A fault diagnosis method, system, device and medium of a GIS switch device

Abstract

The present application relates to a kind of GIS switch equipment fault diagnosis method, system, equipment and medium.GIS switch equipment fault diagnosis method includes obtaining the multiple frames of GIS switch equipment thermograph, and pre-processing;According to DBSCAN unsupervised clustering algorithm, the multiple frames of thermograph after pre-processing are carried out image segmentation;The image obtained by segmentation is marked, and the switch region of GIS switch equipment is extracted, wherein, switch region includes the temperature difference information and coordinate value information of switch region;The image of switch region is carried out three primary color separation, determines the eigenvalue of image pixel, according to the eigenvalue of image pixel Time correlation analysis, determine the change of corresponding ratio based on the result of time correlation analysis and the change of corresponding ratio, judge the thermal fault condition of GIS switch equipment, help to ensure the thermal fault analysis result precision of GIS switch equipment, it is also favorable to improve detection efficiency.

Description

Technical Field

[0001] This invention relates to the field of power equipment technology, and in particular to a fault diagnosis method, system, device and medium for GIS switchgear. Background Technology

[0002] Gas-insulated switchgear (GIS) is a critical electrical device widely used in power systems for current switching, control, and protection. Because GIS equipment often operates in high-voltage and complex environments, its installation and maintenance requirements are extremely stringent. Tool management and safety monitoring are two crucial aspects of GIS equipment installation.

[0003] However, GIS is usually installed in a closed metal grounding enclosure. Because the metal grounding enclosure isolates the GIS from the external environment, it is difficult for maintenance personnel to know the relevant data of the GIS when it is currently working. This results in a lack of relevant data to assist maintenance personnel in GIS inspection, leading to low inspection efficiency, which needs further improvement.

[0004] Therefore, there is an urgent need to develop a means to effectively detect faults in GIS switchgear. Summary of the Invention

[0005] This application provides a fault diagnosis method, system, device, and medium for GIS switchgear to solve the problem of low detection efficiency in the prior art.

[0006] The fault diagnosis method for GIS switchgear provided in this application includes:

[0007] S1. Obtain multi-frame heat maps of GIS switchgear and perform preprocessing;

[0008] S2. Based on the DBSCAN unsupervised clustering algorithm, perform image segmentation on the preprocessed multi-frame heatmap;

[0009] S3. Mark the segmented image and extract the switching area of ​​the GIS switchgear, wherein the switching area includes temperature difference information and coordinate value information of the switching area;

[0010] S4. Perform three-primary-color separation on the image of the switching area to determine the feature values ​​of the image pixels, and perform time correlation analysis based on the feature values ​​of the image pixels to determine the change of the corresponding ratio.

[0011] S5. Based on the results of the time correlation analysis and the changes in the corresponding ratios, determine the thermal fault status of the GIS switchgear.

[0012] Furthermore, the feature values ​​of the image pixels include feature values ​​of different color channels. The process of performing three-primary-color separation on the image of the switching region to determine the feature values ​​of the image pixels, and performing time correlation analysis based on the feature values ​​of the image pixels to determine the changes in the corresponding ratios, specifically includes:

[0013] The image of the switching area is subjected to three primary colors to generate an RGB three-channel image. The average value of the pixels in each channel image is taken and the average value is used as the feature value of the frame image in the corresponding color channel to generate the signals tR(i), tG(i) and tB(i) of the green, blue and red channels.

[0014] The signals tR(i), tG(i), and tB(i) are standardized and separated from the blind source to generate three new signals. The signals are then identified and screened, and the desired signals are selected using time correlation analysis. The changes in the corresponding ratios are determined based on the selected signals.

[0015] Furthermore, the formula for calculating the time correlation is as follows:

[0016]

[0017] Where X and Y are the time series data of the two parameters whose time correlation is to be calculated, and N is the length of the time series.

[0018] Furthermore, the time correlation judgment specifically involves determining whether the calculated result of the time correlation is less than a preset threshold. If it is, it indicates that the two parameters are not correlated; otherwise, it indicates that the two parameters are correlated.

[0019] Furthermore, the standardization process for signals tR(i), tG(i), and tB(i) includes the following steps: Assuming that signals tR(i), tG(i), and tB(i) are real random variables with zero mean, the original signals are zero-mean processed using the following formula: t'x(i) = (tx(i) - μx) / σi, where x = 1, 2, 3, μi and σx are the mean and standard deviation of t(i), respectively. After standardization, t'(i) has a mean of zero and a variance of 1.

[0020] Furthermore, when averaging the pixels of each channel image, the specific steps include: constructing a spatial average sequence X = [xjR, xjG, xjB]Z, where xjR, xjG, and xjB are the R, G, and B channel signals of the pixels in region j after spatial averaging, and Z is the time point corresponding to the acquisition of each channel image.

[0021] This application also provides a fault diagnosis system for GIS switchgear, including:

[0022] The heat map acquisition module is used to acquire multiple frames of heat maps of GIS switchgear.

[0023] The image segmentation module is used to preprocess multiple frames of heatmaps and segment the preprocessed multiple frames of heatmaps according to the DBSCAN unsupervised clustering algorithm.

[0024] The label extraction module is used to label the segmented image and extract the switching area of ​​the GIS switchgear;

[0025] The analysis module is used to determine the feature values ​​of image pixels by performing three-primary-color separation on the image of the switching area, and to perform time correlation analysis on the feature values ​​of image pixels to determine the change of the corresponding ratio.

[0026] The fault diagnosis module is used to determine the thermal fault status of the GIS switchgear based on the results of the time correlation analysis and the change of the corresponding ratio.

[0027] This application also provides an electronic device, the device including a processor and a memory;

[0028] The memory is used to store program code and transmit the program code to the processor;

[0029] The processor is used to execute the above-described fault diagnosis method for GIS switchgear according to the instructions in the program code.

[0030] This application also provides a computer-readable storage medium for storing program code, wherein the instructions in the program code execute the above-described fault diagnosis method for GIS switchgear.

[0031] This application innovatively segments multi-frame heat maps of GIS switchgear to obtain temperature difference and coordinate information of the switch area, and performs three-primary-color separation on the images of the switch area. Based on the results of the time correlation analysis and the changes in the corresponding ratios, the thermal fault status of the GIS switchgear is determined. This provides a thermal fault analysis mechanism for GIS switchgear, helping to ensure the accuracy of thermal fault analysis results.

[0032] The beneficial technical effects of this application mainly include: performing image segmentation on multi-frame heat maps of GIS switchgear, obtaining temperature difference information and coordinate value information of the switch area, and performing time correlation analysis and corresponding ratio changes to determine the thermal fault status of GIS switchgear, thereby improving the accuracy of thermal fault detection of GIS switchgear, and enabling targeted handling of thermal fault status of GIS switchgear. Attached Figure Description

[0033] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0034] Figure 1 This is a flowchart of a fault diagnosis method for GIS switchgear provided in an embodiment of this application;

[0035] Figure 2 This is a system diagram of a fault diagnosis system for GIS switchgear provided in an embodiment of this application. Detailed Implementation

[0036] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0037] The flowchart shown in the attached diagram is for illustrative purposes only and does not necessarily include all content and operations / steps, nor does it necessarily have to be performed in the order described. For example, some operations / steps can be broken down, combined, or partially merged, so the actual execution order may change depending on the actual situation.

[0038] In the description of this application, the terms "first / second" are used merely to distinguish similar objects and do not represent a specific order of objects. It is understood that "first / second" may be interchanged in a specific order or sequence where permitted, so that the embodiments of this application described herein can be implemented in an order other than that illustrated or described herein.

[0039] It should be understood that the orientation or positional relationship is based on the orientation or positional relationship shown in the accompanying drawings. These orientation terms are only for the convenience of describing this application and simplifying the description, and are not intended to indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as limiting this application.

[0040] This application provides a fault diagnosis method for GIS switchgear. Please refer to the following embodiments. Figure 1 This figure is a flowchart of an embodiment of this application. The following is in conjunction with... Figure 1 This application provides a detailed description of a fault diagnosis method for GIS switchgear.

[0041] The fault diagnosis method for GIS switchgear according to the embodiments of this application includes:

[0042] S1. Obtain multi-frame heat maps of GIS switchgear and perform preprocessing;

[0043] Specifically, first, multiple frames of heat maps of the GIS switchgear are acquired, for example, using an infrared thermal imager. The acquired heat maps may then require preprocessing, such as noise reduction and enhancement, to improve image quality.

[0044] S2. Based on the DBSCAN unsupervised clustering algorithm, perform image segmentation on the preprocessed multi-frame heatmap;

[0045] Image segmentation was performed on the preprocessed heatmap using the DBSCAN (Density-Based Spatial Clustering of Applications with Noise) unsupervised clustering algorithm. This algorithm clusters pixels into different regions based on pixel density and distance.

[0046] It should be noted that the DBSCAN unsupervised clustering algorithm applies spatial clustering based on density-based noise. "Unsupervised" means it does not use pre-labeled targets to cluster data points. Clustering refers to the attempt to group similar data points into manually determined groups or clusters. Compared to popular clustering algorithms such as K-Means and hierarchical clustering, K-Means is particularly susceptible to outliers. When the algorithm traverses centroids, outliers significantly affect the movement of centroids before achieving stability and convergence. K-Means also faces the challenge of accurate clustering when cluster sizes and densities vary. K-Means can only apply to spherical clusters; its accuracy is affected if the data is not spherical. K-Means requires us to first select the number of clusters we want to find.

[0047] The implementation process of DBSCAN is as follows:

[0048] First, we define the epsilon and the minimum point, two parameters required when applying the DBSCAN algorithm, and some additional parameters. Epsilon (ε): The maximum radius of the community. Data points are considered to belong to the same class if the distance between them is less than or equal to the specified epsilon. In other words, it's the distance DBSCAN uses to determine whether two points are similar and belong to the same class. A larger epsilon will produce a larger cluster (containing more data points), and a smaller epsilon will build smaller clusters. Generally, we prefer smaller values ​​because we only need a small subset of data points to be within each other's distance. However, if it's too small, you'll end up with increasingly smaller clusters.

[0049] Minimum Points (minPts): A neighborhood of minPts values ​​within a given radius is considered a cluster. Remember that the initial point is included in minPts. A lower minPts helps the algorithm build more clusters with more noise or outliers. A higher minPts ensures more robust clusters, but if the clusters are too large, smaller clusters will be merged into larger ones. If "minPts" = 4, then any four or more points within a certain distance from each other are considered a cluster. DBSCAN Core Point: The minimum number of data points within a core data point's nearest neighbor distance.

[0050] S3. Mark the segmented image and extract the switching area of ​​the GIS switchgear, wherein the switching area includes temperature difference information and coordinate value information of the switching area;

[0051] The DBSCAN unsupervised clustering algorithm is used for image segmentation. The segmented binarized image is labeled using the four-neighborhood region growing method, and the switching area of ​​the GIS switch equipment is extracted, including the area, temperature difference information and coordinate value information of the switching area, thus obtaining the switching area and its coordinate value.

[0052] Temperature difference information is used to describe the temperature difference between any detection area in the two heat maps, while coordinate information is used to describe the size difference, shape change, and rate of change of the range of any detection area in the two heat maps.

[0053] S4. Perform three-primary-color separation on the image of the switching area to determine the feature values ​​of the image pixels, and perform time correlation analysis based on the feature values ​​of the image pixels to determine the change of the corresponding ratio.

[0054] Specifically, the principle of three primary colors lies in the fact that the three primary colors are independent of each other; none of them can be obtained by mixing the other two. Most colors in nature can be obtained by mixing the three primary colors in a certain proportion, or most colors can be decomposed into three primary colors. The mixing ratio of the three primary colors determines the hue and saturation of the mixed color, and the brightness of the mixed color is equal to the sum of the brightness of each primary color that constitutes the mixed color. By performing three primary color separation on the image of the switching region, the feature values ​​of the image pixels are determined. Based on these feature values, time correlation analysis is performed to determine the changes in the corresponding ratios. This can help to understand the temperature changes in the switching region at different points in time.

[0055] S5. Based on the results of time correlation analysis and the changes in the corresponding ratios, determine the thermal fault status of GIS switchgear.

[0056] Based on the results of time correlation analysis and the changes in corresponding ratios, the thermal fault status of GIS switchgear is determined. If the analysis results show abnormal temperature changes or large ratio changes in the switch area, it may indicate a thermal fault in the equipment.

[0057] It should be noted that existing methods for detecting thermal fault conditions in GIS switchgear are inefficient, rely on manual judgment of results, require highly experienced personnel, and have poor accuracy in determining results. This affects the normal operation of GIS and consequently the stable operation of the power grid.

[0058] This application innovatively segments multi-frame heat maps of GIS switchgear to obtain temperature difference and coordinate information of the switch area, and performs three-primary-color separation on the images of the switch area. Based on the results of the time correlation analysis and the changes in the corresponding ratios, the thermal fault status of the GIS switchgear is determined. This provides a thermal fault analysis mechanism for GIS switchgear, which helps ensure the accuracy of thermal fault analysis results.

[0059] By segmenting the multi-frame heat map of GIS switchgear, the temperature difference and coordinate information of the switch area are obtained. The results of time correlation analysis and the change of corresponding ratios are used to determine the thermal fault status of GIS switchgear, thereby improving the accuracy of thermal fault detection of GIS switchgear and enabling targeted handling of thermal fault status of GIS switchgear.

[0060] In some embodiments, the feature values ​​of the image pixels include feature values ​​of different color channels. The process of performing three-primary-color separation on the image of the switching region to determine the feature values ​​of the image pixels, and performing time correlation analysis based on the feature values ​​of the image pixels to determine the changes in corresponding ratios, specifically includes:

[0061] S41. Perform three-primary-color separation on the image of the switching area to generate an RGB three-channel image, and take the average value of the pixels of each channel image. Use the average value as the feature value of the frame image in the corresponding color channel to generate the signals tR(i), tG(i) and tB(i) of the green, blue and red channels.

[0062] S42. Standardize and separate the signals tR(i), tG(i) and tB(i) to generate three new signals. Then, identify and screen the signals, use time correlation analysis to select the required signals, and determine the change of the corresponding ratio based on the selected signals.

[0063] In some embodiments, when averaging the pixels of each channel image, the following steps are specifically included: constructing a spatial average sequence X = [xjR, xjG, xjB]Z, where xjR, xjG, and xjB are the R, G, and B channel signals of the pixels in region j after spatial averaging, and Z is the time point corresponding to the acquisition of each channel image.

[0064] In some embodiments, the standardization of signals tR(i), tG(i), and tB(i) includes the following steps: Assuming that signals tR(i), tG(i), and tB(i) are real random variables with zero mean, the original signals are zero-mean processed using the following formula: t'x(i) = (tx(i) - μx) / σi, where x = 1, 2, 3, μi and σx are the mean and standard deviation of t(i), respectively, and the mean of t'(i) after standardization is zero and the variance is 1.

[0065] For example, signals tR(i), tG(i), and tB(i) are real random variables with zero mean. The signals are zero-mean processed using the following formula: x'i(t) = (xi(t) - μi) / σi, where i = 1, 2, 3, μi and σi are the mean and standard deviation of x(t), respectively. After standardization, x'(t) has a mean of zero and a variance of 1. The obtained standardized signal is then used as the input signal for blind source separation, generating three new signals, denoted as Ta(t), Tb(t), and Tc(t). The separated signals Ta(t), Tb(t), and Tc(t) are then identified and filtered, and the desired signal is selected using a correlation analysis method based on channel signals. In this embodiment, correlation analysis is used to perform correlation analysis between the three independent signals after blind source separation and the green channel signal in the source signal. This involves studying the similarity between the two signals and identifying the most relevant independent signal as the desired signal, thus completing the signal selection process. Alternatively, correlation analysis methods using two other color channel signals can also be used to select the desired signal in this embodiment.

[0066] In some embodiments, the formula for calculating the time correlation is:

[0067]

[0068] Where X and Y are the time series data of the two parameters whose time correlation is to be calculated, and N is the length of the time series.

[0069] Correlation analysis refers to the analysis of two or more correlated variables to measure the degree of correlation between the two factors. For correlated elements to be valid, there must be a certain relationship or probability between them. The calculation of time correlation is usually done by analyzing changes in data at different points in time. Data is collected at different time points, and this data can be continuous or discrete. The correlation coefficient is then calculated, where a value between -1 and 1 indicates the degree of linear correlation between the two variables. A correlation coefficient of 1 indicates a perfect positive correlation; -1 indicates a perfect negative correlation; and 0 indicates no correlation.

[0070] When calculating time correlation, different correlation coefficient calculation methods can be used, such as Pearson correlation coefficient, Spearman rank correlation coefficient, or Kendall rank correlation coefficient. By calculating time correlation, we can understand the trend and pattern of data changes over time, and thus determine whether time correlation exists. If the time correlation is strong, it indicates that the data changes with time according to certain patterns and trends, which can be used to predict and analyze future development trends.

[0071] By performing correlation analysis on the color channel signals, all independent signals are sorted according to their correlation with the desired signal, and the independent signal with the highest correlation is selected as the desired signal. This allows us to filter out the independent signals that are most relevant to the desired signal, which helps to improve the accuracy and reliability of the signal.

[0072] In some embodiments, the time correlation determination specifically involves determining whether the calculated result of the time correlation is less than a preset threshold. If it is, it indicates that the two parameters are not correlated; otherwise, it indicates that the two parameters are correlated.

[0073] Specifically, time correlation is determined based on a preset threshold. For example, if the calculated time correlation result is less than the preset threshold, it indicates that the two parameters are uncorrelated, and the system proceeds to GIS switchgear fault judgment. If the judgment result is that they are correlated, no warning is issued, and monitoring continues.

[0074] like Figure 2As shown, this embodiment of the invention provides a fault diagnosis system 100 for GIS switchgear, including a heat map acquisition module 110, an image segmentation module 120, a marker extraction module 130, an analysis module 140, and a fault judgment module 150. The heat map acquisition module 110 acquires multiple frames of heat maps of the GIS switchgear; the image segmentation module 120 preprocesses the multiple frames of heat maps and segments them using the DBSCAN unsupervised clustering algorithm; the marker extraction module 130 marks the segmented images and extracts the switching areas of the GIS switchgear; the analysis module 140 performs three-primary-color separation on the images of the switching areas to determine the feature values ​​of image pixels and performs time correlation analysis based on the feature values ​​to determine changes in corresponding ratios; the fault judgment module 150 determines the thermal fault status of the GIS switchgear based on the results of the time correlation analysis and the changes in corresponding ratios.

[0075] Specifically, the heat map acquisition module 110 acquires multiple frames of heat maps of the GIS switchgear using an infrared thermal imager. The acquired heat maps may require preprocessing operations, such as noise reduction and enhancement, to improve image quality.

[0076] The image segmentation module 120 uses the DBSCAN (Density-Based Spatial Clustering of Applications with Noise) unsupervised clustering algorithm to segment the preprocessed heatmap, which can cluster the pixels in the image according to the density and distance of the pixels, and divide the pixels into different regions.

[0077] The binary image segmented by the label extraction module 130 is labeled using the four-neighborhood region growing method, and the switching area of ​​the GIS switchgear is extracted, including the area, temperature difference information, and coordinate value information of the switching area, thus obtaining the switching area and its coordinate values. The temperature difference information is used to describe the temperature difference between any detection area in the two heat maps, and the coordinate value information is used to describe the size difference, shape change degree, and rate of change of the range of any detection area in the two heat maps.

[0078] The analysis module 140 performs three-primary-color separation on the image of the switching area, determines the feature values ​​of the image pixels, and performs time correlation analysis based on these feature values ​​to determine the changes in the corresponding ratios, which is helpful for analyzing the temperature changes of the switching area at different time points.

[0079] The fault diagnosis module 150 determines the thermal fault status of the GIS switchgear based on the results of time correlation analysis and changes in the corresponding ratios. If the analysis results show abnormal temperature changes or large ratio changes in the switch area, it may indicate a thermal fault in the equipment.

[0080] For example, environmental factors can cause thermal failures in GIS equipment. Because GIS equipment operates in harsh environments for extended periods, this exacerbates the heating of GIS switching devices, leading to abnormal overheating. Alternatively, sudden changes in power load can have adverse effects. For instance, if the power load remains within a normal range, the resulting temperature rise in GIS equipment generally won't exceed a preset threshold. However, a sharp increase in current due to short circuits or other reasons can significantly increase the power load, and this high-load surge will accelerate the temperature rise of the GIS switching devices. Excessive temperature can easily lead to problems such as oxidation, deformation, and accelerated aging of the GIS switching device materials. The fault diagnosis module 150 issues different warning messages based on the temperature changes in different switching areas, such as flashing red lights, voice alarms, and yellow warning lights.

[0081] This invention also provides an electronic device, which includes a processor and a memory; the memory is used to store program code and transmit the program code to the processor; the processor is used to execute the above-described fault diagnosis method for GIS switchgear according to the instructions in the program code.

[0082] This invention also provides a computer-readable storage medium for storing program code, wherein the instructions in the program code execute the aforementioned fault diagnosis method for GIS switchgear.

[0083] It should be noted that the storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, ROM (Read-Only Memory), RAM (Random Access Memory), magnetic disks, or optical disks.

[0084] Those skilled in the art will understand that all or some of the steps, systems, and apparatuses disclosed above, and their functional modules / units, can be implemented as software, firmware, hardware, or suitable combinations thereof. In hardware implementations, the division between functional modules / units mentioned above does not necessarily correspond to the division of physical components; for example, a physical component may have multiple functions, or a function or step may be performed collaboratively by several physical components. Some or all physical components may be implemented as software executed by a processor, such as a central processing unit, digital signal processor, or microprocessor, or as hardware, or as an integrated circuit, such as an application-specific integrated circuit (ASIC). Such software can be distributed on a computer-readable storage medium, which may include computer-readable storage media (or non-transitory media) and communication media (or transient media).

[0085] As is known to those skilled in the art, the term computer-readable storage medium includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, program modules, or other data). Computer-readable storage media includes, but is not limited to, RAM, ROM, EEPROM, flash memory or other memory technologies, CD-ROM, digital versatile disc (DVD) or other optical disc storage, magnetic cartridges, magnetic tape, disk storage or other magnetic storage devices, or any other medium that can be used to store desired information and is accessible to a computer. Furthermore, it is known to those skilled in the art that communication media typically contain computer-readable instructions, data structures, program modules, or other data in modulated data signals such as carrier waves or other transmission mechanisms, and may include any information delivery medium.

[0086] For example, the computer-readable storage medium may be an internal storage unit of the electronic device described in the foregoing embodiments, such as a hard disk or memory of the electronic device. The computer-readable storage medium may also be an external storage device of the electronic device, such as a plug-in hard disk, smart media card (SMC), secure digital card (SD), flash card, etc., provided on the electronic device.

[0087] The above are merely specific embodiments of this application, enabling those skilled in the art to understand or implement this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features claimed herein.

Claims

1. A fault diagnosis method for GIS switchgear, characterized in that, include: S1. Obtain multi-frame heat maps of GIS switchgear and perform preprocessing; S2. Based on the DBSCAN unsupervised clustering algorithm, perform image segmentation on the preprocessed multi-frame heatmap; S3. Mark the segmented image and extract the switching area of ​​the GIS switchgear, wherein the switching area includes temperature difference information and coordinate value information of the switching area; S4. Perform three-primary-color separation on the image of the switching area to determine the feature values ​​of the image pixels, and perform time correlation analysis based on the feature values ​​of the image pixels to determine the change of the corresponding ratio. S5. Based on the results of the time correlation analysis and the changes in the corresponding ratio, determine the thermal fault status of the GIS switchgear. The feature values ​​of the image pixels include feature values ​​of different color channels. The process involves performing three-primary-color separation on the image of the switching region to determine the feature values ​​of the image pixels, and then performing time correlation analysis based on the feature values ​​of the image pixels to determine the changes in the corresponding ratios. Specifically, this includes: The image of the switching area is subjected to three primary colors to generate an RGB three-channel image. The average value of the pixels in each channel image is taken and the average value is used as the feature value of the frame image in the corresponding color channel to generate the signals tR(t), tG(t) and tB(t) of the green, blue and red channels. The signals tR(t), tG(t), and tB(t) are standardized and separated from the blind source to generate three new signals. The signals are then identified and screened, and the desired signals are selected using time correlation analysis. The changes in the corresponding ratios are determined based on the selected signals. The formula for calculating the time correlation is: Where X and Y are the time series data of the two parameters whose time correlation is to be calculated, and N is the length of the time series; The time correlation determination specifically involves determining whether the calculated result of the time correlation is less than a preset threshold. If it is, it indicates that the two parameters are not correlated; otherwise, it indicates that the two parameters are correlated.

2. The fault diagnosis method according to claim 1, characterized in that, The standardization process for the preprocessed original signal includes the following steps: Assuming the original signal is a real random variable with a mean of zero, the original signal is zero-mean processed using the following formula: x'i(t)=(xi(t)-μi) / σi, where i=1,2,3, μi and σi are the mean and standard deviation of x(t) respectively, and the mean of x'(t) after standardization is zero and the variance is 1.

3. The fault diagnosis method according to claim 1, characterized in that, The process of averaging the pixels of each channel image includes the following steps: constructing a spatial averaging sequence X = [xiR, xiG, xiB]Z, where xiR, xiG, and xiB are the R, G, and B channel signals of the pixels in region i after spatial averaging, and Z is the time point corresponding to the acquisition of each channel image.

4. A fault diagnosis system for GIS switchgear, characterized in that, include: A heat map acquisition module is used to execute the process described in any one of claims 1 to 3 for acquiring multiple frames of heat maps of GIS switchgear and performing preprocessing. The image segmentation module is used to perform image segmentation of the preprocessed multi-frame heatmap according to the DBSCAN unsupervised clustering algorithm as described in any one of claims 1 to 3; The label extraction module is used to perform the process of labeling the segmented image as described in any one of claims 1 to 3 and extracting the switching area of ​​the GIS switchgear, wherein the switching area includes temperature difference information and coordinate value information of the switching area; The analysis module is used to perform the process described in any one of claims 1 to 3, which involves separating the three primary colors of the image of the switching area, determining the feature values ​​of the image pixels, performing time correlation analysis based on the feature values ​​of the image pixels, and determining the change of the corresponding ratio. The fault diagnosis module is used to execute the process described in any one of claims 1 to 3, which is based on the change in the result of the time correlation analysis and the corresponding ratio, to determine the thermal fault status of the GIS switchgear.

5. An electronic device, characterized in that, The device includes a processor and a memory; The memory is used to store program code and transmit the program code to the processor; The processor is used to execute the fault diagnosis method for GIS switchgear according to any one of the claims 1-3 according to the instructions in the program code.

6. A computer-readable storage medium, characterized in that, The computer-readable storage medium is used to store program code, wherein the instructions in the program code execute the fault diagnosis method for GIS switchgear according to any one of claims 1-3.

Images