optional)

[0207] According to the verification results, necessary adjustments and optimizations are made to the model, such as adjusting model parameters, increasing training data, etc., to improve recognition accuracy and generalization ability.

[0208] It can be understood that step 4 realizes the automatic recognition and temperature information extraction of the arrester and its subcomponents in the grayscale image by combining two advanced image recognition models, YOLACT++ and FastInst. This method not only improves the monitoring efficiency, but also ensures the accuracy and reliability of the data, providing strong support for the safe and stable operation of the power system.

[0209] Step 5: Remove the noise in the temperature information of the arrester subcomponent by a denoising algorithm to obtain the denoised subcomponent temperature information.

[0210] In the power system, the temperature information of the arrester sub-components obtained by image recognition technology often contains certain noise, which may come from environmental interference, sensor errors or imperfect algorithm processing during the image acquisition process. In order to obtain more accurate and reliable sub-component temperature information, this step will introduce in detail how to process the original temperature information through a denoising algorithm to remove the noise component.

[0211] In this step, the noise in the temperature information of the arrester sub-component is removed by a denoising algorithm to obtain the denoised sub-component temperature information, which specifically includes the following steps:

[0212] First choose the denoising algorithm:

[0213] In view of the characteristics of temperature information data, this embodiment selects a denoising algorithm suitable for one-dimensional data (such as temperature series) - a method of median filtering combined with wavelet transform. Median filtering can effectively remove random noise, while wavelet transform can more finely process noise components of different frequencies.

[0214] Then complete the denoising step:

[0215] Step 501: Data preparation

[0216] The temperature information of the arrester sub-components is extracted from the image recognition results in the above steps to form a one-dimensional temperature data sequence.

[0217] Conduct a preliminary check on the data to confirm its completeness and rationality.

[0218] Step 502: Median filtering

[0219] Choose an appropriate window size (an odd number like 3, 5, or 7) that will be used to calculate the median of the temperature values ​​around each temperature data point.

[0220] Traverse the entire temperature data sequence and apply the median filter algorithm to each data point, that is, replace the original data point with the median of the temperature value in the surrounding window.

[0221] This step aims to remove random noise and spike noise in the temperature data.

[0222] Step 503: Wavelet transform denoising

[0223] Perform wavelet transform on the temperature data sequence processed by median filtering, and select the appropriate mother wavelet and decomposition level.

[0224] In the wavelet transform domain, according to the different performances of noise and signal at different scales, appropriate thresholds are set to perform threshold processing on high-frequency coefficients to remove noise components.

[0225] The processed high-frequency coefficients and the unchanged low-frequency coefficients are subjected to inverse wavelet transform to obtain the denoised temperature data sequence.

[0226] This step can further remove low-frequency noise and periodic noise hidden in the data.

[0227] Step 504: Result verification and evaluation

[0228] Compare the temperature data series before and after denoising to see whether the noise is effectively suppressed.

[0229] The signal-to-noise ratio (SNR) and mean square error (MSE) of the denoised data are calculated to quantitatively evaluate the denoising effect.

[0230] If necessary, the denoising algorithm parameters can be adjusted according to the actual situation to achieve the best denoising effect.

[0231] Finally, the application of the denoised subcomponent temperature information is completed:

[0232] Using the denoised arrester subcomponent temperature information in subsequent analysis, monitoring and early warning systems can significantly improve the accuracy and reliability of the system. For example, a temperature threshold can be set based on the denoised temperature data, and an alarm will be automatically issued when the subcomponent temperature exceeds the threshold, reminding the operation and maintenance personnel to take timely measures to prevent equipment failure.

[0233] It can be understood that step 5 effectively removes the noise components in the temperature information of the arrester subcomponents by combining the two denoising algorithms of median filtering and wavelet transform, providing more accurate and reliable data support for subsequent power system monitoring and early warning. This denoising method is simple to operate and has significant effects, and is suitable for data processing needs in similar scenarios.

[0234] Step 6: Obtain the arrester thermal fault specification, and obtain the fault analysis result by analyzing the casing regional temperature or analyzing the casing centerline temperature curve according to the arrester thermal fault specification, the arrester basic ledger data and the denoised sub-component temperature information.

[0235] As we all know, in the power system, the arrester is an important protective device, and its operating status is directly related to the safety and stability of the power grid. Thermal fault is one of the common fault types of the arrester. By monitoring the temperature information of the arrester subcomponents, thermal faults can be discovered and diagnosed in time. This embodiment will describe in detail how to combine the arrester thermal fault specification with the denoised subcomponent temperature information, and obtain the analysis results of the arrester thermal fault through two methods: casing regional temperature analysis and casing central axis temperature curve analysis.

[0236] In this step, the fault analysis result is obtained by analyzing the temperature of the casing in different regions or analyzing the temperature curve of the central axis of the casing. The specific implementation includes the following processes:

[0237] Prerequisites:

[0238] The above steps complete the collection and denoising of the temperature information of the arrester sub-components, and obtain the denoised temperature data information.

[0239] Master the relevant specifications and standards for thermal failure of lightning arresters, including the temperature thresholds of different components, temperature change trends, etc.

[0240] Fault analysis steps:

[0241] Step 601: Casing temperature analysis by region

[0242] Area division: According to the structural design of the arrester, the bushing is divided into several temperature monitoring areas, such as top, middle, bottom, etc. Each area corresponds to a certain number of temperature measurement points.

[0243] Temperature data extraction: The temperature values ​​of the corresponding measuring points in each area are extracted from the temperature data sequence corresponding to the denoised basic inventory data of the lightning arrester (used to indicate the identification and attribute information of the lightning arrester) to form a temperature data set for each area.

[0244] Temperature comparison analysis: Compare and analyze the temperature data of each area with the temperature threshold in the arrester thermal fault specification. If the temperature of a certain area is continuously higher than the threshold, or the temperature rise trend is abnormal, it is preliminarily judged that there may be a thermal fault in the area.

[0245] Fault location: Combined with the physical structure and operating conditions of the arrester, further analyze the specific location of the abnormal temperature area and preliminarily determine the possible location of the thermal fault.

[0246] Step 602: Analysis of temperature curve of casing centerline

[0247] Temperature curve drawing: along the central axis of the casing, extract the temperature data of each measuring point and draw a temperature curve.

[0248] Trend analysis: Observe the changing trend of the temperature curve and analyze whether there are temperature mutation points, abnormal temperature gradients, etc. Temperature mutation points may indicate local overheating areas, while abnormal temperature gradients may reflect obstructed heat conduction or poor heat dissipation.

[0249] Specification comparison: Compare the characteristics of the temperature curve with the temperature change trend in the arrester thermal fault specification. If the temperature curve matches the failure mode in the specification, the existence of the thermal fault is further confirmed.

[0250] Comprehensive judgment: Combined with the results of the temperature analysis of the bushing in different areas, a comprehensive analysis is conducted on the temperature curve of the bushing centerline to form a complete diagnostic result of the arrester thermal fault.

[0251] Application of fault analysis results:

[0252] According to the fault analysis results, take appropriate maintenance or replacement measures to ensure the normal operation of the arrester. At the same time, the fault analysis results are recorded to provide a reference for subsequent fault prevention and maintenance.

[0253] It can be understood that this step realizes effective analysis and diagnosis of arrester thermal faults by combining the arrester thermal fault specification and the denoised sub-component temperature information through two methods: casing regional temperature analysis and casing central axis temperature curve analysis. This method has the characteristics of simple operation, accurate analysis and strong practicality, which is of great significance for improving the operating reliability and extending the service life of the arrester.

[0254] S203: storing the fault analysis result in the form of a graph file.

[0255] S204: Displaying the fault analysis result, wherein the fault analysis result is used to indicate warning status information, defect nature information, commissioning time information and manufacturer information.

[0256] The specific implementation can be achieved through the following steps:

[0257] Step 1: Generate fault analysis results

[0258] Data collection: First, collect the operating data, fault alarm information and historical maintenance records of the target equipment through sensor networks, monitoring systems and other channels.

[0259] Fault analysis: The specific implementation method of step S202 may be used to perform analysis and processing, identify the specific type, location and possible cause of the equipment fault, and generate detailed fault analysis results.

[0260] Step 2: Build the graph file

[0261] Information extraction: Extract key information from the fault analysis report, including but not limited to warning status information (such as emergency, warning, normal, etc.), defect nature information (such as mechanical failure, electrical failure, software failure, etc.), commissioning time information (the time point when the equipment is put into use) and manufacturer information.

[0262] Graph design: Design the graph structure based on the extracted information. The graph can be in the form of tree diagrams, flow charts, heat maps, etc., to intuitively display the association and hierarchical relationship between various information points. For example, with the equipment as the central node, sub-nodes such as warning status and defect nature radiate outward, and specific information is displayed in detail under each sub-node. For example, the commissioning time and manufacturer can be attached to the corresponding node in the form of labels or links.

[0263] Data mapping: Map the extracted information to the corresponding nodes of the graph to ensure the accuracy of the information. For complex data relationships, visual elements such as color coding and size changes can be used to distinguish and emphasize them.

[0264] File generation: Save the constructed map as a map file, supporting multiple formats (such as SVG, PNG, PDF, etc.) to facilitate viewing and sharing on different platforms and devices.

[0265] Step 3: Fault analysis results display

[0266] Interface design: Develop or integrate a user-friendly display interface for loading and displaying atlas files. The interface should provide interactive functions such as zooming, panning, and searching so that users can flexibly view atlas details.

[0267] Data visualization: Load the graph file on the display interface to visually display the fault analysis results in a graphical manner. Ensure that each node and connection line in the graph is clearly identifiable and the information displayed is accurate.

[0268] Interactive functions: Add interactive functions to the display interface, such as clicking a node to view detailed information, filtering specific types of faults, etc., to improve user experience and data query efficiency.

[0269] Of course, you can also Figure 1 The embodiment provides an infrared spectrum-based intelligent diagnosis system for lightning arresters to display fault analysis results.

[0270] It can be understood that steps S203 and S204 realize the graphical and structured display of fault information by constructing a graph file and displaying the fault analysis results. This method not only improves the intuitiveness and efficiency of troubleshooting, but also provides strong data support for subsequent equipment maintenance and performance evaluation. At the same time, the storability and shareability of the graph file also promotes the accumulation and inheritance of fault knowledge.

[0271] Those skilled in the art will appreciate that all or some of the steps in the methods disclosed above, and the functional modules / units in the systems and devices may be implemented as software, firmware, hardware, or a suitable combination thereof.

[0272] Those skilled in the art will readily appreciate other embodiments of the present application after considering the specification and practicing the invention disclosed herein. The present application is intended to cover any modification, use or adaptation of the present application, which follows the general principles of the present application and includes common knowledge or customary techniques in the art that are not disclosed in the present application. The specification and examples are intended to be exemplary only, and the true scope and spirit of the present application are indicated by the following claims.

[0273] It should be understood that the present application is not limited to the precise structures that have been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present application is limited only by the appended claims.

Claims

1. An intelligent diagnosis system for lightning arresters based on infrared spectrum, characterized in that: The system comprises: A data acquisition unit, the data acquisition unit is used to collect basic ledger data of lightning arresters and infrared spectrum data of lightning arresters; A data analysis unit, the data analysis unit comprising an infrared spectrum analysis service module, the infrared spectrum analysis service module being used to analyze the basic ledger data of the lightning arrester and the infrared spectrum data of the lightning arrester and obtain a fault analysis result; A data storage unit, the data storage unit includes a graph file storage FTP module, the graph file storage FTP module is used to store the fault analysis result in the form of a graph file; An application display unit, the application display unit includes a macro display module, the macro display module is used to display the fault analysis result, the fault analysis result is used to indicate the warning status information, defect nature information, commissioning time information and manufacturer information; Wherein, the data acquisition unit, the data analysis unit, the data storage unit and the application display unit are connected via a service bus.

2. The system according to claim 1, characterized in that The data acquisition unit comprises: A basic ledger data collection module, which is used to collect basic ledger data of lightning arresters; The lightning arrester infrared spectrum data acquisition module is used to collect the lightning arrester infrared spectrum data.

3. The system according to claim 1, characterized in that The infrared spectrum analysis service module includes: An analysis module, the analysis module is used to analyze the infrared spectrum data of the lightning arrester to obtain a temperature matrix of the lightning arrester; A filtering module, the filtering module is used to filter out the temperature data of the low temperature area in the temperature matrix to obtain the temperature data after filtering out the low temperature area; A LUT mapping processing module, wherein the LUT mapping processing module processes the temperature data after filtering out the low temperature area through LUT mapping to obtain a grayscale image; An identification module is used to identify the arrester and the arrester sub-components in the grayscale image through analysis by a recognition algorithm to obtain temperature information of the arrester sub-components.

4. The system according to claim 3, characterized in that The infrared spectrum analysis service module also includes: A denoising module, wherein the denoising module removes noise from the temperature information of the arrester subcomponent by a denoising algorithm to obtain a denoised subcomponent temperature information; The analysis acquisition module is used to obtain the arrester thermal fault specification and obtain the fault analysis result by analyzing the casing regional temperature or analyzing the casing central axis temperature curve according to the arrester thermal fault specification and the de-noised sub-component temperature information.

5. The system according to claim 1, characterized in that The data analysis unit also includes: A data management service module, the data management service module is used to manage the basic ledger data of the lightning arrester and the fault analysis results; A macro display service module, wherein the macro display service module is used to display the basic ledger data of the lightning arrester and the fault analysis results.

6. The system according to claim 1, characterized in that The data storage unit also includes: The structured data storage MySql module is used to store the collected arrester basic ledger data and the fault analysis results in a structured form.

7. The system according to claim 1, characterized in that The application display unit also includes: Arrester information management module, the arrester information management module is used to manage the arrester basic ledger data and the arrester infrared spectrum data; The infrared spectrum analysis management module is used to analyze the infrared spectrum data of the arrester and confirm, display and export the analysis results.

8. An intelligent diagnosis method for lightning arresters based on infrared spectrum, characterized in that: The method is applied to the intelligent diagnosis system for lightning arresters based on infrared spectrum according to any one of claims 1 to 7, and the diagnosis method comprises: Collect basic data of lightning arresters and infrared spectrum data of lightning arresters; Analyze the basic ledger data of the lightning arrester and the infrared spectrum data of the lightning arrester and obtain the fault analysis result; Storing the fault analysis results in the form of a graph file; Display fault analysis results, which are used to indicate warning status information, defect nature information, commissioning time information and manufacturer information.

9. The method according to claim 8, characterized in that The analyzing the basic ledger data of the lightning arrester and the infrared spectrum data of the lightning arrester and obtaining the fault analysis result includes: Analyze infrared spectrum data of the lightning arrester to obtain a temperature matrix of the lightning arrester; Filtering out the temperature data of the low temperature region in the temperature matrix to obtain the temperature data after filtering out the low temperature region; Processing the temperature data after filtering out the low temperature area through LUT mapping to obtain a grayscale image; According to the grayscale image, the arrester and the arrester sub-components are identified by a recognition algorithm analysis to obtain temperature information of the arrester sub-components, wherein the arrester sub-components are used to indicate the arrester joint, the switch contact, the insulator, the bushing and the threaded area; Remove the noise in the temperature information of the arrester subcomponent by a denoising algorithm to obtain the denoised subcomponent temperature information; Obtain the arrester thermal fault specification, and obtain the fault analysis result by analyzing the casing regional temperature or analyzing the casing centerline temperature curve according to the arrester thermal fault specification, the arrester basic ledger data and the de-noised sub-component temperature information.

10. The method according to claim 9, characterized in that The method of identifying the arrester and the arrester sub-components by analyzing the grayscale image through a recognition algorithm to obtain temperature information of the arrester sub-components includes: Processing the grayscale image through a YOLACT++ model to identify the lightning arrester; Processing the arrester through a FastInst model to identify subcomponents of the arrester; The temperature spectrum of the arrester sub-component is analyzed to obtain temperature information of the arrester sub-component.

Images