Power transformation main equipment fault diagnosis and repair system, method and equipment and storage medium

Abstract

The invention discloses a power transformation main equipment fault diagnosis and repair system, method and equipment and a storage medium. The system comprises a data acquisition module, a data preprocessing module, a data fusion module, a fault diagnosis module, a fault analysis module and a fault repair module. The data acquisition module is used for acquiring multi-source operation state data of the power transformation main equipment; the data preprocessing module is used for preprocessing the multi-source operation state data to obtain target multi-source operation state data; the data fusion module is used for performing data fusion on the target multi-source operation state data to obtain fused data; the fault diagnosis module is used for determining basic fault information of the power transformation main equipment according to the fused data; the fault analysis module is used for determining deep fault information of the power transformation main equipment according to the basic fault information; and the fault repairing module is used for determining a fault repairing scheme according to the basic fault information and the deep fault information. According to the invention, the accuracy of fault diagnosis and repair is improved.

Description

Technical Field

[0001] This application relates to the technical field of power system fault maintenance, and particularly to a fault diagnosis and repair system, method, device and storage medium for main substation equipment. Background Art

[0002] With the continuous expansion of the scale of the power system, the stable operation of the power system faces increasing challenges. Among them, as an important node in power transmission, the stable operation of the main substation equipment is crucial to the safety of the power grid. Therefore, fault diagnosis and repair of the main substation equipment are important issues in the existing power system.

[0003] In the existing fault diagnosis and repair methods for main substation equipment, the operation status data of the main substation equipment is mainly collected by setting sensors. For example, an infrared thermal imager is used to monitor temperature anomalies, and ultrasonic detection is used to detect internal cracks. Then, the fault type and repair plan of the main substation equipment are determined by analyzing the operation status data.

[0004] However, in the prior art, the data collected by each sensor can only be analyzed at a time. Since the data collected by the sensor is affected by the environment and the equipment structure, the collected data has poor adaptability to complex working conditions. In the case of multiple faults affecting each other, it is easy to miss or misdetect, resulting in low accuracy of fault diagnosis and repair. Summary of the Invention

[0005] In order to solve the above problems existing in the prior art, the embodiments of this application provide a fault diagnosis and repair system, method, device and storage medium for main substation equipment. After preprocessing the multi-source operation status data of the main substation equipment, data fusion is performed, and fault analysis is performed on the main substation equipment according to the fused data, so as to analyze all the basic fault information of the main substation equipment under the multi-source operation status data, avoid missing or misdetecting, improve the accuracy of fault diagnosis, and furthermore, reasoning can be performed based on the basic fault information to obtain the deep fault information of the main substation equipment, so as to determine the fault repair plan according to the basic fault information and the deep fault information, improving the accuracy of fault repair.

[0006] In a first aspect, the embodiments of this application provide a fault diagnosis and repair system for main substation equipment, the system includes: a data acquisition module, a data preprocessing module, a data fusion module, a fault diagnosis module, a fault analysis module and a fault repair module;

[0007] The data acquisition module is used to acquire multi-source operation status data of the main substation equipment;

[0008] The data preprocessing module is used to preprocess the multi-source operation status data to obtain target multi-source operation status data;

[0009] The data fusion module is used to perform data fusion on the target multi-source operation status data to obtain fused data;

[0010] The fault diagnosis module is used to determine the basic fault information of the main substation equipment according to the fused data;

[0011] The fault analysis module is used to determine the deep fault information of the main substation equipment according to the basic fault information;

[0012] The fault repair module is used to determine a fault repair plan according to the basic fault information and the deep fault information.

[0013] In a second aspect, an embodiment of the present application provides a method for diagnosing and repairing faults of main substation equipment, including:

[0014] Collect multi-source operation status data of the main substation equipment;

[0015] Preprocess the multi-source operation status data to obtain target multi-source operation status data;

[0016] Perform data fusion on the target multi-source operation status data to obtain fused data;

[0017] Determine the basic fault information of the main substation equipment according to the fused data;

[0018] Determine the deep fault information of the main substation equipment according to the basic fault information;

[0019] Determine a fault repair plan according to the basic fault information and the deep fault information.

[0020] In a third aspect, an embodiment of the present application provides an electronic device, including: a processor and a memory, the processor is connected to the memory, the memory is used to store a computer program, and the processor is used to execute the computer program stored in the memory so that the electronic device executes the method described in the second aspect.

[0021] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, the computer-readable storage medium stores a computer program, and the computer program is executed by a processor to implement the method described in the second aspect.

[0022] In a fifth aspect, an embodiment of the present application provides a computer program product, the computer program product includes a non-transitory computer-readable storage medium storing a computer program, and the computer program is executed by a processor to implement the method described in the second aspect.

[0023] Implementing the embodiments of the present application has the following beneficial effects:

[0024] The power transformation main equipment fault diagnosis and repair system described in this application includes: a data acquisition module, a data preprocessing module, a data fusion module, a fault diagnosis module, a fault analysis module, and a fault repair module. Among them: The data acquisition module is used to collect multi-source operation status data of the power transformation main equipment; the data preprocessing module is used to preprocess the multi-source operation status data to obtain target multi-source operation status data; the data fusion module is used to perform data fusion on the target multi-source operation status data to obtain fusion data; the fault diagnosis module is used to determine the basic fault information of the power transformation main equipment according to the fusion data; the fault analysis module is used to determine the deep fault information of the power transformation main equipment according to the basic fault information; the fault repair module is used to determine a fault repair plan according to the basic fault information and the deep fault information. In this way, through data fusion after preprocessing the multi-source operation status data of the power transformation main equipment, the fused data can reduce the influence of the environment and the complex internal structure of the power transformation main equipment, thereby improving the accuracy of fault diagnosis. According to the fusion data, fault analysis is performed on the power transformation main equipment, and all basic fault information of the power transformation main equipment under multi-source operation status data can be analyzed, avoiding missed detection or misdetection, further improving the accuracy of fault diagnosis. Moreover, based on the basic fault information, reasoning can be carried out to obtain the deep fault information of the power transformation main equipment, so that according to the basic fault information and the deep fault information, a fault repair plan is determined, improving the accuracy of fault repair. Brief Description of the Drawings

[0025] In order to more clearly illustrate the technical solutions in the embodiments of this application, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the drawings in the following description are some embodiments of this application. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained according to these drawings.

[0026] Figure 1 It is an application scenario diagram of a power transformation main equipment fault diagnosis and repair system provided by an embodiment of this application;

[0027] Figure 2 It is a schematic diagram of the functional modules of a power transformation main equipment fault diagnosis and repair system provided by an embodiment of this application;

[0028] Figure 3 It is a schematic diagram of the functional units of a data preprocessing module provided by an embodiment of this application;

[0029] Figure 4 It is a schematic diagram of the functional units of a data fusion module provided by an embodiment of this application;

[0030] Figure 5Schematic diagram of the functional units of a fault diagnosis module provided by an embodiment of the present application;

[0031] Figure 6 Schematic diagram of the functional units of a fault analysis module provided by an embodiment of the present application;

[0032] Figure 7 Schematic diagram of the functional units of a fault repair module provided by an embodiment of the present application;

[0033] Figure 8 Schematic flowchart of a fault diagnosis and repair method for main substation equipment provided by an embodiment of the present application;

[0034] Figure 9 Schematic diagram of the structure of an electronic device provided by an embodiment of the present application. Detailed implementation manners

[0035] In order to enable those skilled in the art to better understand the solution of the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present application.

[0036] The terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but optionally further includes steps or units not listed, or optionally further includes other steps or units inherent to these processes, methods, products or devices.

[0037] It should be understood that the term " / and" in this article is only a description of the association relationship between associated objects, indicating that three relationships may exist. For example, A and / or B may represent: A exists alone, A and B exist simultaneously, and B exists alone. In addition, the character " / " in this article indicates that the associated objects before and after are in an "or" relationship. The term "multiple" as used in the embodiments of the present application refers to two or more.

[0038] The "at least one (piece)" or its similar expression in the embodiments of the present application refers to any combination of these items, including any combination of a single item (piece) or plural items (pieces), meaning one or more, and "more than one" means two or more than two. For example, at least one (piece) of a, b, or c can represent the following seven cases: a, b, c, a and b, a and c, b and c, a, b, and c. Each of a, b, and c can be an element or a set containing one or more elements.

[0039] The "connection" mentioned in the embodiments of the present application refers to various connection methods such as direct connection or indirect connection to achieve communication between devices, and the embodiments of the present application do not make any limitation thereto.

[0040] Referring to "embodiment" in this context means that the specific features, structures, or characteristics described in connection with the embodiment can be included in at least one embodiment of the present application. The occurrence of this phrase at various positions in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment mutually exclusive with other embodiments. Those skilled in the art explicitly and implicitly understand that the embodiments described herein can be combined with other embodiments.

[0041] The relevant content, concepts, meanings, technical problems, technical solutions, beneficial effects, etc. involved in the embodiments of the present application are described below.

[0042] First, some professional terms involved in the present application are explained:

[0043] Main substation equipment: Equipment that plays a key role in a power system substation and undertakes core functions such as transformation, transmission, and distribution of electric energy. For example, equipment such as transformers, switchgear, instrument transformers, reactors, and capacitors. The stable operation of the main substation equipment is crucial for the stable operation of the power system and grid security.

[0044] Wavelet decomposition: An algorithm that decomposes a signal or an image into different frequency components, revealing the local characteristics of the signal through multi-resolution analysis (MRA). Different from the global frequency analysis corresponding to the Fourier transform, wavelet decomposition can capture both the time-domain and frequency-domain information of the signal and is commonly used to process non-stationary signals. Wavelet decomposition can be applied to the denoising process. Among them, the wavelet coefficients after wavelet decomposition can be used to restore the data through inverse wavelet transform after threshold processing.

[0045] Scale-Invariant Feature Transform (SIFT) algorithm: A feature extraction algorithm in the field of computer vision, used to detect local features in images to extract feature points corresponding to the local features. The core features of the SIFT algorithm are scale invariance and rotation invariance, and it is widely used in technical fields such as image matching, object recognition, and 3D reconstruction.

[0046] Contrast enhancement algorithm: A key algorithm in image processing for enhancing the contrast of images. By adjusting the distribution of pixel values, it makes the features in the image clearer and the colors more vivid. Common contrast enhancement algorithms include: histogram equalization, Gamma Correction algorithm, etc.

[0047] Image edge sharpening algorithm: An algorithm in image processing for enhancing the edge features of images. By highlighting the high-frequency information in the image, such as contours, textures, etc., it makes the image features more obvious. Common image edge sharpening algorithms include: unsharp masking algorithm, Sobel operator sharpening algorithm, Prewitt operator sharpening algorithm, etc.

[0048] Convolutional Neural Network (CNN) model: A neural network architecture in the field of deep learning specifically used to process grid-structured data, such as images, videos, and audio. It can process the spatial or temporal features of data through local perception, weight sharing, and hierarchical feature extraction, and predict the changes in data features.

[0049] Random forest model: A machine learning algorithm based on parallel ensemble learning. By constructing multiple decision trees and combining the prediction results of each decision tree, it significantly improves the accuracy and robustness of the model. The random forest model is widely used in tasks such as classification, regression, and feature selection, and is especially suitable for high-dimensional data and non-linear problems.

[0050] Knowledge graph: A structured semantic network used to represent and manage knowledge relationships, consisting of nodes and paths. Nodes represent entities, and paths represent the relationships between entities. For example, in the knowledge graph of this application, nodes represent components in the main substation equipment, and paths represent the influence relationships, fault types, and fault triggering conditions of the operating state data between two components.

[0051] Please refer to Figure 1 , Figure 1An application scenario diagram of a substation main equipment fault diagnosis and repair system provided in an embodiment of the present application. Among them, the substation main equipment fault diagnosis and repair system can be connected to the substation main equipment in the power system for communication. The substation main equipment, as an important node participating in power conversion, transmission and distribution in the power system, plays a vital role in the stable operation of the power system. Optionally, the substation main equipment fault diagnosis and repair system can communicate with the substation main equipment through a physical interface or a communication bus. The substation main equipment fault diagnosis and repair system can also be connected to the substation main equipment for remote communication through remote communication technology, such as local area network, long-distance transmission and other technologies. The substation main equipment fault diagnosis and repair system is provided with a data acquisition module, such as a variety of sensors, which can collect the operating status data of the substation main equipment through the sensors. The substation main equipment fault diagnosis and repair system can analyze the operating status data of the substation main equipment to determine the fault information of the substation main equipment, and analyze and determine the fault repair plan for the fault information.

[0052] It should be noted that in some existing fault diagnosis and repair methods, each sensor collects a piece of operating status data from the main substation equipment. Each piece of operating status data is then analyzed separately to determine a single fault message for the main substation equipment. However, because the data collected by the sensors is easily affected by the environment and the complex structure of the main substation equipment, the quality of the collected data is poor. Analysis based on this data results in low accuracy in fault diagnosis and repair. Furthermore, only one piece of fault information can be determined at a time. When faced with the impact of multiple faults, missed detections or false detections are likely to occur, resulting in low accuracy in fault diagnosis and repair.

[0053] To this end, in a substation main equipment fault diagnosis and repair system provided in the present application, the system includes: a data acquisition module, a data preprocessing module, a data fusion module, a fault diagnosis module, a fault analysis module and a fault repair module; the data acquisition module is used to collect multi-source operating status data of the substation main equipment; the data preprocessing module is used to preprocess the multi-source operating status data to obtain target multi-source operating status data; the data fusion module is used to fuse the target multi-source operating status data to obtain fused data; the fault diagnosis module is used to determine the basic fault information of the substation main equipment based on the fused data; the fault analysis module is used to determine the deep fault information of the substation main equipment based on the basic fault information; the fault repair module is used to determine the fault repair plan based on the basic fault information and the deep fault information.

[0054] It can be seen that in the system provided by this application, after preprocessing the multi-source operation status data of the main substation equipment and then performing data fusion, the fused data can reduce the influence of the environment and the complex internal structure of the main substation equipment, thereby improving the accuracy of fault diagnosis. Based on the fused data, fault analysis is carried out on the main substation equipment, and all basic fault information of the main substation equipment under multi-source operation status data can be analyzed, avoiding missed detections or false detections, and further improving the accuracy of fault diagnosis. Moreover, reasoning can be carried out based on the basic fault information to obtain the deep fault information of the main substation equipment, so that based on the basic fault information and the deep fault information, a fault repair plan can be determined, improving the accuracy of fault repair.

[0055] The following will specifically introduce the fault diagnosis and repair system for the main substation equipment provided by this application.

[0056] Refer to Figure 2 , Figure 2 , which is a schematic diagram of the functional modules of a fault diagnosis and repair system for a main substation equipment provided by an embodiment of this application. The fault diagnosis and repair system for the main substation equipment includes: a data acquisition module, a data preprocessing module, a data fusion module, a fault diagnosis module, a fault analysis module, and a fault repair module. Among them, the data acquisition module, the data preprocessing module, the data fusion module, the fault diagnosis module, the fault analysis module, and the fault repair module are connected through a data bus or remotely communicated.

[0057] The data acquisition module is used to acquire the multi-source operation status data of the main substation equipment.

[0058] In the embodiment of this application, the data acquisition module of the fault diagnosis and repair system for the main substation equipment can be communicatively connected to the main substation equipment and is used to acquire the multi-source operation status data of the main substation equipment.

[0059] Exemplarily, the data acquisition module can be various types of sensors deployed at key nodes of the main substation equipment. For example, the data acquisition module can include: high-definition cameras, infrared thermal imagers, ultrasonic sensors, vibration sensors, current sensors, etc. The multi-source operation status data can include: visible light images, infrared thermal imaging images, ultrasonic images, mechanical vibration data, electrical data, etc. The mechanical vibration data can include mechanical vibration frequency and mechanical vibration amplitude. The electrical data can include the current of the main substation equipment. Among them, the visible light image is collected by a high-definition camera, the infrared thermal imaging image is collected by an infrared thermal imager, the ultrasonic image is collected by an ultrasonic sensor, the mechanical vibration is collected by a vibration sensor, and the electrical data is collected by electrical data sensors such as current sensors. The visible light image can be used to analyze the fault information on the surface of the main substation equipment, such as appearance defects like cracks, deformations, and rust. The infrared thermal imaging image can be used to analyze the temperature distribution on the surface of the main substation equipment to determine whether there are fault information such as local overheating in high-load areas. The ultrasonic image can be used to analyze the fault information corresponding to the small structural changes inside the main substation equipment, such as cavities or cracks. The mechanical vibration data can be used to analyze the abnormal changes in the mechanical vibration of the main substation equipment to determine whether there is mechanical looseness or fatigue in the equipment. The electrical data can be used to analyze the fault information such as current fluctuations and load abnormalities of the main substation equipment. For example, in the fault diagnosis of a 500 kV transformer, a high-definition camera is installed above the shell to capture surface cracks. An infrared thermal imager is placed on the radiator to monitor the temperature distribution. An ultrasonic sensor is attached to the inside of the transformer to detect internal cracks. A vibration sensor is fixed to the base to record mechanical vibrations. An electrical sensor is connected to the transformer terminal to monitor load fluctuations.

[0060] It should be noted that any operation status data collected by the data acquisition module is accompanied by a timestamp, and the timestamp is used to indicate the acquisition moment corresponding to the operation status data. Optionally, a high-precision time synchronization protocol, such as the IEEE 1588 Precision Time Protocol (PTP) protocol, can be used to stamp each piece of data to ensure that the data collected by different types of sensors is aligned at the same acquisition moment, avoiding analysis errors caused by moment misalignment. By analyzing the multi-source operation status data under the same timestamp, multiple different-dimensional fault information of the main substation equipment at the acquisition moment corresponding to the timestamp can be analyzed, realizing the coverage of multi-dimensional data diagnosis of the surface defects, internal structure, and electrical performance of the main substation equipment, ensuring that reliable operation status data can be collected under different fault types and environments, and improving the comprehensiveness and accuracy of fault diagnosis and repair.

[0061] The data preprocessing module is used to preprocess the multi-source operation status data to obtain the target multi-source operation status data.

[0062] In an embodiment of the present application, the preprocessing may include: denoising processing, data alignment, missing data filling, image enhancement, etc. The data preprocessing module receives multi-source operating state data collected by the data acquisition module, and performs denoising processing, data alignment, missing data filling, image enhancement, etc. on the multi-source operating state data in sequence, so as to obtain target multi-source operating state data and ensure the accuracy of the data.

[0063] Exemplarily, the multi-source operating state data may include multiple device images and multiple device state data of main substation equipment. For example, the multiple device images may include: visible light images, infrared thermal imaging images, ultrasonic images, etc. The multiple device state data may include: mechanical vibration data, electrical data, etc. The target multi-source operating state data includes multiple target device images and multiple target device state data.

[0064] Refer to Figure 3 , in terms of preprocessing the multi-source operating state data to obtain the target multi-source operating state data, the data preprocessing module may include: a denoising processing unit, a data alignment unit, a missing data filling unit, and an image enhancement unit, where:

[0065] The denoising processing unit is configured to perform wavelet decomposition on the multi-source operating state data according to a preset wavelet basis function and a preset decomposition level to obtain low-frequency wavelet coefficients and high-frequency wavelet coefficients; perform threshold processing on the high-frequency wavelet coefficients according to a preset threshold type to obtain target high-frequency wavelet coefficients; perform inverse wavelet transform according to the low-frequency wavelet coefficients and the target high-frequency wavelet coefficients to obtain denoised data; where the denoised data includes multiple denoised device images and multiple denoised device state data.

[0066] The data alignment unit is configured to extract multiple feature point sets corresponding to the multiple denoised device images through the SIFT algorithm, and each denoised device image corresponds to one feature point set; align the multiple denoised device images spatially according to the multiple feature point sets to obtain multiple aligned device images;

[0067] The missing data filling unit is configured to obtain the missing part data of each denoised device state data in the multiple denoised device state data, and fill the missing part data with the nearest preset number of non-missing part data of the missing part data to obtain multiple target device state data;

[0068] The image enhancement unit is configured to use a contrast enhancement algorithm to enhance the contrast of the multiple aligned device images to obtain multiple enhanced images; perform edge sharpening on the multiple enhanced images using an image edge sharpening algorithm to obtain multiple target device images.

[0069] In a specific implementation, the denoising processing unit first obtains a preset wavelet basis function and a preset decomposition level. Optionally, the preset wavelet basis function can be a Haar wavelet basis function, a Daubechies wavelet basis function, a Symlets wavelet basis function, etc. The decomposition level can be preset based on the signal type of the multi-source operating state data, usually three to five layers. The more the decomposition level, the more accurate the denoising processing of the signal. The denoising processing unit performs wavelet decomposition on each piece of data in the multi-source operating state data according to the preset wavelet basis function and the preset decomposition level, obtaining low-frequency wavelet coefficients and high-frequency wavelet coefficients.

[0070] It should be noted that the low-frequency wavelet coefficients include the low-frequency wavelet coefficients corresponding to each piece of data in the multi-source operating state data, which are used to represent the low-frequency information corresponding to each piece of data. The high-frequency wavelet coefficients include the high-frequency wavelet coefficients corresponding to each piece of data in the multi-source operating state data, which are used to represent the high-frequency information corresponding to each piece of data. Among them, the noise of the signal exists in the high-frequency components of the signal. By performing threshold processing on the high-frequency wavelet coefficients, the high-frequency noise in the multi-source operating state data can be filtered out, such as wind noise, environmental vibration, random interference signals, and so on. Optionally, the noise with a frequency higher than 50 Hz can be removed.

[0071] Therefore, the denoising processing unit obtains a preset threshold type and, according to the preset threshold type, performs threshold processing on the high-frequency wavelet coefficients to obtain target high-frequency wavelet coefficients. Among them, the preset threshold type can include: hard threshold, soft threshold, etc. The threshold used can include any one of the following: VisuShrink threshold, SureShrink threshold, etc. The denoising processing unit can perform threshold processing on the high-frequency wavelet coefficients through the threshold corresponding to the preset threshold type to obtain target high-frequency wavelet coefficients. Finally, the denoising processing unit can perform wavelet inverse transform according to the low-frequency wavelet coefficients and the target high-frequency wavelet coefficients to reconstruct the signal of the multi-source operating state data and obtain the denoised processing data.

[0072] Then, the data alignment unit can extract the feature points of each denoised device image in multiple denoised device images through the SIFT algorithm, obtaining a feature point set corresponding to each denoised device image. Feature point matching is performed according to the feature point set corresponding to each denoised device image to obtain a matching result. According to the matching result, each pixel position in the multiple denoised device images is spatially aligned to obtain multiple aligned device images. Optionally, multiple denoised device images can be aligned to a pre-established spatial coordinate system to align the multiple device images spatially.

[0073] Furthermore, the missing data filling unit may obtain the missing partial data of each denoising device status data according to the time stamp corresponding to each denoising device status data in multiple denoising device status data. By obtaining a preset number of non-missing partial data corresponding to the time stamp closest to the missing partial data of each denoising device status data to fill the missing partial data, each target device status data can be obtained. In this way, multiple target device status data can be obtained in sequence. Optionally, the missing partial data can be filled with the average or median of a preset number of non-missing partial data corresponding to the time stamp closest to the missing partial data of each denoising device status data, so that the filled data is closer to the actually collected data, avoiding analysis errors caused by inaccurate data filling and improving the accuracy of fault diagnosis and repair.

[0074] Optionally, a quadratic polynomial fitting calculation can be performed on a preset number of non-missing partial data corresponding to the time stamp closest to the missing partial data of each denoising device status data to obtain an intermediate value, and a smooth transition value of the missing partial data can be obtained. The missing partial data can be filled with the smooth transition value. In this way, the continuity and trend rationality of the ultrasonic signal can be ensured, and the gradual change characteristics of the internal structure of the device can be better reflected compared with simple linear interpolation.

[0075] Finally, the image enhancement unit can enhance the contrast of multiple aligned device images through a contrast enhancement algorithm. For example, the contrast of multiple aligned device images can be enhanced through a histogram equalization algorithm to obtain multiple enhanced images. And an image edge sharpening algorithm is used to perform edge sharpening on the multiple enhanced images to obtain multiple target device images. Optionally, multiple enhanced images can be edge-sharpened through a Gaussian sharpening filter to obtain multiple target device images. In this way, through image enhancement processing such as contrast enhancement and image edge sharpening, the image features can be made more obvious, making it easier to extract image features.

[0076] It can be seen that by denoising the multi-source operation status data through the denoising processing unit, the interference of high-frequency environmental noise and random noise in the multi-source operation status data can be removed, thereby improving the accuracy of the multi-source operation status data and making the fault diagnosis and repair more accurate. By aligning the data of multiple denoised device images, the multiple denoised device images can be spatially aligned, facilitating comprehensive fault diagnosis based on the multiple denoised device images and improving the comprehensiveness of fault diagnosis and repair. By filling in the missing data in the status data of multiple denoised devices, the integrity of data analysis can be ensured to improve the accuracy of fault diagnosis. By enhancing the contrast and sharpening the edges of the multiple aligned device images after data alignment, the image features can be made more obvious and easier to extract. In this way, the preprocessed data can be made more complete and with obvious features, so that fault diagnosis and analysis can be carried out based on this data, improving the accuracy of fault diagnosis.

[0077] The data fusion module is used to perform data fusion on the target multi-source operation status data to obtain fused data.

[0078] In the embodiment of the present application, the target multi-source operation status data includes: preprocessed visible light images, preprocessed infrared thermal imaging images, preprocessed ultrasonic images, preprocessed mechanical vibration data, preprocessed electrical data, etc. The data fusion module can perform data fusion on the target multi-source operation status data so that the target multi-source operation status data can be represented by the same data type, thereby analyzing the fault information of the main substation equipment in multiple dimensions based on the target multi-source operation status data and improving the comprehensiveness of fault diagnosis.

[0079] Exemplarily, each device status data corresponds to a spatial position. The spatial position corresponding to each device status data is the acquisition position of the device status data.

[0080] Refer to Figure 4 , in terms of performing data fusion on the target multi-source operation status data to obtain fused data, the data fusion module may include: a weight assignment unit, a data fusion unit, and a reliability evaluation unit, where:

[0081] The weight assignment unit is used to determine the environmental parameters of the main substation equipment; according to the environmental parameters, determine multiple image weights corresponding to multiple target device images, and each target device image corresponds to an image weight;

[0082] A data fusion unit is configured to determine display parameters corresponding to each target device image among multiple target device images according to multiple image weights, so as to obtain multiple display parameters; display the display content corresponding to each target device image among the multiple target device images in a multi-dimensional space coordinate system according to the multiple display parameters, so as to obtain a multi-dimensional fusion model corresponding to the main substation equipment; fuse the target device status data and the multi-dimensional fusion model according to the spatial position corresponding to the target device status data, so as to obtain initial fusion data;

[0083] A reliability evaluation unit is configured to perform reliability evaluation on each data in the target multi-source operation status data through a Bayesian inference model to obtain an evaluation result; determine fusion data according to the evaluation result and the initial fusion data.

[0084] It should be noted that the data fusion module can fuse the target multi-source operation status data into a multi-dimensional fusion model corresponding to the main substation equipment under the same space coordinate system, so as to intuitively display the target multi-source operation status data through the multi-dimensional fusion model, which is convenient for feature extraction.

[0085] In specific implementation, the weight allocation unit first obtains the environmental parameters of the main substation equipment. The environmental parameters may include: light intensity, temperature, and load fluctuation amount. According to the environmental parameters, image weights are set for each target device image among the multiple target device images to obtain multiple image weights. Taking the environmental parameter as the light intensity as an example for illustration, the weight allocation unit first determines the light intensity interval where the light intensity is located. Each light intensity interval corresponds to an image weight of one target device image, and the corresponding relationship between the light intensity interval and the image weight of each target device image can be preset in advance. In this way, multiple image weights corresponding to multiple target device images can be determined.

[0086] It can be understood that adjusting the image weights according to different environmental parameters can enable the target device image that better matches the environment to retain more image features, so that the extracted image features are more accurate, and the result of fault diagnosis based on the image features is more accurate. For example, under good daylight conditions, the system sets a higher weight for the data of the high-definition camera and combines it with the infrared data to capture temperature anomalies.

[0087] Then, according to multiple image weights, the data fusion unit determines the display parameters corresponding to each target device image among the multiple target device images, that is, according to the multiple image weights, retains the image features corresponding to each target device image among the multiple target device images. The greater the image weight, the more image features are retained, thereby obtaining multiple display parameters. The data fusion unit will construct a multi-dimensional fusion model in the multi-dimensional space coordinate system according to the multiple display parameters. Among them, the multi-dimensional fusion model displays the display content of each target device image among the multiple target device images corresponding to the main power transformation equipment. For example, in the multi-dimensional fusion model, there are displayed multi-dimensional images of the surface of the main power transformation equipment corresponding to the visible light image, the temperature distribution of the surface of the main power transformation equipment corresponding to the infrared thermal imaging image, and multi-dimensional images of the internal structure of the main power transformation equipment corresponding to the ultrasonic image. Then, according to the spatial position corresponding to the target device state data, the target device state data is aligned with each spatial position on the multi-dimensional fusion model to obtain initial fusion data, thereby obtaining a fusion data that includes the surface features, temperature features, internal structure features, mechanical vibration features, and electrical data features of the main power transformation equipment.

[0088] Furthermore, the reliability evaluation unit performs a reliability evaluation on each data in the target multi-source operation state data through a Bayesian inference model to obtain an evaluation result, and the evaluation result includes the reliability of each data in the target multi-source operation state data. In a specific implementation, the reliability evaluation unit obtains preset reliability parameters. For example, the preset reliability parameters can include any one of the following parameters: constant failure rate, time-varying failure rate, etc. Obtain the likelihood function corresponding to the preset reliability parameter. For example, the likelihood function corresponding to the constant failure rate is an exponential distribution function. Then, according to each data in the target multi-source operation state data, determine the prior distribution corresponding to each data to obtain multiple prior distributions. According to the likelihood function and the multiple prior distributions, determine multiple posterior distributions. According to the reliability function corresponding to the likelihood function and the multiple posterior distributions, determine multiple reliabilities, that is, obtain the reliability of each data in the target multi-source operation state data. The reliability evaluation unit can adjust the initial fusion data according to the reliability of each data in the target multi-source operation state data to obtain the fusion data. For example, data with a reliability less than the preset reliability can be deleted from the initial fusion data to obtain the fusion data, thereby ensuring the accuracy and reliability of the fusion data.

[0089] It can be seen that by assigning image weights to each target device image according to the environmental parameters of the main substation equipment and constructing a multi-dimensional fusion model corresponding to multiple target device images based on the image weights, the target device images matching the environmental parameters can retain more image features, so that more accurate fault diagnosis can be performed through these image features, improving the accuracy of fault diagnosis. By performing reliability evaluation on the target multi-source operation status data, the accuracy of the target multi-source operation status data can be ensured, so that more accurate diagnosis and repair results can be obtained according to the target multi-source operation status data for fault diagnosis and repair.

[0090] A fault diagnosis module, configured to determine the basic fault information of the main substation equipment according to the fusion data.

[0091] In the embodiment of the present application, the basic fault information may include the fault type and the fault spatial location of the main substation equipment. The fault diagnosis module can obtain the basic fault information of the main substation equipment by extracting key features from the fusion data and processing the extracted features.

[0092] Exemplarily, as Figure 5 shown, in terms of determining the basic fault information of the main substation equipment according to the fusion data, the fault diagnosis module includes a feature extraction unit and a feature analysis unit, where:

[0093] The feature extraction unit is configured to extract key features from the fusion data; perform data augmentation on the key features to obtain potential fault features; the potential fault features include: surface features, temperature distribution features, internal structure features, mechanical vibration features, and electrical data features of the main substation equipment;

[0094] The feature analysis unit is configured to determine a temperature fault diagnosis result according to the temperature distribution features; input the surface features and internal structure features into a preset convolutional neural network model to obtain a structural fault diagnosis result of the main substation equipment; analyze the mechanical vibration features and electrical data features through a preset random forest model to obtain an electrical load fault diagnosis result; determine the basic fault information according to the temperature fault diagnosis result, the structural fault diagnosis result, and the electrical load fault diagnosis result.

[0095] In specific implementation, the feature extraction unit first extracts key features from the fusion data, and the key features are used to represent possible fault features. Then, data augmentation is performed on the key features to obtain potential fault features including surface features, temperature distribution features, internal structure features, mechanical vibration features, and electrical data features of the main substation equipment. Optionally, the data augmentation may include: performing rotation transformation on the surface features and internal structure features to simulate image features from multiple perspectives; adding Gaussian noise to the mechanical vibration features to enhance the robustness to noise; performing time series smoothing on the temperature distribution features and electrical data features to highlight the change trend.

[0096] Further, the feature analysis unit can analyze whether there is a temperature anomaly in the main substation equipment according to the temperature distribution characteristics, and determine the temperature fault diagnosis result. The temperature fault diagnosis result may include the abnormal hot spot temperature on the surface of the main substation equipment and the spatial position corresponding to the abnormal hot spot temperature.

[0097] Exemplarily, each device status data corresponds to an electrical data.

[0098] In terms of determining the temperature fault diagnosis result according to the temperature distribution characteristics, the feature analysis unit is specifically used for:

[0099] Determine multiple temperatures corresponding to multiple spatial positions in the main substation equipment according to the temperature distribution characteristics;

[0100] Obtain the ambient temperature and working duration corresponding to the main substation equipment;

[0101] Determine the difference between each temperature in the multiple temperatures and the ambient temperature to obtain multiple temperature differences;

[0102] Determine the reference temperature difference according to the working duration and the electrical data corresponding to each spatial position in the multiple spatial positions;

[0103] Determine multiple influence degrees corresponding to the multiple temperatures according to the multiple temperature differences and the reference temperature difference;

[0104] Determine at least one temperature fault position from the multiple spatial positions according to the multiple influence degrees;

[0105] Determine the temperature fault diagnosis result according to at least one temperature fault position, the temperature corresponding to each temperature fault position, and the influence degree.

[0106] Wherein, each spatial position corresponds to a temperature. Each temperature corresponds to an influence degree.

[0107] In specific implementation, the feature analysis unit first determines the temperature corresponding to each spatial position in the multiple spatial positions in the main substation equipment according to the extracted temperature distribution characteristics, and obtains multiple temperatures. Then, it obtains the ambient temperature and working duration of the main substation equipment.

[0108] It can be understood that the surface temperature change of the main substation equipment is associated with the ambient temperature and the working duration. When the electrical data of the main substation equipment is within the preset electrical data range, the difference between the surface temperature of the main substation equipment and the ambient temperature changes in accordance with a preset distribution with the change of the working duration. The preset distribution can be preset according to the actual test results. Therefore, the feature analysis unit will determine the difference between each temperature in the multiple temperatures and the ambient temperature to obtain multiple temperature differences.

[0109] Further, the feature analysis unit determines a reference temperature difference according to the working duration and the electrical data corresponding to each of the multiple spatial positions. The reference temperature difference represents the surface temperature of the main substation equipment when the electrical data of the main substation equipment is within a preset electrical data range. The feature analysis unit can determine the reference temperature difference according to the working duration, the ambient temperature, and the preset distribution of the main substation equipment.

[0110] Further, the feature analysis unit determines the difference between each of the multiple temperature differences and the reference temperature difference to obtain multiple difference offsets. And determines multiple influence degrees corresponding to the multiple difference offsets. Each difference offset corresponds to an influence degree. The mapping relationship between the difference offset and the influence degree can be preset. In this way, the influence degree corresponding to the temperature of each of the multiple spatial positions can be obtained. The spatial positions where the corresponding influence degrees are within the preset influence degree range are used as temperature fault positions. Thus, at least one temperature fault position can be determined from the multiple spatial positions according to the multiple influence degrees.

[0111] Finally, at least one temperature fault position, the temperature corresponding to each temperature fault position, and the influence degree are used as the temperature fault diagnosis result.

[0112] It can be seen that through the temperature distribution characteristics, multiple temperatures corresponding to multiple spatial positions in the main substation equipment can be determined. By determining the difference between each of the multiple temperatures and the ambient temperature, and by determining the reference temperature difference according to the working duration and the electrical data corresponding to each of the multiple spatial positions, multiple influence degrees corresponding to the multiple temperatures can be determined according to the multiple temperature differences and the reference temperature difference. Thus, at least one temperature fault position can be determined from the multiple spatial positions according to the multiple influence degrees. In this way, the temperature fault diagnosis result of the main substation equipment can be analyzed based on the temperature distribution characteristics. Combining the temperature fault diagnosis result with other fault diagnosis results can improve the comprehensiveness of fault diagnosis and repair.

[0113] In some feasible embodiments, the feature analysis unit can determine multiple temperatures corresponding to multiple spatial positions in the main substation equipment according to the temperature distribution characteristics, and determine at least one abnormal hot spot temperature and at least one abnormal hot spot temperature position according to the multiple temperatures. Each abnormal hot spot temperature corresponds to an abnormal hot spot temperature position. In a specific implementation, the feature analysis unit takes the temperature greater than the hot spot temperature of the main substation equipment among the multiple temperatures as the abnormal hot spot temperature, and takes the spatial position corresponding to the abnormal hot spot temperature as the abnormal hot spot temperature position, so as to obtain at least one abnormal hot spot temperature and at least one abnormal hot spot temperature position. In this way, the abnormal hot spot temperature and the abnormal hot spot temperature position of the main substation equipment can be determined according to the temperature distribution characteristics, improving the comprehensiveness of fault diagnosis.

[0114] Further, the feature analysis unit inputs the surface features and internal structure features into a preset convolutional neural network model to obtain the structural fault diagnosis result of the main power transformation equipment. The preset convolutional neural network model is a pre-trained convolutional neural network model that can predict the structural fault diagnosis result of the main power transformation equipment based on the surface features and internal structure features of the main power transformation equipment. The structural fault diagnosis result may include: the type of structural fault, the location of the structural fault, the morphological features of the structural fault, and so on. The morphological features of the structural fault are used to represent the features such as the size and shape of the structure where the main power transformation equipment fails. For example, the width, depth, direction, and other features of the surface crack of the main power transformation equipment.

[0115] Optionally, the surface features and internal structure features can be separately input into two pre-trained convolutional neural network models, and the surface structure fault diagnosis result corresponding to the surface features and the internal structure fault diagnosis result corresponding to the internal structure features are respectively determined through the two pre-trained convolutional neural network models. In this way, each feature can be independently analyzed to make full use of each feature for analysis and improve the reliability of the fault diagnosis result.

[0116] Further, the feature analysis unit inputs the mechanical vibration features and electrical data features into a preset random forest model to determine the electrical load fault diagnosis result through the preset random forest model. Among them, the preset random forest model is a pre-trained random forest model that can predict the electrical load fault diagnosis result of the main power transformation equipment based on the mechanical vibration features and electrical data features. Optionally, the preset random forest model may include 100 decision trees. The electrical load fault diagnosis result may include a mechanical fault diagnosis result and an electrical fault diagnosis result. The mechanical fault diagnosis result may include: the abnormal frequency and abnormal frequency location of the mechanical vibration, the abnormal amplitude and abnormal amplitude location of the mechanical vibration, and so on. The electrical fault diagnosis result may include: abnormal electrical fluctuations and abnormal electrical fluctuation locations, overload amounts and overload locations, and so on.

[0117] Optionally, the mechanical vibration features and electrical data features can be separately input into two pre-trained random forest models to respectively determine the mechanical fault diagnosis result corresponding to the mechanical vibration features and the electrical fault diagnosis result corresponding to the electrical data features through the two pre-trained random forest models. In this way, each feature can be independently analyzed to make full use of each feature for analysis and improve the reliability of the fault diagnosis result.

[0118] Optionally, a preset Long Short Term Memory (LSTM) model can be used to analyze the mechanical vibration characteristics and electrical data characteristics to obtain the electrical load fault diagnosis result. The LSTM model can include 2 hidden layers, with 64 neurons in each layer, and use the ReLU activation function. Through the LSTM model, the long-term dependence relationship between features can be used to accurately determine the electrical load fault diagnosis result.

[0119] Finally, the temperature fault diagnosis result, the structure fault diagnosis result, and the electrical load fault diagnosis result are jointly used as the basic fault information.

[0120] It can be seen that by extracting the key features of the fusion data and performing data augmentation on the key features, the surface features, temperature distribution features, internal structure features, mechanical vibration features, and electrical data features of the main substation equipment can be obtained. By analyzing the temperature distribution features, the temperature fault diagnosis result can be determined. By analyzing the surface features and internal structure features, the structure fault diagnosis result of the main substation equipment can be determined. By analyzing the mechanical vibration features and electrical data features, the electrical load fault diagnosis result can be obtained, thereby obtaining the basic fault information. In this way, the surface features, temperature distribution features, internal structure features, mechanical vibration features, and electrical data features of the main substation equipment can be comprehensively analyzed to obtain a comprehensive fault diagnosis result, improving the comprehensiveness of fault diagnosis. Moreover, by performing fusion analysis on multiple features, the influence of the environment and equipment structure on the data collected by a single sensor can be reduced, improving the accuracy of fault diagnosis.

[0121] The fault analysis module is used to determine the deep fault information of the main substation equipment according to the basic fault information.

[0122] In the embodiment of the present application, the basic fault information may further include: the spatial position of the faulty component, the basic fault type. The basic fault type may include: temperature fault, structure fault, mechanical fault, electrical fault, and so on. Optionally, the basic fault information may further include the severity of the fault. Optionally, the basic fault type can be obtained by classifying the potential fault features through a Support Vector Machine (SVM) model or a Deep Neural Networks (DNN) model.

[0123] Exemplarily, as Figure 6 shown, in terms of determining the deep fault information of the main substation equipment according to the basic fault information, the fault analysis module may include: a knowledge graph construction unit, an association analysis unit, and a deep fault analysis unit, where:

[0124] A knowledge graph construction unit is used to obtain the historical multi-source operation status data of the main substation equipment, historical fault information, and the spatial positions of all components of the main substation equipment; construct a knowledge graph based on the historical multi-source operation status data, historical fault information, and the spatial positions of all components of the main substation equipment; wherein, the knowledge graph includes multiple nodes and multiple paths, each node represents the spatial position of a component, and each path represents the influence relationship, fault type, and fault trigger condition of the operation status data between two components.

[0125] A linkage analysis unit is used to determine at least one associated path corresponding to the basic fault type from multiple paths according to the basic fault type and the knowledge graph; determine the fault source node from multiple nodes according to the at least one associated path.

[0126] A deep fault analysis unit is used to determine at least one potential fault node from multiple nodes and at least one potential fault path from multiple paths according to the fault source node and the at least one associated path; use the fault source node, the at least one associated path, the at least one potential fault node, and the at least one potential fault path as deep fault information.

[0127] In a specific implementation, the knowledge graph construction unit first obtains the historical multi-source operation status data of the main substation equipment, historical fault information, and the spatial positions of all components of the main substation equipment. The historical multi-source operation status data, historical fault information, and the spatial positions of all components of the main substation equipment are all stored in the database, and the knowledge graph construction unit can directly obtain them from the database. The knowledge graph construction unit can construct multiple nodes according to the spatial positions of all components of the main substation equipment, and construct paths between any two nodes according to the historical multi-source operation status data and historical fault information, obtaining multiple nodes and multiple paths, and thus constructing a knowledge graph according to the multiple nodes and multiple paths.

[0128] Then, the linkage analysis unit matches the basic fault type with multiple paths in the knowledge graph, and determines at least one associated path corresponding to the basic fault type from multiple paths. Perform reverse lookup according to the at least one associated path to obtain the fault source node. For example, the mechanical fault of the first component may be caused by the structural fault of the second component, and the node corresponding to the second component is the fault source node.

[0129] Furthermore, the deep fault analysis unit can determine at least one potential fault node from multiple nodes and at least one potential fault path from multiple paths according to the fault source node and at least one associated path. For example, abnormal mechanical vibration frequency and cracks in the third component may cause electrical fluctuations in the fourth component. Then, the node corresponding to the fourth component is a potential fault node, and the path between the node of the third component and the node of the fourth component is a potential fault path. Finally, the fault source node, at least one associated path, at least one potential fault node, and at least one potential fault path are jointly used as the deep fault information of the main substation equipment.

[0130] It can be seen that by constructing a knowledge graph through historical multi-source operation status data, historical fault information, and the spatial positions of all components of the main substation equipment, and through the linkage analysis of the knowledge graph and the basic fault types, the linkage relationship between components can be analyzed to obtain at least one associated path. Through the fault source node and the at least one associated path, fault deduction can be carried out to obtain at least one potential fault node and at least one potential fault path, so as to obtain the deep fault information of the main substation equipment. In this way, on the basis of determining the basic fault information of the main substation equipment, the potential faults of the main substation equipment can be analyzed to obtain the deep fault information, which improves the comprehensiveness of fault diagnosis.

[0131] The fault repair module is used to determine a fault repair plan according to the basic fault information and the deep fault information.

[0132] In the embodiment of the present application, the fault repair plan may include: repair steps, repair tools, replaced components, repair suggestions, and so on.

[0133] Exemplarily, as Figure 7 shown, in terms of determining a fault repair plan according to the basic fault information and the deep fault information, the fault repair module includes: a fault cause analysis unit and a repair plan determination unit, where:

[0134] The fault cause analysis unit is used to determine the node fault type of each associated node in the multiple associated nodes corresponding to at least one associated path according to the fault source node, at least one associated path, and the basic fault type, to obtain multiple node fault types; and to determine the fault cause according to the multiple node fault types;

[0135] A repair plan determination unit is configured to obtain multiple associated operation status data corresponding to multiple associated nodes, and obtain at least one potential fault operation status data corresponding to at least one potential fault node, where each associated node corresponds to one associated operation status data, and each potential fault node corresponds to one potential fault operation status data; determine at least one candidate fault repair plan from multiple preset fault repair plans corresponding to the fault cause according to the multiple associated operation status data; determine at least one potential fault type corresponding to at least one potential fault node according to the at least one potential fault operation status data and at least one associated path; and determine a fault repair plan from the at least one candidate fault repair plan according to the at least one potential fault type.

[0136] In a specific implementation, the fault cause analysis unit first determines multiple associated nodes according to the fault source node and at least one associated path, where the fault source node is one of the multiple associated nodes. According to the fault triggering conditions of each node in the knowledge graph, determine the node fault types of each associated node in the multiple associated nodes, and obtain multiple node fault types. Determine the fault cause according to the multiple node fault types, where the fault cause at least includes the node fault types of each associated node and the fault triggering conditions corresponding to each node fault type.

[0137] Then, the repair plan determination unit obtains multiple associated operation status data corresponding to multiple associated nodes, that is, the operation status data at the spatial position corresponding to each associated node in the multiple associated nodes. And obtain at least one potential fault operation status data corresponding to at least one potential fault node, that is, the operation status data at the spatial position corresponding to each potential fault node in the at least one potential fault node. Determine the abnormal data in the multiple associated operation status data to obtain at least one associated abnormal data. Determine at least one associated abnormal type according to the at least one associated abnormal data. Determine at least one candidate fault repair plan from multiple preset fault repair plans corresponding to the fault cause according to the at least one associated abnormal type, where each associated abnormal type corresponds to one candidate fault repair plan, and the mapping relationship between the associated abnormal type and the candidate fault repair plan can be preset in advance.

[0138] Furthermore, the repair plan determination unit determines at least one potential fault type corresponding to at least one potential fault node according to the at least one potential fault operation status data and at least one associated path, where each potential fault node corresponds to one potential fault type. Screen out the candidate fault repair plans that can solve the at least one potential fault type from the at least one candidate fault repair plan as the fault repair plan according to the at least one potential fault type.

[0139] Optionally, there may be multiple fault repair solutions. Based on the matching degree between each fault repair solution and at least one potential fault type, the multiple fault repair solutions can be sorted and presented to the user. In this way, by comprehensively analyzing the basic fault information and in-depth fault information, the fault repair solution can be determined, thereby improving the comprehensiveness and accuracy of fault diagnosis and repair.

[0140] Optionally, the severity level corresponding to each potential fault type in at least one potential fault type can be determined. The corresponding relationship between the potential fault type and the severity level can be preset in advance. From the at least one candidate fault repair solution, the candidate fault repair solution corresponding to the potential fault type with the highest corresponding severity level is selected as the fault repair solution, so that the fault with the highest severity level can be repaired preferentially, improving the safety of the main substation equipment.

[0141] In summary, the main substation equipment fault diagnosis and repair system described in this application includes: a data acquisition module, a data preprocessing module, a data fusion module, a fault diagnosis module, a fault analysis module, and a fault repair module, where: the data acquisition module is used to acquire multi-source operation status data of the main substation equipment; the data preprocessing module is used to preprocess the multi-source operation status data to obtain target multi-source operation status data; the data fusion module is used to perform data fusion on the target multi-source operation status data to obtain fusion data; the fault diagnosis module is used to determine the basic fault information of the main substation equipment according to the fusion data; the fault analysis module is used to determine the in-depth fault information of the main substation equipment according to the basic fault information; the fault repair module is used to determine the fault repair solution according to the basic fault information and the in-depth fault information. In this way, after preprocessing the multi-source operation status data of the main substation equipment and then performing data fusion, the fused data can reduce the influence of the environment and the complex internal structure of the main substation equipment, thereby improving the accuracy of fault diagnosis. Analyzing the main substation equipment for faults based on the fusion data can analyze all the basic fault information of the main substation equipment under the multi-source operation status data, avoiding missed detection or misdetection, and further improving the accuracy of fault diagnosis. Moreover, the in-depth fault information of the main substation equipment can be obtained by reasoning based on the basic fault information, so that the fault repair solution can be determined according to the basic fault information and the in-depth fault information, improving the accuracy of fault repair.

[0142] Refer to Figure 8 , Figure 8 which is a schematic flow chart of a method for diagnosing and repairing faults of a main substation equipment provided by an embodiment of this application. This method is applied to the main substation equipment fault diagnosis and repair system in any of the above embodiments. This method may include the following steps:

[0143] 801: Acquire multi-source operation status data of the main substation equipment;

[0144] 802: Preprocess the multi-source operation status data to obtain the target multi-source operation status data;

[0145] 803: Perform data fusion on the target multi-source operation status data to obtain the fused data;

[0146] 804: Determine the basic fault information of the main substation equipment according to the fused data;

[0147] 805: Determine the deep fault information of the main substation equipment according to the basic fault information;

[0148] 806: Determine the fault repair plan according to the basic fault information and the deep fault information.

[0149] In specific implementation, the method for diagnosing and repairing faults of the main substation equipment described in the embodiments of the present application may further include the functional implementation steps described in the system for diagnosing and repairing faults of the main substation equipment provided in any of the above embodiments, which will not be elaborated here.

[0150] Refer to Figure 9 , Figure 9 which is a schematic structural diagram of an electronic device provided in the embodiments of the present application. The electronic device 900 may include a processor 902 and a memory 903, and the processor 902 is connected to the memory 903. The electronic device 900 further includes a transceiver 901, and the transceiver 901, the processor 902, and the memory 903 may be interconnected through a bus 904. The memory 903 is used to store computer programs and data, and the processor 902 is used to execute the computer programs stored in the memory 903 so that the electronic device 900 executes the functional implementation steps and methods as described in any of the above embodiments. In the embodiments of the present application, the above computer programs include parts or all of the steps for executing the method for diagnosing and repairing faults of the main substation equipment.

[0151] The embodiments of the present application further provide a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and the computer program is executed by a processor to implement parts or all of the steps of the method as described in any of the above embodiments. The above computer includes the electronic device 900.

[0152] The embodiments of the present application further provide a computer program product. The above computer program product includes a non-transitory computer-readable storage medium storing a computer program, and the computer program is executed by a processor to implement parts or all of the steps of the method as described in any of the above embodiments. The computer program product may be a software installation package, and the above computer includes the electronic device 900.

[0153] It should be noted that, for the foregoing method embodiments, for the sake of simple description, they are all expressed as a series of action combinations. However, those skilled in the art should know that this application is not limited by the described action sequence, because according to this application, certain steps can be performed in other sequences or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily essential to this application.

[0154] In the above embodiments, the descriptions of the respective embodiments have their own focuses. For the parts not detailed in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0155] In several embodiments provided by this application, it should be understood that the disclosed device can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the above division of units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed couplings or direct couplings or communication connections to each other can be through some interfaces. The indirect couplings or communication connections of the devices or units can be in electrical or other forms.

[0156] Those of ordinary skill in the art can understand all or part of the processes in the above method embodiments. This process can be completed by relevant hardware instructed by a computer program. This program can be stored in a computer-readable storage medium. When this program is executed, it can include the processes of the above method embodiments. The foregoing storage media include: ROM or random access memory RAM, magnetic disks or optical disks and other various media that can store program codes.

[0157] The steps of the methods or algorithms described in the embodiments of this application can be implemented in a hardware manner or by a processor executing software instructions. The software instructions can be composed of corresponding software modules. The software modules can be stored in RAM, flash memory, ROM, EPROM, electrically erasable programmable read-only memory (EEPROM), registers, hard disks, removable hard disks, compact disc read-only memory (CD-ROM) or any other form of storage medium well-known in the art. An exemplary storage medium is coupled to the processor so that the processor can read information from the storage medium and write information to the storage medium. Of course, the storage medium can also be a component of the processor. The processor and the storage medium can be located in an ASIC. In addition, the ASIC can be located in a terminal device or a management device. Of course, the processor and the storage medium can also exist as discrete components in the terminal device or the management device.

[0158] Those skilled in the art should be able to realize that in one or more of the above examples, the functions described in the embodiments of the present application can be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the processes or functions described in the embodiments of the present application are generated in whole or in part.

[0159] The above computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from a website, computer, server, or data center to another website, computer, server, or data center in a wired manner (such as coaxial cable, optical fiber, digital subscriber line (DSL)) or wirelessly (such as infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that includes one or more integrated available media.

[0160] Among them, the available medium can be a magnetic medium (such as a floppy disk, hard disk, magnetic tape), an optical medium (such as a digital video disc (DVD)), or a semiconductor medium (such as a solid state disk (SSD)), etc.

[0161] Each device and product described in the above embodiments, and each module / unit included therein, may be a software module / unit, a hardware module / unit, or may be partly a software module / unit and partly a hardware module / unit. For example, for each device and product applied to or integrated into a chip, each module / unit included therein may be implemented in a hardware manner such as a circuit, or at least some of the modules / units may be implemented in the form of a software program that runs on a processor integrated inside the chip, and the remaining (if any) part of the modules / units may be implemented in a hardware manner such as a circuit; for each device and product applied to or integrated into a chip module, each module / unit included therein may be implemented in a hardware manner such as a circuit, and different modules / units may be located in the same component (such as a chip, a circuit module, etc.) or different components of the chip module, or at least some of the modules / units may be implemented in the form of a software program that runs on a processor integrated inside the chip module, and the remaining (if any) part of the modules / units may be implemented in a hardware manner such as a circuit; for each device and product applied to or integrated into a terminal device, each module / unit included therein may be implemented in a hardware manner such as a circuit, and different modules / units may be located in the same component (such as a chip, a circuit module, etc.) or different components inside the terminal device, or at least some of the modules / units may be implemented in the form of a software program that runs on a processor integrated inside the terminal device, and the remaining (if any) part of the modules / units may be implemented in a hardware manner such as a circuit.

[0162] The specific embodiments described above further elaborate on the objectives, technical solutions, and beneficial effects of the embodiments of the present application. It should be understood that the above description is only the specific embodiments of the embodiments of the present application and is not used to limit the protection scope of the embodiments of the present application. Any modifications, equivalent replacements, improvements, etc. made on the basis of the technical solutions of the embodiments of the present application shall be included in the protection scope of the embodiments of the present application.

Claims

1. A main substation equipment fault diagnosis and repair system, characterized in that, The system includes: a data acquisition module, a data preprocessing module, a data fusion module, a fault diagnosis module, a fault analysis module, and a fault repair module; The data acquisition module is used to acquire multi-source operation status data of the main substation equipment; The data preprocessing module is used to preprocess the multi-source operation status data to obtain target multi-source operation status data; The data fusion module is used to perform data fusion on the target multi-source operation status data to obtain fusion data; The fault diagnosis module is used to determine the basic fault information of the main substation equipment according to the fusion data; The fault analysis module is used to determine the deep fault information of the main substation equipment according to the basic fault information; The fault repair module is used to determine a fault repair plan according to the basic fault information and the deep fault information.

2. The system according to claim 1, wherein The multi-source operation status data includes multiple device images and multiple device status data of the main substation equipment; the target multi-source operation status data includes multiple target device images and multiple target device status data; In terms of preprocessing the multi-source operation status data to obtain target multi-source operation status data, the data preprocessing module includes: a denoising processing unit, a data alignment unit, a missing data filling unit, and an image enhancement unit, where: The denoising processing unit is used to perform wavelet decomposition on the multi-source operation status data according to a preset wavelet basis function and a preset decomposition level to obtain low-frequency wavelet coefficients and high-frequency wavelet coefficients; perform threshold processing on the high-frequency wavelet coefficients according to a preset threshold type to obtain target high-frequency wavelet coefficients; perform inverse wavelet transform according to the low-frequency wavelet coefficients and the target high-frequency wavelet coefficients to obtain denoised data; the denoised data includes multiple denoised device images and multiple denoised device status data; The data alignment unit is used to extract multiple feature point sets corresponding to the multiple denoised device images through the SIFT algorithm, and each denoised device image corresponds to a feature point set; align the multiple denoised device images spatially according to the multiple feature point sets to obtain multiple aligned device images; The missing data filling unit is used to obtain the missing part data of each denoised device status data in the multiple denoised device status data, and fill the missing part data with the preset number of nearest non-missing part data of the missing part data to obtain the multiple target device status data; The image enhancement unit is used to improve the contrast of the multiple aligned device images by using a contrast enhancement algorithm to obtain multiple enhanced images; perform edge sharpening on the multiple enhanced images by using an image edge sharpening algorithm to obtain the multiple target device images.

3. The system according to claim 2, characterized in that Each device status data corresponds to a spatial position; in terms of performing data fusion on the target multi-source operation status data to obtain fusion data, the data fusion module includes: a weight assignment unit, a data fusion unit, and a reliability evaluation unit, where: The weight allocation unit is used to determine the environmental parameters of the main substation equipment; according to the environmental parameters, determine multiple image weights corresponding to the multiple target device images, and each target device image corresponds to an image weight; The data fusion unit is used to determine display parameters corresponding to each target device image in the multiple target device images according to the multiple image weights, and obtain multiple display parameters; according to the multiple display parameters, display the display content corresponding to each target device image in the multiple target device images in a multi-dimensional space coordinate system, and obtain the multi-dimensional fusion model corresponding to the main substation equipment; according to the spatial position corresponding to the target device status data, fuse the target device status data and the multi-dimensional fusion model to obtain initial fusion data; The reliability evaluation unit is used to perform reliability evaluation on each data in the target multi-source operation status data through a Bayesian inference model to obtain an evaluation result; according to the evaluation result and the initial fusion data, determine the fusion data.

4. The system according to claim 2, wherein In terms of determining the basic fault information of the main substation equipment according to the fusion data, the fault diagnosis module includes a feature extraction unit and a feature analysis unit, where: The feature extraction unit is used to extract key features from the fusion data; perform data enhancement on the key features to obtain potential fault features; the potential fault features include: surface features, temperature distribution features, internal structure features, mechanical vibration features, and electrical data features of the main substation equipment; The feature analysis unit is used to determine a temperature fault diagnosis result according to the temperature distribution feature; input the surface feature and the internal structure feature into a preset convolutional neural network model to obtain a structural fault diagnosis result of the main substation equipment; analyze the mechanical vibration feature and the electrical data feature through a preset random forest model to obtain an electrical load fault diagnosis result; according to the temperature fault diagnosis result, the structural fault diagnosis result, and the electrical load fault diagnosis result, determine the basic fault information.

5. The system according to claim 4, characterized in that, Each device status data corresponds to an electrical data; in terms of determining the temperature fault diagnosis result according to the temperature distribution feature, the feature analysis unit specifically is used to: According to the temperature distribution feature, determine multiple temperatures corresponding to multiple spatial positions in the main substation equipment, and each spatial position corresponds to a temperature; Obtain the environmental temperature and working duration corresponding to the main substation equipment; Determine the difference between each temperature in the multiple temperatures and the environmental temperature to obtain multiple temperature differences; Determine a reference temperature difference according to the working duration and the electrical data corresponding to each spatial position in the multiple spatial positions; According to the multiple temperature differences and the reference temperature difference, determine multiple influence degrees corresponding to the multiple temperatures, and each temperature corresponds to an influence degree; According to the multiple influence degrees, determine at least one temperature fault position from the multiple spatial positions; Determine the temperature fault diagnosis result according to the at least one temperature fault location, the temperature corresponding to each temperature fault location, and the degree of influence.

6. The system according to any one of claims 1-5, characterized in that, The basic fault information includes: the spatial location of the faulty component and the basic fault type; in terms of determining the deep fault information of the main substation equipment according to the basic fault information, the fault analysis module includes: a knowledge graph construction unit, a linkage analysis unit, and a deep fault analysis unit, where: The knowledge graph construction unit is used to obtain the historical multi-source operation status data of the main substation equipment, the historical fault information, and the spatial locations of all components of the main substation equipment; construct a knowledge graph according to the historical multi-source operation status data, the historical fault information, and the spatial locations of all components of the main substation equipment; wherein, the knowledge graph includes a plurality of nodes and a plurality of paths, each node represents the spatial location of a component, and each path represents the influence relationship, fault type, and fault trigger condition of the operation status data between two components; The linkage analysis unit is used to determine at least one associated path corresponding to the basic fault type from the plurality of paths according to the basic fault type and the knowledge graph; determine the fault source node from the plurality of nodes according to the at least one associated path; The deep fault analysis unit is used to determine at least one potential fault node from the plurality of nodes and at least one potential fault path from the plurality of paths according to the fault source node and the at least one associated path; use the fault source node, the at least one associated path, the at least one potential fault node, and the at least one potential fault path as the deep fault information.

7. The system according to claim 6, wherein In terms of determining the fault repair plan according to the basic fault information and the deep fault information, the fault repair module includes: a fault cause analysis unit and a repair plan determination unit, where: The fault cause analysis unit is used to determine the node fault type of each associated node in the plurality of associated nodes corresponding to the at least one associated path according to the fault source node, the at least one associated path, and the basic fault type, to obtain a plurality of node fault types; determine the fault cause according to the plurality of node fault types; The repair plan determination unit is used to obtain the plurality of associated operation status data corresponding to the plurality of associated nodes, and obtain at least one potential fault operation status data corresponding to the at least one potential fault node, each associated node corresponds to one associated operation status data, and each potential fault node corresponds to one potential fault operation status data; determine at least one candidate fault repair plan from the plurality of preset fault repair plans corresponding to the fault cause according to the plurality of associated operation status data; determine at least one potential fault type corresponding to the at least one potential fault node according to the at least one potential fault operation status data and the at least one associated path; determine the fault repair plan from the at least one candidate fault repair plan according to the at least one potential fault type.

8. A fault diagnosis and repair method for main substation equipment, characterized in that, Include: Collect multi-source operation status data of main substation equipment; Preprocess the multi-source operation status data to obtain target multi-source operation status data; Perform data fusion on the target multi-source operation status data to obtain fusion data; Determine the basic fault information of the main substation equipment according to the fusion data; Determine the deep fault information of the main substation equipment according to the basic fault information; Determine a fault repair plan according to the basic fault information and the deep fault information.

9. An electronic device, characterized in that, Including: A processor and a memory, the processor is connected to the memory, the memory is used to store a computer program, and the processor is used to execute the computer program stored in the memory so that the electronic device executes the method according to claim 8.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, and the computer program is executed by a processor to implement the method according to claim 8.

Images