GIL equipment fault detection and positioning method and system based on automatic feature learning

Abstract

The invention provides a GIL equipment fault detection and positioning method and system based on automatic feature learning, and the method comprises the steps: carrying out the feature extraction of data within a preset time period before the occurrence of a fault based on the time point of the occurrence of various faults, and obtaining the abnormal features of the data; the abnormal features serve as input of a binary-classification early warning model, whether a fault occurs or not serves as an output label, and the binary-classification early warning model is trained; acquiring data in the GIL equipment in real time; performing automatic feature extraction on the data obtained in real time to obtain comprehensive feature representation; taking the comprehensive feature representation as the input of a dichotomy early warning model, obtaining the probability of fault occurrence, and triggering early warning when the probability of fault occurrence exceeds a preset threshold value; and starting fault diagnosis based on early warning, inputting the comprehensive feature representation for triggering early warning into the fault diagnosis model, and obtaining the fault type and position information. According to the application, the data features before the occurrence of the fault are utilized, the potential fault is early warned in advance, the fault type and position are quickly positioned, and preventive maintenance is realized.

Description

Technical Field

[0001] The present application relates to the technical field of GIL equipment fault detection and positioning, and specifically to a GIL equipment fault detection and positioning method and system based on automatic feature learning. Background Art

[0002] Current GIL equipment fault detection and location technology mainly focuses on immediate response after a fault occurs. Once a GIL device fails, the specific location of the fault is determined by using specialized detection tools and methods. However, this type of fault detection and location is often performed after the fault has occurred and has had an actual impact, which means that before the fault is detected, production or life may have been disrupted or damaged to varying degrees. For many critical industries, this delay may result in serious economic losses or safety risks.

[0003] Therefore, how to accurately determine the type and potential location of GIL equipment failure before it occurs has become a technical problem that needs to be solved urgently. This fault prediction and location capability can not only significantly improve maintenance efficiency and reduce maintenance costs, but more importantly, it can minimize the impact of failures on actual production and life, ensuring the stable operation of GIL equipment. Summary of the invention

[0004] In order to overcome the problem in the prior art that fault detection and location are often performed only after the fault has occurred and has produced actual impact, the present application provides a GIL equipment fault detection and location method and system based on automatic feature learning, which adopts the following technical solutions:

[0005] In a first aspect, the present application provides a GIL equipment fault detection and positioning method based on automatic feature learning, comprising:

[0006] Determine the time points when various types of faults occurred in historical operation data;

[0007] Based on the time points at which the various types of faults occur, feature extraction is performed on the data within a preset time period before the fault occurs, to obtain abnormal features of one or more of the partial discharge data, temperature data, voltage data, and current data;

[0008] The abnormal feature is used as the input of the binary classification early warning model, whether a fault occurs is used as the output label, and the binary classification early warning model is trained;

[0009] Obtain partial discharge data, temperature data, voltage data and current data in GIL equipment in real time;

[0010] Automatically extracting features from the partial discharge data, temperature data, voltage data and current data acquired in real time to obtain a comprehensive feature representation;

[0011] The comprehensive feature representation is used as the input of the binary classification warning model to obtain the probability of a fault occurring, and a warning is triggered when the probability of the fault occurring exceeds a preset threshold;

[0012] Fault diagnosis is initiated based on the early warning, and the comprehensive feature representation that triggers the early warning is input into the fault diagnosis model to obtain the fault type and location information.

[0013] Furthermore, the determining of the time points at which various types of faults occurred in the historical operation data includes:

[0014] Acquire historical operation data of the GIL device, wherein the historical operation data includes partial discharge data, temperature data, voltage data, and current data;

[0015] The historical operation data is marked according to the actual fault conditions, and the time points when various faults occurred are determined.

[0016] Furthermore, based on the time points at which the various types of faults occur, feature extraction is performed on the data within a preset time period before the fault occurs to obtain abnormal features of one or more of the partial discharge data, temperature data, voltage data, and current data, specifically including:

[0017] Feature extraction is performed on the partial discharge data, temperature data, voltage data and current data respectively. Based on the threshold values ​​of normal ranges of different features, when the feature value of a feature does not meet the threshold value of its corresponding normal range, the feature is an abnormal feature.

[0018] Furthermore, the abnormal feature is used as the input of the binary classification early warning model, whether a fault occurs is used as the output label, and the binary classification early warning model is trained, including:

[0019] Based on the actual occurrence of faults in historical operation data, the fault labels are marked for abnormal feature data. For abnormal data with faults, they are marked as "1"; for abnormal data without faults, they are marked as "0";

[0020] Performing peak analysis, spectrum analysis, statistical analysis, and trend analysis on the abnormal characteristics to obtain abnormal characteristic indicators;

[0021] The abnormal feature index is used as the input feature of the binary classification warning model, and the labeled abnormal data is used to train the binary classification warning model to learn the features before the fault occurs.

[0022] Furthermore, the automatic feature extraction of the partial discharge data, temperature data, voltage data and current data to obtain a comprehensive feature representation includes:

[0023] Acquire spatial features of the partial discharge data, temperature data, voltage data and current data based on a convolutional neural network;

[0024] Acquire the time series characteristics of the partial discharge data, temperature data, voltage data and current data based on a long short-term memory network;

[0025] The spatial feature is fused with the time series feature to obtain a comprehensive feature representation.

[0026] Further, the said starting fault diagnosis based on early warning, inputting the said comprehensive characteristic representation that triggers the early warning into the fault diagnosis model, and obtaining the fault type and location information, comprises:

[0027] Collect various types of fault data that have occurred in GIL equipment from historical operation data, including abnormal characteristics, fault types, and fault location annotation information;

[0028] Establish a mapping relationship between abnormal characteristics, fault types and fault locations based on the annotation information;

[0029] Based on the mapping relationship and the fault diagnosis model, the fault type and fault location represented by the comprehensive features that trigger the early warning are obtained.

[0030] In a second aspect, the present application also provides a GIL equipment fault detection and positioning system based on automatic feature learning, comprising:

[0031] A module for determining the time point at which various faults occur, used to determine the time points at which various faults occur in historical operation data;

[0032] An abnormal feature acquisition module is used to extract features of data within a preset time period before the occurrence of the fault based on the time point of occurrence of the various faults, and obtain abnormal features of one or more of the partial discharge data, temperature data, voltage data and current data;

[0033] A binary classification warning model training module is used to use the abnormal features as the input of the binary classification warning model and whether a fault occurs as the output label to train the binary classification warning model;

[0034] Real-time data acquisition module, used to acquire partial discharge data, temperature data, voltage data and current data in GIL equipment in real time;

[0035] A comprehensive feature representation acquisition module is used to automatically extract features from the partial discharge data, temperature data, voltage data and current data acquired in real time to obtain a comprehensive feature representation;

[0036] An early warning trigger module is used to use the comprehensive feature representation as the input of the binary early warning model to obtain the probability of a fault occurring, and trigger an early warning when the probability of a fault occurring exceeds a preset threshold;

[0037] The fault type and location information acquisition module is used to start fault diagnosis based on the early warning, input the comprehensive feature representation that triggers the early warning into the fault diagnosis model, and obtain the fault type and location information.

[0038] In a third aspect, the present application provides an electronic device, including:

[0039] One or more processors; a memory; and one or more computer programs, wherein the one or more computer programs are stored in the memory, and the one or more computer programs include instructions, which, when executed by the device, cause the device to perform the method as described in the first aspect.

[0040] In a fourth aspect, the present application provides a computer-readable storage medium, in which a computer program is stored. When the computer-readable storage medium is run on a computer, the computer executes the method described in the first aspect.

[0041] In a fifth aspect, the present application provides a computer program, which, when executed by a computer, is used to execute the method described in the first aspect.

[0042] In one possible design, the program in the fifth aspect may be stored in whole or in part on a storage medium packaged together with the processor, or may be stored in whole or in part on a memory not packaged together with the processor.

[0043] This application has the following beneficial effects:

[0044] 1. This application determines the time points when various types of failures occur in historical operation data. This application specifically studies the data characteristics before the failure occurs, avoids the analysis of irrelevant data, and improves the efficiency of data processing.

[0045] 2. This application extracts features from data within a preset time period before the occurrence of various types of faults based on the time point at which the fault occurs, and obtains abnormal features of one or more of the partial discharge data, temperature data, voltage data, and current data; this application helps to extract key features related to the fault by focusing on the data within a preset time period before the fault occurs. By analyzing partial discharge, temperature, voltage, and current data, the abnormal features of the equipment before the fault can be captured, and these abnormal features can more accurately reflect the potential problems of the equipment.

[0046] 3. This application trains a binary classification warning model by taking abnormal features as the input of the binary classification warning model and whether a fault occurs as the output label. This application trains a binary classification model by taking abnormal features as the input and whether a fault occurs as the output label. It can directly establish the relationship between abnormal features and the occurrence of faults. The binary classification warning model can learn the relationship between these feature patterns and the possibility of faults, providing a data basis for real-time warnings.

[0047] 4. This application studies historical fault data and monitors real-time data, and by utilizing data features before a fault occurs, it can provide early warning of potential faults, and can provide timely warnings before a fault occurs, providing the possibility of taking preventive measures and achieving preventive maintenance, which helps to reduce the frequency of GIL equipment failures, extend the service life of GIL equipment, and reduce maintenance costs and production losses. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 is an exemplary system architecture diagram to which the embodiments of the present application can be applied;

[0049] Figure 2 This is a flow chart of a GIL equipment fault detection and positioning method based on automatic feature learning according to an embodiment of the present application;

[0050] Figure 3 A flow chart for determining the time points at which various faults occur in the embodiments of the present application;

[0051] Figure 4 This is a flow chart of training a binary classification warning model according to an embodiment of the present application;

[0052] Figure 5 A flowchart of obtaining comprehensive features according to an embodiment of the present application;

[0053] Figure 6 A flowchart of obtaining fault type and location information according to an embodiment of the present application;

[0054] Figure 7 This is a system flow chart of an embodiment of the present application;

[0055] Figure 8 It is a schematic diagram of a computer device according to an embodiment of the present application. DETAILED DESCRIPTION

[0056] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by technicians in the technical field of the present application; the terms used in the specification of the application herein are only for the purpose of describing specific embodiments and are not intended to limit the present application; the terms "including" and "having" and any variations thereof in the specification and claims of the present application and the above-mentioned drawings are intended to cover non-exclusive inclusions. The terms "first", "second", etc. in the specification and claims of the present application or the above-mentioned drawings are used to distinguish different objects, not to describe a specific order.

[0057] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0058] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the accompanying drawings.

[0059] like Figure 1 As shown, the system architecture 100 may include terminal devices 101, 102, 103, a network 104 and a server 105. The network 104 is used to provide a medium for communication links between the terminal devices 101, 102, 103 and the server 105. The network 104 may include various connection types, such as wired, wireless communication links or optical fiber cables, etc.

[0060] Users can use terminal devices 101, 102, 103 to interact with server 105 through network 104 to receive or send messages, etc. Various communication client applications can be installed on terminal devices 101, 102, 103, such as web browser applications, shopping applications, search applications, instant messaging tools, email clients, social platform software, etc.

[0061] Terminal devices 101, 102, 103 can be various electronic devices with display screens and supporting web browsing, including but not limited to smart phones, tablet computers, e-book readers, MP3 players (Moving Picture Experts Group Audio Layer III, Moving Picture Experts Group Audio Layer 3), MP4 (Moving Picture Experts Group Audio Layer IV, Moving Picture Experts Group Audio Layer 4) players, laptop computers and desktop computers, etc.

[0062] The server 105 may be a server that provides various services, such as a background server that provides support for web pages displayed on the terminal devices 101 , 102 , and 103 .

[0063] It should be noted that the GIL equipment fault detection and positioning method based on automatic feature learning provided in the embodiment of the present application is generally executed by a server / terminal device, and accordingly, the GIL equipment fault detection and positioning system based on automatic feature learning is generally set in the server / terminal device.

[0064] It should be understood that Figure 1 The number of terminal devices, networks and servers in the embodiment is only for illustration. Any number of terminal devices, networks and servers may be provided according to implementation requirements.

[0065] Continue to refer Figure 2 , the figure shows a flow chart of a GIL equipment fault detection and positioning method based on automatic feature learning of the present application, the method comprising the following steps:

[0066] Step 201, determine the time point when various faults occurred in the historical operation data. The acquired historical operation data includes the collected value, time and location.

[0067] In a possible implementation manner, the time point at which each type of fault occurs in the historical operation data is determined, please refer to Figure 3 , the specific contents include:

[0068] Step 31, obtaining historical operation data of the GIL device, wherein the historical operation data includes partial discharge data, temperature data, voltage data and current data.

[0069] In a possible implementation, historical operation data of the GIL device is collected from an operation monitoring system, maintenance records, and a historical database of the GIL device.

[0070] Step 32: annotate the historical operation data according to the actual fault conditions, and determine the time points when various faults occurred.

[0071] Step 202, based on the time points at which the various types of faults occur, feature extraction is performed on the data within a preset time period before the fault occurs, to obtain abnormal features of one or more of the partial discharge data, temperature data, voltage data, and current data;

[0072] In one possible implementation, for each fault occurrence time point, a preset time period before the fault occurs is determined, and the length of the time period can be determined based on the characteristics of the fault type; for example, for some relatively slow-developing faults, such as faults caused by insulation aging, data from several hours or even days before the fault occurs can be selected; for some sudden faults, such as short circuit faults, data from several minutes before the fault occurs may be selected.

[0073] In a possible implementation, based on the time points at which the various types of faults occur, feature extraction is performed on data within a preset time period before the fault occurs to obtain abnormal features of one or more of the partial discharge data, temperature data, voltage data, and current data, specifically including:

[0074] Feature extraction is performed on the partial discharge data, temperature data, voltage data and current data respectively. Based on the threshold values ​​of normal ranges of different features, when the feature value of a feature does not meet the threshold value of its corresponding normal range, the feature is an abnormal feature.

[0075] Step 203, taking the abnormal features as input of a binary classification warning model, taking whether a fault occurs as an output label, and training the binary classification warning model; the binary classification warning model in the present application can learn the features before a fault occurs.

[0076] In a possible implementation, the abnormal feature is used as the input of the early warning model, whether a fault occurs is used as the output label, and a two-classification early warning model is trained. Please refer to Figure 4 , the specific contents include:

[0077] Step 41, based on the actual occurrence of faults in the historical operation data, the abnormal characteristic data is labeled with a fault label, and the abnormal data with faults is labeled as "1"; the abnormal data without faults is labeled as "0".

[0078] Step 42, performing peak analysis, spectrum analysis, statistical analysis and trend analysis on the abnormal characteristics to obtain abnormal characteristic indicators;

[0079] Step 43, using the abnormal feature index as the input feature of the binary classification warning model, using the labeled abnormal data to train the binary classification warning model, and learning the features before the fault occurs.

[0080] Step 204, obtaining partial discharge data, temperature data, voltage data and current data in the GIL device in real time; the obtained real-time data includes the collected value, time and position.

[0081] In a possible implementation, partial discharge sensors, temperature sensors, voltage sensors and current sensors are installed at key locations of the GIL equipment; partial discharge data, temperature data, voltage data and current data are collected based on preset time periods.

[0082] For example, for partial discharge data, since it may change rapidly and is sensitive to early fault detection, a shorter collection time interval can be set, such as once every microsecond; temperature data changes relatively slowly and can be collected once every second; the collection time interval for voltage and current data can be set according to the electrical characteristics and operating frequency of the equipment, such as once every millisecond.

[0083] Step 205, automatically extracting features from the partial discharge data, temperature data, voltage data and current data acquired in real time to obtain a comprehensive feature representation;

[0084] In a possible implementation manner, the local discharge data, temperature data, voltage data and current data are automatically extracted to obtain a comprehensive feature representation. Please refer to Figure 5 , specifically:

[0085] Step 51, obtaining spatial features of the partial discharge data, temperature data, voltage data and current data based on a convolutional neural network;

[0086] Step 52, obtaining time series characteristics of the partial discharge data, temperature data, voltage data and current data based on a long short-term memory network;

[0087] Step 53: Fusing the spatial feature with the time series feature to obtain a comprehensive feature representation.

[0088] In a possible implementation, the spatial features are fused with the time series features to obtain a comprehensive feature representation. A simple splicing method can be used to sequentially splice the spatial feature vector extracted by the convolutional neural network and the time series feature vector extracted by the long short-term memory network to form a comprehensive feature vector. For example, if the dimension of the spatial feature vector output by the convolutional neural network is 128, and the dimension of the time series feature vector output by the long short-term memory network is 64, then the dimension of the spliced ​​comprehensive feature vector is 192; or weighted summation, attention mechanism, etc. are used for fusion. In the weighted summation method, a weight vector is learned to perform weighted summation on the features of the convolutional neural network and the long short-term memory network; the attention mechanism automatically assigns different weights according to the importance of different features to achieve more effective feature fusion.

[0089] Step 206, using the comprehensive feature representation as the input of the binary classification warning model to obtain the probability of a fault occurring, and triggering a warning when the probability of a fault occurring exceeds a preset threshold;

[0090] Step 207 , starting fault diagnosis based on the early warning, inputting the comprehensive feature representation that triggers the early warning into a fault diagnosis model, and acquiring fault type and location information.

[0091] In a possible implementation, the fault diagnosis is initiated based on the early warning, and the comprehensive feature representation that triggers the early warning is input into the fault diagnosis model to obtain the fault type and location information. Please refer to Figure 6 , the specific contents include:

[0092] Step 61, collecting various fault data that have occurred in the GIL equipment from the historical operation data, including annotated information of abnormal characteristics, fault types and fault locations;

[0093] Step 62, establishing a mapping relationship between abnormal characteristics, fault types and fault locations based on the annotation information;

[0094] Step 63: Based on the mapping relationship and the fault diagnosis model, the fault type and fault location represented by the comprehensive features that trigger the early warning are obtained.

[0095] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program, and the computer program can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, the aforementioned storage medium can be a non-volatile storage medium such as a disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM).

[0096] It should be understood that, although the steps in the flowchart of the accompanying drawings are displayed in sequence as indicated by the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least a part of the steps in the flowchart of the accompanying drawings may include multiple sub-steps or multiple stages, and these sub-steps or stages are not necessarily executed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be executed in turn or alternately with other steps or at least a part of the sub-steps or stages of other steps.

[0097] Continue to refer Figure 7The GIL equipment fault detection and positioning system based on automatic feature learning described in this embodiment includes:

[0098] A module 701 for determining the time point at which each type of fault occurs, for determining the time point at which each type of fault occurs in the historical operation data;

[0099] The abnormal feature acquisition module 702 is used to extract features of data within a preset time period before the occurrence of the fault based on the time point of occurrence of each type of fault, and obtain abnormal features of one or more of the partial discharge data, temperature data, voltage data and current data;

[0100] A binary classification warning model training module 703 is used to use the abnormal features as the input of the binary classification warning model and whether a fault occurs as the output label to train the binary classification warning model;

[0101] A real-time data acquisition module 704 is used to acquire partial discharge data, temperature data, voltage data and current data in the GIL device in real time;

[0102] A comprehensive feature representation acquisition module 705 is used to automatically extract features from the partial discharge data, temperature data, voltage data and current data acquired in real time to obtain a comprehensive feature representation;

[0103] The warning trigger module 706 is used to use the comprehensive feature representation as the input of the binary classification warning model to obtain the probability of the fault occurring, and trigger a warning when the probability of the fault occurring exceeds a preset threshold;

[0104] The fault type and location information acquisition module 707 is used to start fault diagnosis based on the early warning, input the comprehensive feature representation that triggers the early warning into the fault diagnosis model, and obtain the fault type and location information.

[0105] To solve the above technical problems, the present application also provides a computer device. Figure 8 , Figure 8 This is a basic structural block diagram of the computer device in this embodiment.

[0106] The computer device 8 includes a memory 8a, a processor 8b, and a network interface 8c that are interconnected and communicated through a system bus. It should be noted that the figure only shows a computer device 8 with components 8a-8c, but it should be understood that it is not required to implement all the components shown, and more or fewer components can be implemented instead. Among them, those skilled in the art can understand that the computer device here is a device that can automatically perform numerical calculations and / or information processing according to pre-set or stored instructions, and its hardware includes but is not limited to microprocessors, application specific integrated circuits (Application Specific Integrated Circuit, ASIC), programmable gate arrays (Field-Programmable Gate Array, FPGA), digital processors (Digital Signal Processor, DSP), embedded devices, etc.

[0107] The computer device may be a computing device such as a desktop computer, a notebook, a PDA, a cloud server, etc. The computer device may interact with a user through a keyboard, a mouse, a remote controller, a touch pad, or a voice control device.

[0108] The memory 8a includes at least one type of readable storage medium, and the readable storage medium includes flash memory, hard disk, multimedia card, card-type memory (for example, SD or DX memory, etc.), random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic memory, disk, optical disk, etc. In some embodiments, the memory 8a can be an internal storage unit of the computer device 8, such as a hard disk or memory of the computer device 8. In other embodiments, the memory 8a can also be an external storage device of the computer device 8, such as a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (FlashCard), etc. equipped on the computer device 8. Of course, the memory 8a can also include both the internal storage unit of the computer device 8 and its external storage device. In this embodiment, the memory 8a is generally used to store the operating system and various application software installed on the computer device 8, such as the program code of the GIL device fault detection and positioning method based on automatic feature learning, etc. In addition, the memory 8a can also be used to temporarily store various types of data that have been output or are to be output.

[0109] The processor 8b may be a central processing unit (CPU), a controller, a microcontroller, a microprocessor, or other data processing chips in some embodiments. The processor 8b is generally used to control the overall operation of the computer device 8. In this embodiment, the processor 8b is used to run the program code stored in the memory 8a or process data, such as running the program code of the GIL device fault detection and positioning method based on automatic feature learning.

[0110] The network interface 8c may include a wireless network interface or a wired network interface. The network interface 8c is generally used to establish a communication connection between the computer device 8 and other electronic devices.

[0111] The present application also provides another embodiment, namely, providing a non-volatile computer-readable storage medium, wherein the non-volatile computer-readable storage medium stores a program of a GIL device fault detection and positioning method based on automatic feature learning, and the GIL device fault detection and positioning method based on automatic feature learning can be executed by at least one processor so that the at least one processor performs the steps of the GIL device fault detection and positioning method based on automatic feature learning as described above.

[0112] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes a number of instructions for a terminal device (which can be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in each embodiment of the present application.

[0113] Obviously, the embodiments described above are only some embodiments of the present application, rather than all embodiments. The preferred embodiments of the present application are given in the accompanying drawings, but they do not limit the patent scope of the present application. The present application can be implemented in many different forms. On the contrary, the purpose of providing these embodiments is to make the understanding of the disclosure of the present application more thorough and comprehensive. Although the present application is described in detail with reference to the aforementioned embodiments, for those skilled in the art, it is still possible to modify the technical solutions recorded in the aforementioned specific implementation methods, or to perform equivalent replacement of some of the technical features therein. Any equivalent structure made using the contents of the specification and drawings of this application, directly or indirectly used in other related technical fields, is similarly within the scope of patent protection of this application.

Claims

1. A GIL equipment fault detection and location method based on automatic feature learning, characterized in that: include: Determine the time points when various types of faults occurred in historical operation data; Based on the time points at which the various types of faults occur, feature extraction is performed on the data within a preset time period before the fault occurs, to obtain abnormal features of one or more of the partial discharge data, temperature data, voltage data, and current data; The abnormal feature is used as the input of the binary classification early warning model, whether a fault occurs is used as the output label, and the binary classification early warning model is trained; Obtain partial discharge data, temperature data, voltage data and current data in GIL equipment in real time; Automatically extracting features from the partial discharge data, temperature data, voltage data and current data acquired in real time to obtain a comprehensive feature representation; The comprehensive feature representation is used as the input of the binary classification warning model to obtain the probability of a fault occurring, and a warning is triggered when the probability of the fault occurring exceeds a preset threshold; Fault diagnosis is initiated based on the early warning, and the comprehensive feature representation that triggers the early warning is input into the fault diagnosis model to obtain the fault type and location information.

2. The GIL equipment fault detection and positioning method based on automatic feature learning according to claim 1 is characterized in that: Determining the time points at which various types of faults occurred in the historical operation data includes: Acquire historical operation data of the GIL device, wherein the historical operation data includes partial discharge data, temperature data, voltage data, and current data; The historical operation data is marked according to the actual fault conditions, and the time points when various faults occurred are determined.

3. The GIL equipment fault detection and positioning method based on automatic feature learning according to claim 1 is characterized in that: Based on the time points of occurrence of the various types of faults, feature extraction is performed on the data within a preset time period before the fault occurs to obtain abnormal features of one or more of the partial discharge data, temperature data, voltage data and current data, and the specific contents include: Feature extraction is performed on the partial discharge data, temperature data, voltage data and current data respectively. Based on the thresholds of normal ranges of different features, when the feature value of a feature does not meet the threshold of its corresponding normal range, the feature is an abnormal feature.

4. The GIL equipment fault detection and positioning method based on automatic feature learning according to claim 1 is characterized in that: The abnormal feature is used as the input of the binary classification warning model, whether a fault occurs is used as the output label, and the binary classification warning model is trained, including: Based on the actual occurrence of faults in historical operation data, the fault labels are marked for abnormal feature data. For abnormal data with faults, they are marked as "1"; for abnormal data without faults, they are marked as "0"; Performing peak analysis, spectrum analysis, statistical analysis, and trend analysis on the abnormal characteristics to obtain abnormal characteristic indicators; The abnormal feature index is used as the input feature of the binary classification warning model, and the labeled abnormal data is used to train the binary classification warning model to learn the features before the fault occurs.

5. The GIL equipment fault detection and positioning method based on automatic feature learning according to claim 1 is characterized in that: The automatic feature extraction of the partial discharge data, temperature data, voltage data and current data to obtain a comprehensive feature representation includes: Acquire spatial features of the partial discharge data, temperature data, voltage data and current data based on a convolutional neural network; Acquire the time series characteristics of the partial discharge data, temperature data, voltage data and current data based on a long short-term memory network; The spatial feature is fused with the time series feature to obtain a comprehensive feature representation.

6. The GIL equipment fault detection and positioning method based on automatic feature learning according to claim 1 is characterized in that: The method of starting fault diagnosis based on early warning, inputting the comprehensive feature representation that triggers the early warning into a fault diagnosis model, and obtaining fault type and location information includes: Collect various types of fault data that have occurred in GIL equipment from historical operation data, including abnormal characteristics, fault types, and fault location annotation information; Establish a mapping relationship between abnormal characteristics, fault types and fault locations based on the annotation information; Based on the mapping relationship and the fault diagnosis model, the fault type and fault location represented by the comprehensive features that trigger the early warning are obtained.

7. A GIL equipment fault detection and positioning system based on automatic feature learning, used to implement the GIL equipment fault detection and positioning method based on automatic feature learning of claims 1-6, characterized in that: include: A module for determining the time point at which various faults occur, used to determine the time points at which various faults occur in historical operation data; An abnormal feature acquisition module is used to extract features of data within a preset time period before the occurrence of the fault based on the time point of occurrence of the various faults, and obtain abnormal features of one or more of the partial discharge data, temperature data, voltage data and current data; A binary classification warning model training module is used to use the abnormal features as the input of the binary classification warning model and whether a fault occurs as the output label to train the binary classification warning model; Real-time data acquisition module, used to acquire partial discharge data, temperature data, voltage data and current data in GIL equipment in real time; A comprehensive feature representation acquisition module is used to automatically extract features from the partial discharge data, temperature data, voltage data and current data acquired in real time to obtain a comprehensive feature representation; An early warning trigger module is used to use the comprehensive feature representation as the input of the binary early warning model to obtain the probability of a fault occurring, and trigger an early warning when the probability of a fault occurring exceeds a preset threshold; The fault type and location information acquisition module is used to start fault diagnosis based on the early warning, input the comprehensive feature representation that triggers the early warning into the fault diagnosis model, and obtain the fault type and location information.

8. An electronic device, characterized in that: include: one or more processors; Memory; and one or more computer programs, wherein the one or more computer programs are stored in the memory, and the one or more computer programs include instructions, which, when executed by the device, enable the device to perform the steps of the GIL device fault detection and positioning method based on automatic feature learning as described in any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, which, when executed on a computer, enables the computer to execute the steps of the GIL equipment fault detection and positioning method based on automatic feature learning as claimed in any one of claims 1 to 6.

Images