GIS fault diagnosis method, device and equipment based on multilayer feature imaging

Abstract

The invention belongs to the technical field of GIS fault diagnosis, and provides a GIS fault diagnosis method, device and equipment based on multilayer feature imaging. The method comprises the following steps: acquiring spatial distribution information, expressed feature dimensions and perceived GIS (Geographic Information System) operation state information of multiple types of sensors which are arranged in advance, and constructing a feature map of each sensor; the method comprises the following steps: converting feature maps of sensors in the same feature dimension into corresponding gray values, mapping the corresponding gray values to corresponding pixel points in a GIS cavity image template according to spatial distribution information of the sensors, generating complete gray images of all single feature dimensions, and then randomly superposing the complete gray images to form a multi-layer feature image; according to the method, the preset useful features are extracted from the multi-layer feature image, whether the GIS has a fault or not and the corresponding fault type are predicted according to the mapping relation between the preset useful features and the operation state of the GIS, and the accuracy and practicability of GIS equipment fault diagnosis are improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of GIS fault diagnosis, and in particular relates to a GIS fault diagnosis method, device and equipment based on multi-layer feature imaging. Background Art

[0002] The statements in this section merely provide background information related to the present invention and do not necessarily constitute prior art.

[0003] GIS (gas-insulated switchgear) is a crucial component in power systems, involved in key areas such as high-voltage grid switching and control, as well as ensuring power system reliability and safety. Traditional switchgear in high-voltage grids is bulky and difficult to maintain. GIS replaces traditional air or solid insulating materials with gas insulation, effectively reducing equipment size and improving insulation performance.

[0004] The current technical solutions used for GIS fault diagnosis include: 1) Fault diagnosis methods based on traditional indicators. These methods usually rely on traditional electrical indicators such as voltage, current, and temperature, and perform fault diagnosis by comparing the changes in these indicators. The disadvantage is that these methods lack comprehensive analysis of fault characteristics in multiple dimensions and cannot fully capture the characteristics of complex faults, resulting in low diagnostic accuracy. 2) Local detection sensor data analysis methods. These methods mainly rely on the monitoring and analysis of local sensor data, such as the temperature and pressure data of a certain component. However, these methods ignore the overall system status and the mutual influence between components, which may lead to misjudgment or omission of the overall system status. 3) Fault diagnosis methods based on statistical models. These methods identify potential fault modes by statistically analyzing the distribution and trends of data. The disadvantage is that statistical models may not be able to handle system complexity and the influence of multiple factors. The diagnostic effect of complex faults may be poor, and it is easy to misjudge or miss some fault modes.

[0005] In summary, the current GIS fault diagnosis methods have the problems of high data processing complexity, large time and resources required for analysis, and low stability and reliability of fault detection results. Summary of the Invention

[0006] In order to solve the problems of high data processing complexity, large time and resources required for analysis, and low stability and reliability of fault detection results in current GIS fault diagnosis methods, the present invention provides a GIS fault diagnosis method, device and equipment based on multi-layer feature imaging. Through a multi-feature image fusion method, it is committed to optimizing the fault diagnosis process of the GIS system, ensuring the effective integration and analysis of multiple monitoring variables, and significantly improving the accuracy of fault detection, while effectively reducing the complexity of data processing and the time and resources required for analysis.

[0007] In order to achieve the above object, the present invention adopts the following technical solutions:

[0008] A first aspect of the present invention provides a GIS fault diagnosis method based on multi-layer feature imaging.

[0009] In one or more embodiments, a GIS fault diagnosis method based on multi-layer feature imaging is provided, comprising:

[0010] Obtain spatial distribution information, expressed feature dimensions, and perceived GIS operation status information of pre-deployed multi-type sensors;

[0011] Based on the GIS operation status information perceived by each sensor, a feature map of each sensor is constructed;

[0012] Based on the spatial distribution information of the sensor and the feature dimension to be expressed, the size of the corresponding feature map is determined, and then the feature map is converted into the corresponding grayscale value, which is then mapped to the corresponding pixel points of the preset image template to generate a complete grayscale image of all single feature dimensions;

[0013] Randomly superimpose all complete grayscale images of a single feature dimension to form a multi-layer feature image;

[0014] Preset useful features are extracted from the multi-layer feature images, and the mapping relationship between the preset useful features and the operating status of the GIS is used to predict whether the GIS has a fault and the corresponding fault type.

[0015] As an implementation method, the electromagnetic field, acoustic field and thermal field distribution inside the GIS are simulated to optimize the sensor layout points with the purpose of maximizing the capture of fault signals, thereby obtaining the spatial distribution information of the sensors.

[0016] The advantage of the above technical solution is that it takes into account the various signals generated by partial discharge caused by possible faults inside the GIS chamber under certain conditions, and reasonably arranges the positions of the sensors based on the characteristics of various sensors to ensure that the sensor's monitoring range can cover the entire GIS chamber.

[0017] As an implementation method, each type of sensor is evenly distributed in the GIS chamber.

[0018] The advantage of the above technical solution is that it can effectively monitor fault signals without interfering with the normal operation of the equipment, and can effectively collect information from various parts.

[0019] As an implementation method, there is a one-to-one correspondence between the type of sensor and the characteristic dimension.

[0020] The advantage of the above technical solution is that, since different types of faults generate different signal characteristics, using different types of sensors as a feature dimension can ensure the effective integration and analysis of multiple monitoring variables and improve the accuracy of GIS fault detection.

[0021] As an implementation method, the pixel size of the GIS cavity image template is determined by the aspect ratio of the GIS cavity.

[0022] The advantage of the above technical solution is that it can adapt to the actual chamber conditions and lay the foundation for accurately detecting GIS faults.

[0023] As an implementation method, before constructing the feature maps of each sensor, the following steps are further included:

[0024] The GIS operation status information perceived by each sensor is time aligned, data cleaned, missing information supplemented and normalized.

[0025] The advantage of the above technical solution is that it can ensure that the data of each sensor reflects the equipment status at the same time and guarantee the accuracy of the perception data, thereby providing a data basis for subsequent GIS fault diagnosis.

[0026] As an implementation method, the fault prediction results of GIS, the actual state operation information and the corresponding sensor related information are added to the data set as training samples to update the mapping relationship between the preset useful features and the operation state of GIS.

[0027] The advantage of the above technical solution is that it can adapt to new failure modes and equipment changes and improve the accuracy of GIS fault diagnosis.

[0028] A second aspect of the present invention provides a GIS fault diagnosis device based on multi-layer feature imaging.

[0029] In one or more embodiments, a GIS fault diagnosis device based on multi-layer feature imaging includes:

[0030] The sensor-related information acquisition module is used to obtain the spatial distribution information, characteristic dimensions of the expression, and perceived GIS operation status information of pre-deployed multi-type sensors;

[0031] The sensor feature map construction module is used to construct the feature map of each sensor based on the GIS operation status information perceived by each sensor;

[0032] The single feature image generation module is used to convert the feature maps of each sensor of the same feature dimension into corresponding grayscale values, map the corresponding grayscale values ​​to the corresponding pixel points in the GIS chamber image template according to the spatial distribution information of the sensor, and generate a complete grayscale image of all single feature dimensions;

[0033] A multi-layer feature image generation module, which is used to randomly superimpose all complete grayscale images of a single feature dimension to form a multi-layer feature image;

[0034] The GIS operation status prediction module is used to extract preset useful features from multi-layer feature images, and use the mapping relationship between the preset useful features and the operation status of the GIS to predict whether the GIS has a fault and the corresponding fault type.

[0035] A third aspect of the present invention provides a computer-readable storage medium.

[0036] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the above-mentioned GIS fault diagnosis method based on multi-layer feature imaging.

[0037] A fourth aspect of the present invention provides an electronic device.

[0038] An electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, the steps of the GIS fault diagnosis method based on multi-layer feature imaging as described above are implemented.

[0039] Compared with the prior art, the present invention has the following beneficial effects:

[0040] The present invention proposes a GIS fault diagnosis method based on multi-feature fusion imaging, which solves the problem of multivariate correlation in gas-insulated switchgear fault diagnosis. It uses the spatial distribution information of multiple types of sensors and the feature dimensions that need to be expressed to form a feature map and convert it into corresponding grayscale values, which are mapped to the corresponding pixel points of a preset image template to generate a complete grayscale image of all single feature dimensions, and then superimpose to obtain multi-layer feature images. Preset useful features are then extracted from the multi-layer feature images, and the mapping relationship between the preset useful features and the operating status of the GIS is used to predict whether the GIS has a fault and the corresponding fault type, thereby ensuring the effective integration and analysis of multiple monitoring variables, improving the accuracy and practicality of GIS equipment fault diagnosis, meeting the high performance requirements of real engineering applications, and being able to effectively improve the stability and reliability of the GIS system, playing a key role in the practical application of the power industry. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] The accompanying drawings, which constitute a part of the present invention, 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 improper limitations on the present invention.

[0042] Figure 1 is a schematic diagram of an electronic device according to an embodiment of the present invention;

[0043] Figure 2 1 is a flow chart of a GIS fault diagnosis method based on multi-layer feature imaging according to an embodiment of the present invention;

[0044] Figure 3 Schematic diagram of possible defects inside the GIS chamber according to an embodiment of the present invention;

[0045] Figure 4 This is a schematic diagram of the placement of sensors inside the GIS chamber according to an embodiment of the present invention;

[0046] Figure 5 is a flow chart of multi-layer feature image generation according to an embodiment of the present invention;

[0047] Figure 6 1 is a schematic structural diagram of a GIS fault diagnosis device based on multi-layer feature imaging according to an embodiment of the present invention;

[0048] Figure 7 is a flow chart of machine learning model training in an embodiment of the present invention;

[0049] Figure 8 This is a grayscale heat map of the ultrasonic sensor distribution positions and measured results according to an embodiment of the present invention. DETAILED DESCRIPTION

[0050] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0051] It should be noted that the following detailed descriptions are illustrative and intended to provide further explanation of the present invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which the present invention belongs.

[0052] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present invention. As used herein, unless the context clearly indicates otherwise, the singular form is intended to include the plural form. In addition, it should be understood that when the terms "comprise" and / or "include" are used in this specification, they indicate the presence of features, steps, operations, devices, components and / or combinations thereof.

[0053] In power systems, the role of GIS is primarily reflected in spatial efficiency and reliability, anti-interference capabilities, energy conservation and environmental protection, long-distance communication, and intelligence. GIS equipment is small and lightweight, allowing for high-density deployment within limited spaces, improving the spatial utilization of power equipment while reducing the power system's footprint and enhancing system reliability and stability. GIS equipment exhibits excellent anti-interference capabilities, effectively preventing the impact of external environmental factors on equipment and ensuring the normal operation of the power system. Furthermore, by utilizing environmentally friendly gas insulation media such as sulfur hexafluoride, GIS can significantly reduce the amount of insulation used, minimizing adverse environmental impacts and meeting modern environmental protection requirements. GIS equipment also enables remote monitoring and communication, enhancing the intelligence level of power systems, helping to promptly identify and resolve problems and improving power system operational efficiency.

[0054] The application of GIS in power systems has become an indispensable component, not only improving their efficiency and reliability but also providing crucial support for their safe operation. With the continuous advancement of science and technology, GIS technology is also evolving and improving to meet the increasingly complex and challenging demands of power systems. The widespread use of GIS equipment will continue to drive the power industry forward, achieving a more efficient and stable power supply.

[0055] In order to solve the problems in the background technology of the current GIS fault diagnosis method, such as high data processing complexity, large time and resources required for analysis, and low stability and reliability of fault detection results, the present invention provides a GIS fault diagnosis method, device and equipment based on multi-layer feature imaging.

[0056] Reference Figure 1 , a schematic diagram of an electronic device is given. It should be noted that, Figure 1 The electronic device 100 shown is only an example and should not limit the functions and scope of use of the embodiments of the present invention.

[0057] like Figure 1 As shown, the electronic device 100 includes a central processing unit (CPU) 101, which can perform various appropriate actions and processes according to the program stored in the read-only memory (ROM) 102 or the program loaded from the storage part 108 into the random access memory (RAM) 103. Various programs and data required for system operation are also stored in the RAM 103. The central processing unit 101, the ROM 102, and the RAM 103 are connected to each other via a bus 104. An input / output (I / O) interface 105 is also connected to the bus 104.

[0058] The following components are connected to the I / O interface 105: an input section 106 including a keyboard, a mouse, and the like; an output section 107 including devices such as a cathode ray tube (CRT), a liquid crystal display (LCD), and a speaker; a storage section 108 including devices such as a hard disk; and a communication section 109 including a network interface card such as a local area network (LAN) card or a modem. The communication section 109 performs communication processing via a network such as the Internet. A drive 110 is also connected to the I / O interface 105 as needed. A removable medium 111, such as a magnetic disk, an optical disk, a magneto-optical disk, or a semiconductor memory, is installed in the drive 110 as needed, so that computer programs read therefrom can be installed into the storage section 108 as needed.

[0059] The electronic device 100 mentioned in the embodiment of the present invention may include an electronic device of a power system or other Internet device running an APP application, and communicate with a federal server via the Internet.

[0060] When the central processing unit 101 in the electronic device of this embodiment executes the program, the following is achieved: Figure 2 The steps in the GIS fault diagnosis method based on multi-layer feature imaging are shown.

[0061] Figure 2 FIG. 1 is a flow chart of a GIS fault diagnosis method based on multi-layer feature imaging according to an embodiment of the present invention. Figure 2 The GIS fault diagnosis method based on multi-layer feature imaging in this embodiment may include:

[0062] S201, obtaining spatial distribution information, expressed feature dimensions, and perceived GIS operation status information of pre-deployed multiple types of sensors.

[0063] In the specific implementation process, Figure 3 The possible defects inside the GIS cavity are listed. These defects include but are not limited to metal tip defects, metal suspension defects, metal adhesion defects, air bubble defects inside the basin insulator, and poor conductor contact defects.

[0064] Different fault types generate distinct signal characteristics. For example, metal tip defects typically generate high-frequency electromagnetic waves, while bubbles within a basin insulator can cause ultrasonic signal anomalies. Considering the various signals generated by partial discharges caused by potential faults within the GIS equipment chamber under certain conditions, and rationally arranging sensors based on their characteristics, ensures that the sensor monitoring range covers the entire component. For example, ultra-high frequency (UHF) sensors are used for ultra-high frequency signals, ultrasonic sensors for ultrasonic signals, fluorescent fiber sensors for optical signals, vibration sensors for vibration signals, and infrared sensors for temperature signals.

[0065] In one or more embodiments, simulations of the electromagnetic, acoustic, and thermal field distribution within the GIS are employed to optimize sensor placement with the goal of maximizing the capture of fault signals, thereby obtaining information on the spatial distribution of sensors. For example, computer-aided design (CAD) software and simulation tools, such as ANSYS and COMSOL Multiphysics, are used to perform virtual simulations of sensor layout. These tools can predict and optimize sensor performance under real-world operating conditions, ensuring comprehensive monitoring coverage of all components within the GIS.

[0066] In some other embodiments, the structural characteristics of the equipment and the convenience of operation and maintenance must also be considered to ensure that the sensor installation position can effectively monitor fault signals without interfering with the normal operation of the equipment.

[0067] In this embodiment, if Figure 4 As shown, the sensors are evenly distributed, which can effectively collect information from various parts.

[0068] In the specific implementation process, there is a one-to-one correspondence between sensor types and feature dimensions. For example, five feature dimensions are selected: infrared sensors, vibration sensors, fiber optic sensors, ultrasonic sensors, and ultra-high frequency sensors. Because different fault types produce distinct signal characteristics, using different sensor types as a feature dimension ensures the effective integration and analysis of multiple monitoring variables, improving the accuracy of GIS fault detection.

[0069] S202: construct a feature map of each sensor based on the GIS operation status information sensed by each sensor.

[0070] Before constructing the characteristic maps for each sensor, the GIS operating status information sensed by each sensor is time-aligned, cleaned, missing data is supplemented, and normalized. This ensures that each sensor's data reflects the device status at the same moment and guarantees the accuracy of the sensed data, providing a data foundation for subsequent GIS fault diagnosis.

[0071] During data cleaning of GIS operational status information sensed by various sensors, filtering techniques are used to remove data noise introduced by electromagnetic interference, environmental noise, and other factors. Outlier detection algorithms (such as statistical outlier detection and density-based local outlier factor (LOF) detection) are employed to identify and remove outliers, such as sudden data jumps caused by equipment failures or data anomalies introduced by operational errors.

[0072] In some specific implementations, missing data can be supplemented by linear interpolation.

[0073] Normalize all feature data to between 0 and 1 using the formula:

[0074]

[0075] The x in the formula is the original data, x min and x max Represent the minimum and maximum values ​​of the characteristic data respectively. This step is to eliminate the influence of different dimensions and numerical ranges on subsequent analysis.

[0076] S203, based on the spatial distribution information of the sensor and the feature dimension to be expressed, determine the size of the corresponding feature map, then convert the feature map into the corresponding grayscale value, and then map it to the corresponding pixel points of the preset image template to generate a complete grayscale image of all single feature dimensions.

[0077] In the specific implementation process, the normalized data is then scaled to a range of 0-255, and the normalized data values ​​are converted to corresponding grayscale levels. 0 is the darkest color, indicating the smallest eigenvalue, while 255 is the lightest color, indicating the largest eigenvalue.

[0078] The pixel size of the GIS chamber image template is determined by the aspect ratio of the GIS chamber. This adapts to the actual chamber conditions and lays the foundation for accurate GIS fault detection.

[0079] In this embodiment, the GIS chamber image template is a 160*120 pure black image. Figure 8 , 16 groups of ultrasonic sensors are evenly arranged on the plane of the GIS chamber, Figure 8 The black dot in the figure is the position of the infrared sensor in the chamber plane, and the surrounding grayscale blocks are the pixel mapping sets obtained after the temperature values ​​measured by the sensor are processed.

[0080] S204, randomly superimposing all complete grayscale images of a single feature dimension to form a multi-layer feature image.

[0081] Combine Figure 5 All complete grayscale images of a single feature dimension are randomly superimposed, and the single-layer feature images representing different sensor data are connected in a certain order to form a multi-layer feature image. Each layer is equivalent to an independent feature channel and can be represented as an image layer of different colors or grayscale levels.

[0082] The order of images in a data set must be consistent. For example, all images in the same data set must be stacked from top to bottom in the order of infrared sensor, vibration sensor, fiber optic sensor, ultrasonic sensor and UHF sensor to form a multi-layer feature image.

[0083] S205 , extracting preset useful features from the multi-layer feature image, and using the mapping relationship between the preset useful features and the operating status of the GIS to predict whether the GIS has a fault and the corresponding fault type.

[0084] In practice, an integrated monitoring system is designed to aggregate data collected by different types of sensors into a central processing unit (CPU). Within the CPU, the data is analyzed and processed to identify and locate faults. The CPU employs data fusion techniques, such as machine learning algorithms, to improve the accuracy and efficiency of fault detection. For example, a convolutional neural network is used to extract pre-defined useful features from multi-layer feature images. The extracted features are then used to train a machine learning model to distinguish between normal states and different fault conditions. The machine learning model characterizes the mapping between pre-defined useful features and the operating status of the GIS.

[0085] The GIS fault prediction results, actual operating status information and corresponding sensor related information are added to the dataset as training samples to update the mapping relationship between the preset useful features and the operating status of the GIS. This adapts to new fault modes and equipment changes and improves the accuracy of GIS fault diagnosis. Specifically, the training process of the machine learning model, such as Figure 7 As shown in Figure 2, the trained model is used to analyze new multi-layer images to achieve real-time or near real-time fault diagnosis.

[0086] In this embodiment, useful features can be predetermined based on the correlation between the information sensed by each sensor and the GIS fault information.

[0087] In particular, according to an embodiment of the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present application includes a computer program product, which includes a computer program carried on a computer readable medium, the computer program including a computer program for executing Figure 2 In such an embodiment, the computer program can be downloaded and installed from a network via the communication section 109 and / or installed from the removable medium 111. When the computer program is executed by the central processing unit 101, various functions defined in the apparatus of the present application are performed.

[0088] in, Figure 2 The computer program instructions corresponding to the method shown can also be stored in a computer readable memory that can guide a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0089] Those skilled in the art will appreciate that all or part of the processes in the above-described method embodiments can be implemented by instructing related hardware through a computer program. The program can be stored in a computer-readable storage medium, and when executed, the program can include the processes in the above-described method embodiments. The storage medium can be a magnetic disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM).

[0090] Figure 6 This is a schematic diagram of the structure of a GIS fault diagnosis device based on multi-layer feature imaging in an embodiment of the present invention. Figure 2 The GIS fault diagnosis method based on multi-layer feature imaging corresponds to Figure 6 As shown, the GIS fault diagnosis device based on multi-layer feature imaging in this embodiment may include:

[0091] The sensor-related information acquisition module 301 is used to obtain the spatial distribution information, the characteristic dimensions expressed, and the perceived GIS operation status information of the pre-deployed multiple types of sensors;

[0092] A sensor feature map construction module 302 is used to construct a feature map of each sensor based on the GIS operation status information sensed by each sensor;

[0093] The single feature image generation module 303 is used to convert the feature maps of each sensor of the same feature dimension into corresponding grayscale values, map the corresponding grayscale values ​​to the corresponding pixel points in the GIS chamber image template according to the spatial distribution information of the sensor, and generate a complete grayscale image of all single feature dimensions;

[0094] A multi-layer feature image generation module 304 is used to randomly superimpose all complete grayscale images of a single feature dimension to form a multi-layer feature image;

[0095] The GIS operation status prediction module 305 is used to extract preset useful features from the multi-layer feature image, and use the mapping relationship between the preset useful features and the operation status of the GIS to predict whether the GIS has a fault and the corresponding fault type.

[0096] Figure 6 The specific implementation process of modules 301 to 305 in the GIS fault diagnosis device based on multi-layer feature imaging is similar to that of FIG. Figure 2 The specific implementation process of steps S201 to S205 is the same as that of FIG.

[0097] The foregoing description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Those skilled in the art will readily appreciate that various modifications and variations of the present invention are possible. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present invention are intended to be within the scope of protection of the present invention.

Claims

1. A GIS fault diagnosis method based on multi-layer feature imaging, characterized in that: include: Obtain spatial distribution information, expressed feature dimensions, and perceived GIS operation status information of pre-deployed multi-type sensors; Based on the GIS operation status information perceived by each sensor, a feature map of each sensor is constructed; The feature maps of each sensor of the same feature dimension are converted into corresponding grayscale values. According to the spatial distribution information of the sensor, the corresponding grayscale values ​​are mapped to the corresponding pixel points in the GIS chamber image template to generate a complete grayscale image of all single feature dimensions. Randomly superimpose all complete grayscale images of a single feature dimension to form a multi-layer feature image; Preset useful features are extracted from the multi-layer feature images, and the mapping relationship between the preset useful features and the operating status of the GIS is used to predict whether the GIS has a fault and the corresponding fault type.

2. The GIS fault diagnosis method based on multi-layer feature imaging according to claim 1, characterized in that: By simulating the electromagnetic field, acoustic field and thermal field distribution inside the GIS, the sensor layout points are optimized with the purpose of maximizing the capture of fault signals, thereby obtaining the spatial distribution information of the sensors.

3. The GIS fault diagnosis method based on multi-layer feature imaging according to claim 1, characterized in that: Each type of sensor is evenly distributed in the GIS chamber.

4. The GIS fault diagnosis method based on multi-layer feature imaging according to claim 1, characterized in that: There is a one-to-one correspondence between the sensor type and the feature dimension.

5. The GIS fault diagnosis method based on multi-layer feature imaging according to claim 1 is characterized in that: The pixel size of the GIS chamber image template is determined by the aspect ratio of the GIS chamber.

6. The GIS fault diagnosis method based on multi-layer feature imaging according to claim 1, characterized in that: Before constructing the feature maps of each sensor, it also includes: The GIS operation status information perceived by each sensor is time aligned, data cleaned, missing information supplemented and normalized.

7. The GIS fault diagnosis method based on multi-layer feature imaging according to claim 1, characterized in that: The fault prediction results of GIS, the actual operating status information and the corresponding sensor-related information are added to the dataset as training samples to update the mapping relationship between the preset useful features and the operating status of GIS.

8. A GIS fault diagnosis method based on multi-layer feature imaging, characterized in that: include: The sensor-related information acquisition module is used to obtain the spatial distribution information, characteristic dimensions of the expression, and perceived GIS operation status information of pre-deployed multi-type sensors; The sensor feature map construction module is used to construct the feature map of each sensor based on the GIS operation status information perceived by each sensor; The single feature image generation module is used to convert the feature maps of each sensor of the same feature dimension into corresponding grayscale values, map the corresponding grayscale values ​​to the corresponding pixel points in the GIS chamber image template according to the spatial distribution information of the sensor, and generate a complete grayscale image of all single feature dimensions; A multi-layer feature image generation module, which is used to randomly superimpose all complete grayscale images of a single feature dimension to form a multi-layer feature image; The GIS operation status prediction module is used to extract preset useful features from multi-layer feature images, and use the mapping relationship between the preset useful features and the operation status of the GIS to predict whether the GIS has a fault and the corresponding fault type.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps of the GIS fault diagnosis method based on multi-layer feature imaging as described in any one of claims 1 to 7 are implemented.

10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the steps of the GIS fault diagnosis method based on multi-layer feature imaging as described in any one of claims 1 to 7 are implemented.

Images