Transformer internal insulation breakdown risk assessment method, device, equipment and medium
By collecting transformer electric field data in real time and entering the evaluation model to evaluate the health status of insulation, the problem of insufficient sensitivity of traditional detection technology is solved, and efficient and accurate monitoring and fault prediction of internal insulation of transformer is achieved, ensuring the continuous operation of the power system and the extended equipment life. .
Patent Information
- Application Number
- CN202411017720.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-29
- Publication Date
- 2024-12-06
AI Technical Summary
Traditional transformer insulation detection technology has limited sensitivity and prediction capabilities when detecting early insulation deterioration, and cannot effectively foresee the future failure risk of insulating materials. It also requires equipment shutdown to increase operation and maintenance costs, affecting the continuity of power supply.
By obtaining the transformer electric field data collected in real time, inputting the trained evaluation model, evaluating the insulation health status and outputting the insulation health index, triggering an early warning mechanism to achieve real-time, accurate and efficient monitoring of the internal insulation of the transformer.
Without interrupting the operation of the equipment, improve the real-time, accuracy and predictiveness of transformer insulation monitoring, extend the equipment life, and improve the operating efficiency and reliability of the power system.
Smart Images

Figure CN119090256A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of power equipment health monitoring and maintenance, and in particular to a transformer internal insulation breakdown risk assessment method, device, equipment and medium. Background Art
[0002] In the power transmission and distribution system, the transformer is a core component, and the integrity of its internal insulation is crucial to ensure the efficient and safe operation of the system. Degradation or damage of insulation materials is one of the main reasons for transformer performance degradation or even failure. Therefore, accurate evaluation and real-time monitoring of the internal insulation status of the transformer is the key to preventing failures and extending the life of the equipment.
[0003] Traditional transformer insulation testing techniques typically rely on physical sampling and electrical property testing, such as dielectric loss testing and partial discharge measurements. These methods can provide useful information about the condition of insulation materials, but they have many limitations. First, these tests often require equipment shutdown, which not only increases operating and maintenance costs but also affects the continuity of power supply. Second, these methods have limited sensitivity and predictive power in detecting early insulation degradation and cannot effectively foresee the future failure risk of insulation materials.
[0004] With the development of technology, especially in the field of sensor technology and data analysis, the demand for real-time online monitoring of transformer insulation is increasing day by day. In summary, a more efficient and safe transformer insulation monitoring solution is urgently needed to improve the real-time, accuracy and predictability of transformer internal insulation monitoring. Summary of the invention
[0005] The main purpose of the present invention is to provide a transformer internal insulation breakdown risk assessment method, device, equipment and medium, which can improve the real-time, accuracy and predictability of transformer internal insulation monitoring.
[0006] To achieve the above objectives, the present application provides a method for assessing the risk of insulation breakdown inside a transformer, the method comprising:
[0007] Obtain real-time collected transformer electric field data;
[0008] Inputting the transformer electric field data into a trained evaluation model, evaluating the health status of the transformer internal insulation through the trained evaluation model, and outputting a target insulation health index;
[0009] An early warning mechanism is triggered according to the target insulation health index.
[0010] Optionally, before acquiring the transformer electric field data collected in real time, the method further includes:
[0011] Obtain transformer electric field data as sample data;
[0012] Inputting the sample data into a preset model for training to obtain a preliminary evaluation model;
[0013] The preliminary evaluation model is used to cross-validate historical electric field data to obtain the trained evaluation model.
[0014] Optionally, before obtaining the transformer electric field data as sample data, the method further includes:
[0015] The layout plan of the electric field sensors is determined based on the theoretical distribution of the electric field inside the transformer.
[0016] Optionally, determining the layout scheme of the electric field sensors according to the theoretical distribution of the electric field inside the transformer includes but is not limited to the following steps:
[0017] Spatial distribution analysis, identification of key monitoring points, determination of sensor quantity and type, optimal layout strategy, installation and calibration, testing and adjustment.
[0018] Optionally, the spatial distribution analysis specifically includes:
[0019] By building a three-dimensional model inside the transformer and simulating the electric field distribution, the area where the electric field intensity changes most significantly is determined;
[0020] The identification of key monitoring points specifically includes:
[0021] Based on the simulation results of the electric field distribution, the key area with the largest electric field intensity gradient is identified;
[0022] The number and type of sensors are determined, including:
[0023] Determine the number and type of sensors required based on monitoring requirements and the specific size and configuration of the transformer;
[0024] The optimal layout strategy specifically includes:
[0025] An optimization algorithm is used to determine the best placement of the required sensors, wherein the optimization algorithm aims to cover all key monitoring points and maximize overall monitoring efficiency and accuracy.
[0026] Optionally, the preset model adopts a convolutional neural network structure optimized by a residual learning mechanism;
[0027] The specific structure of the preset model includes:
[0028] Input layer, first convolutional layer, first residual block, second residual block, third residual block, fourth residual block, fully connected layer and output layer;
[0029] Each grid of the input layer represents the electric field strength at a measurement point.
[0030] Optionally, the preset model adopts a convolutional neural network structure optimized by a residual learning mechanism;
[0031] The specific structure of the preset model includes:
[0032] Input layer, first convolutional layer, first residual block, second residual block, third residual block, fourth residual block, fully connected layer and output layer;
[0033] Each grid of the input layer represents the electric field strength at a measurement point.
[0034] On the other hand, the present application also provides a transformer internal insulation breakdown risk assessment device, comprising:
[0035] A data acquisition module, used to acquire transformer electric field data collected in real time;
[0036] An evaluation module, used to input the transformer electric field data into a trained evaluation model, evaluate the health status of the transformer internal insulation through the trained evaluation model, and output a target insulation health index;
[0037] The early warning module is used to trigger an early warning mechanism according to the target insulation health index.
[0038] Optionally, the device further comprises a training module, which is used to:
[0039] Before obtaining the transformer electric field data collected in real time, obtaining the transformer electric field data as sample data;
[0040] Inputting the sample data into a preset model for training to obtain a preliminary evaluation model;
[0041] The preliminary evaluation model is used to cross-validate historical electric field data to obtain the trained evaluation model.
[0042] Optionally, the device further includes a layout module, configured to:
[0043] Before obtaining the transformer electric field data as sample data, the electric field sensor layout scheme is determined according to the theoretical distribution of the electric field inside the transformer.
[0044] Optionally, the layout module is specifically used to perform the following steps:
[0045] Spatial distribution analysis, identification of key monitoring points, determination of sensor quantity and type, optimal layout strategy, installation and calibration, testing and adjustment.
[0046] Optionally, the layout module is specifically used to:
[0047] By building a three-dimensional model inside the transformer and simulating the electric field distribution, the area where the electric field intensity changes most significantly is determined;
[0048] Based on the simulation results of the electric field distribution, the key area with the largest electric field intensity gradient is identified;
[0049] Determine the number and type of sensors required based on monitoring requirements and the specific size and configuration of the transformer;
[0050] An optimization algorithm is used to determine the best placement of the required sensors, wherein the optimization algorithm aims to cover all key monitoring points and maximize overall monitoring efficiency and accuracy.
[0051] Optionally, the preset model adopts a convolutional neural network structure optimized by a residual learning mechanism;
[0052] The specific structure of the preset model includes:
[0053] Input layer, first convolutional layer, first residual block, second residual block, third residual block, fourth residual block, fully connected layer and output layer;
[0054] Each grid of the input layer represents the electric field strength at a measurement point.
[0055] Optionally, the early warning module is specifically used to:
[0056] The target insulation health index is compared with an index threshold; when the target insulation health index is lower than the index threshold, the early warning mechanism is triggered to provide data-driven maintenance suggestions to support operation and maintenance decisions.
[0057] On the other hand, the present application provides an electronic device, including a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of the first aspect and any possible implementation method thereof.
[0058] On the other hand, the present application provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, the processor executes each step of the method described in the first aspect.
[0059] The present application provides a method, device, equipment and medium for assessing the risk of transformer internal insulation breakdown, which obtains real-time collected transformer electric field data; inputs the transformer electric field data into a trained assessment model, assesses the health status of the transformer internal insulation through the trained assessment model, and outputs a target insulation health index; triggers an early warning mechanism based on the target insulation health index, and can assess the risk of transformer internal insulation breakdown without interrupting the normal operation of the equipment, achieve continuous monitoring of the transformer insulation performance and fault prediction, so as to improve the operating efficiency and reliability of the power system, and can improve the real-time, accuracy and predictability of transformer internal insulation monitoring. BRIEF DESCRIPTION OF THE DRAWINGS
[0060] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0061] in:
[0062] Figure 1 A schematic flow chart of a transformer internal insulation breakdown risk assessment method provided in an embodiment of the present application;
[0063] Figure 2 A flow chart of another transformer internal insulation breakdown risk assessment method provided in an embodiment of the present application;
[0064] Figure 3 A schematic structural diagram of a transformer internal insulation breakdown risk assessment device provided in an embodiment of the present application. DETAILED DESCRIPTION
[0065] In order to enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in the field without creative work are within the scope of protection of the present application.
[0066] The terms "first", "second", etc. in the specification and claims of this application and the above-mentioned drawings are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. 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 includes steps or units that are not listed, or optionally includes other steps or units inherent to these processes, methods, products or devices.
[0067] 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.
[0068] The basic idea of cross validation mentioned in the embodiments of the present application is to group the original data (dataset) in a certain sense, with one part as a training set (train set) and the other part as a validation set (validation set or test set). The classifier is first trained with the training set, and then the validation set is used to test the trained model, which is used as a performance indicator for evaluating the classifier.
[0069] The embodiments of the present application are described below in conjunction with the drawings in the embodiments of the present application.
[0070] See also Figure 1 , is a flow chart of a transformer internal insulation breakdown risk assessment method provided in an embodiment of the present application. Figure 1 As shown, the method includes:
[0071] 101. Obtaining transformer electric field data collected in real time;
[0072] 102. Input the above transformer electric field data into a trained evaluation model, evaluate the health status of the transformer internal insulation through the trained evaluation model, and output a target insulation health index;
[0073] 103. Trigger an early warning mechanism based on the above target insulation health index.
[0074] The embodiment of the present application mainly proposes a transformer internal insulation breakdown risk assessment method, which is suitable for real-time assessment of the health status of the internal insulation system of a high-voltage power transformer.
[0075] The execution subject of the method in the embodiment of the present application is a transformer internal insulation breakdown risk assessment device. In practical applications, it can be executed on an electronic device such as a terminal device.
[0076] The preset model can adopt a convolutional neural network structure (RES-SKIPCNN) optimized by residual learning mechanism to train the required evaluation model. In practical applications, the real-time collected transformer electric field data can be input into the trained evaluation model to evaluate the health status of the transformer internal insulation and output the corresponding insulation health index.
[0077] The insulation health index in the embodiment of the present application can reflect the comprehensive capabilities of the electrical equipment (power transformer) in terms of breakdown risk, such as operational reliability, maintainability, and maintenance guarantee performance. The insulation health index can vary between values 0 and 1, and 0 and 1 represent the upper and lower limits of the insulation health index, respectively. The upper and lower limits can be distinguished according to different needs. For example, if 1 is used to represent the ideal value of the insulation health state, the closer the health index is to 1, the better the performance of the equipment (power transformer), and the closer the health index is to 0, the worse the performance of the equipment (power transformer), and there is a higher risk of breakdown; vice versa.
[0078] Furthermore, the corresponding early warning mechanism or measures can be implemented according to the output insulation health index to achieve the purpose of preventive maintenance and risk management.
[0079] In an optional implementation, the above step 103 includes:
[0080] Compare the target insulation health index with the index threshold; when the target insulation health index is lower than the index threshold, trigger the early warning mechanism to provide data-driven maintenance recommendations to support operation and maintenance decisions.
[0081] The above-mentioned index threshold can be set as needed. For example, when the insulation health index is 0-1, the index threshold is set to 0.5. Then, when the target insulation health index is lower than the index threshold of 0.5, an early warning mechanism can be triggered. The early warning mechanism may include outputting various prompt information to indicate the risk of insulation breakdown inside the transformer. In addition, corresponding maintenance suggestions can be provided according to the target insulation health index to support operation and maintenance decisions. Further, for the above-mentioned index threshold, multiple thresholds can be set as needed to determine the health level of the target insulation health index, and corresponding early warnings or measures can be executed according to the health level. The embodiments of the present application do not limit this.
[0082] In an optional implementation, before the above step 101, the above method further includes:
[0083] Obtain transformer electric field data as sample data;
[0084] Input the above sample data into the preset model for training to obtain a preliminary evaluation model;
[0085] The above preliminary evaluation model is used to cross-validate the historical electric field data to obtain the above trained evaluation model.
[0086] Specifically, model training can be completed before actual application. The obtained transformer electric field data can be used as sample data and input into the improved deep learning model (RES-SKIP CNN) for training; the deep learning model with self-learning ability (preliminary evaluation model) after training can be used to cross-validate the historical electric field data to ensure the generalization ability of the model.
[0087] The obtained transformer electric field data can be input into the improved deep learning model (RES-SKIP CNN) for training. The deep learning model with self-learning ability after training is used to cross-validate the historical electric field data to ensure the generalization ability of the model. The trained model can be used for real-time prediction scenarios. The model structure will be introduced in detail later.
[0088] See also Figure 2 , is a flow chart of another transformer internal insulation breakdown risk assessment method provided in an embodiment of the present application, such as Figure 2 As shown, the method includes:
[0089] 201. Determine the layout plan of the electric field sensor based on the theoretical distribution of the electric field inside the transformer;
[0090] 202. Use synchronization technology and data encryption technology to obtain real-time collected transformer electric field data;
[0091] 203. Input the transformer electric field data into a trained evaluation model, evaluate the health status of the transformer internal insulation through the trained evaluation model, and output a target insulation health index;
[0092] 204. Trigger an early warning mechanism based on the above target insulation health index.
[0093] The method in the embodiments of the present application is mainly to improve the electric field measurement scheme and the construction of the deep learning model.
[0094] Specifically, in the embodiments of the present application, an optimized electric field sensor layout scheme can be designed based on the theoretical distribution of the electric field inside the transformer to ensure comprehensive monitoring of key insulation parts.
[0095] Based on the layout plan, data collection and analysis are carried out. In the data collection process, synchronization technology is used to ensure the synchronization and accuracy of data collection from multiple sensors, and high-order filtering technology can be used to remove noise components in the data. In addition, data encryption technology can be used to ensure the security and integrity of data transmission from sensors to data processing centers.
[0096] Among them, the above steps 202 to 204 can refer to Figure 1 The specific description of the illustrated embodiments will not be repeated here.
[0097] In an optional implementation, the above step 201 includes but is not limited to the following steps:
[0098] Spatial distribution analysis, identification of key monitoring points, determination of sensor quantity and type, optimal layout strategy, installation and calibration, testing and adjustment.
[0099] Optionally, the above spatial distribution analysis specifically includes:
[0100] By building a three-dimensional model inside the transformer and simulating the electric field distribution, the area where the electric field intensity changes most significantly is determined;
[0101] The identification of the above key monitoring points specifically includes:
[0102] Based on the simulation results of the electric field distribution, the key area with the largest electric field intensity gradient is identified;
[0103] The number and type of the above sensors are determined by:
[0104] Determine the number and type of sensors required based on monitoring requirements and the specific size and configuration of the transformer;
[0105] The above optimal layout strategy specifically includes:
[0106] An optimization algorithm is used to determine the best placement of the above-mentioned sensors, which aims to cover all key monitoring points and maximize overall monitoring efficiency and accuracy.
[0107] In the embodiments of the present application, the improvements for electric field measurement are introduced as follows:
[0108] A design scheme for improving sensor layout is proposed. This scheme combines the internal structural characteristics of the transformer and the electric field distribution law to ensure comprehensive and accurate monitoring of the internal insulation status of the transformer. Specifically, it includes:
[0109] Spatial distribution analysis: First, a detailed analysis of the internal space and electric field distribution of the transformer is conducted. By building a three-dimensional model of the transformer and simulating the electric field distribution, the area with the most significant change in electric field intensity is determined;
[0110] Identification of key monitoring points: Based on the electric field distribution simulation results, identify the key areas with the largest electric field intensity gradient. These areas are usually the locations where insulation aging or damage is most likely to occur first;
[0111] Determination of the number and type of sensors: Determine the number and type of sensors required based on the monitoring requirements and the specific size and configuration of the transformer. This can mainly be done by selecting highly sensitive sensors that can accurately measure small-scale electric field changes;
[0112] Optimal layout strategy: An optimization algorithm, such as genetic algorithm or particle swarm optimization, is used to determine the best placement of sensors. The algorithm aims to cover all key monitoring points and maximize overall monitoring efficiency and accuracy.
[0113] Further, optionally, the above layout scheme may also be considered:
[0114] Installation and calibration: Install the sensor at the optimal location and perform on-site calibration to adapt to the actual electric field environment; the calibration process ensures accurate correspondence between the sensor output and the actual electric field strength;
[0115] Test and adjust: Perform preliminary tests after installation to evaluate whether the performance of the entire sensor network meets expectations. Adjust sensor locations or parameter settings based on test results to optimize monitoring results.
[0116] In an optional implementation, the preset model adopts a convolutional neural network structure optimized by a residual learning mechanism;
[0117] The specific structure of the above preset model includes:
[0118] Input layer, first convolutional layer, first residual block, second residual block, third residual block, fourth residual block, fully connected layer and output layer;
[0119] Each grid in the above input layer represents the electric field intensity at a measurement point.
[0120] In a specific implementation, the network structure may refer to the following description:
[0121] Input Layer:
[0122] Input->256x256 grid (each grid represents the electric field strength of a measurement point)
[0123] First convolutional layer:
[0124] Conv2D->64filters,3x3,ReLU activation
[0125] ->Batch Normalization
[0126] ->MaxPooling2D->2x2
[0127] Residual Block #1:
[0128] ->Conv2D->64 filters, 3x3, ReLU activation
[0129] ->Conv2D->64 filters, 3x3
[0130] +(Skip Connection)
[0131] ->ReLU activation
[0132] ->MaxPooling2D->2x2
[0133] Residual Block #2:
[0134] ->Conv2D->128 filters, 3x3, ReLU activation
[0135] ->Conv2D->128 filters, 3x3
[0136] +(Skip Connection)
[0137] ->ReLU activation
[0138] ->MaxPooling2D->2x2
[0139] Residual Block #3:
[0140] ->Conv2D->256 filters, 3x3, ReLU activation
[0141] ->Conv2D->256 filters, 3x3
[0142] +(Skip Connection)
[0143] ->ReLU activation
[0144] ->MaxPooling2D->2x2
[0145] Residual Block #4:
[0146] ->Conv2D->512filters,3x3,ReLU activation
[0147] ->Conv2D->512filters,3x3
[0148] +(Skip Connection)
[0149] ->ReLU activation
[0150] ->MaxPooling2D->2x2
[0151] Fully connected layer:
[0152] ->Flatten
[0153] ->Dense->1024neurons,ReLU activation
[0154] Output layer:
[0155] ->Dense->1neuron, Sigmoid activation (output electric field health index, between 0 and 1)
[0156] The above is only an example of a specific network structure. Based on the present application, it can be adjusted as needed, and the embodiments of the present application do not limit this.
[0157] In terms of deep learning models, in the embodiments of the present application, a convolutional neural network structure (RES-SKIP CNN) optimized by a residual learning mechanism is used to solve the problem of gradient disappearance in deep network training, improve the network training depth and performance, and introduce skip connections in the network. The input of each residual block is not only passed to the next layer, but also added to the output of this block. This method allows the gradient to flow directly to deeper network layers, thereby increasing the network depth without adding additional parameters, improving training efficiency and results. Through this structure and optimization method, the network can more effectively learn and extract important features in the electric field data, thereby more accurately evaluating the health of the transformer's internal insulation.
[0158] The method in the embodiment of the present application adopts a more accurate and practical electric field measurement method and combines it with a deep learning method for risk assessment. By optimizing the layout of electric field sensors, the electric field data is collected with high precision and encrypted for transmission, and then the acquired electric field data is input into a deep learning model designed and improved for the internal insulation breakdown data of the transformer for training and verification, and real-time risk assessment is achieved.
[0159] The method in the embodiment of the present application aims to achieve continuous monitoring and fault prediction of transformer insulation performance through advanced data processing and analysis technology, so as to improve the operating efficiency and reliability of the power system. The method is mainly applicable to non-intrusive online monitoring of power transformers, emphasizing the analysis of electric field data through advanced algorithms without interrupting the normal operation of the equipment, so as to achieve the purpose of preventive maintenance and risk management.
[0160] Based on the description of the aforementioned method embodiment, the embodiment of the present application further discloses a transformer internal insulation breakdown risk assessment device.
[0161] Figure 3 A schematic diagram of a transformer internal insulation breakdown risk assessment device provided in an embodiment of the present application is shown in FIG. Figure 3 As shown, the transformer internal insulation breakdown risk assessment device 300 may include:
[0162] The data acquisition module 310 is used to acquire the transformer electric field data collected in real time;
[0163] An evaluation module 320 is used to input the above transformer electric field data into a trained evaluation model, evaluate the health status of the transformer internal insulation through the above trained evaluation model, and output a target insulation health index;
[0164] The early warning module 330 is used to trigger an early warning mechanism according to the target insulation health index.
[0165] Optionally, the apparatus 300 further includes a training module 340, which is used to:
[0166] Before obtaining the transformer electric field data collected in real time, obtaining the transformer electric field data as sample data;
[0167] Input the above sample data into the preset model for training to obtain a preliminary evaluation model;
[0168] The above preliminary evaluation model is used to cross-validate the historical electric field data to obtain the above trained evaluation model.
[0169] Optionally, the apparatus 300 further includes a layout module 350, which is used to:
[0170] Before obtaining the transformer electric field data as sample data, the electric field sensor layout scheme is determined based on the theoretical distribution of the electric field inside the transformer.
[0171] Optionally, the layout module 350 is specifically configured to perform the following steps:
[0172] Spatial distribution analysis, identification of key monitoring points, determination of sensor quantity and type, optimal layout strategy, installation and calibration, testing and adjustment.
[0173] Optionally, the layout module 350 is specifically used for:
[0174] By building a three-dimensional model inside the transformer and simulating the electric field distribution, the area where the electric field intensity changes most significantly is determined;
[0175] Based on the simulation results of the electric field distribution, the key area with the largest electric field intensity gradient is identified;
[0176] Determine the number and type of sensors required based on monitoring requirements and the specific size and configuration of the transformer;
[0177] An optimization algorithm is used to determine the best placement of the above-mentioned sensors, which aims to cover all key monitoring points and maximize overall monitoring efficiency and accuracy.
[0178] Optionally, the above preset model adopts a convolutional neural network structure optimized by a residual learning mechanism;
[0179] The specific structure of the above preset model includes:
[0180] Input layer, first convolutional layer, first residual block, second residual block, third residual block, fourth residual block, fully connected layer and output layer;
[0181] Each grid in the above input layer represents the electric field intensity at a measurement point.
[0182] Optionally, the warning module 330 is specifically used for:
[0183] Compare the target insulation health index with the index threshold; when the target insulation health index is lower than the index threshold, trigger the early warning mechanism to provide data-driven maintenance recommendations to support operation and maintenance decisions.
[0184] It can be understood that the relevant contents of each module in the above device have been described in detail in the aforementioned method embodiment, and the details can be referred to the contents in the method embodiment; that is, the transformer internal insulation breakdown risk assessment device provided in the present application can be executed as follows: Figure 1 or Figure 2 Any steps in the illustrated embodiment will not be described in detail here.
[0185] In one embodiment of the present application, an electronic device is also provided. The electronic device may include a processor and a memory, wherein the memory stores a computer program, and when the computer program is executed by the processor, the execution Figure 1 or Figure 2Any step in the method embodiment shown can execute the relevant construction of the model and can also execute the application calculation of the model. The electronic device can also include an input / output device, etc. In a specific implementation, the electronic device can be a server, a terminal device, etc.
[0186] In one embodiment, a computer-readable storage medium is further provided. The computer-readable storage medium stores a computer program. When the computer program is executed by a processor, the processor executes any step in the above method embodiment.
[0187] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the program can be stored in a non-volatile computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).
[0188] The technical features of the above embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.
[0189] The above-mentioned embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.
Claims
1. A method for assessing the risk of internal insulation breakdown of a transformer. It is characterized in that The method comprises: Obtain real-time collected transformer electric field data; Inputting the transformer electric field data into a trained evaluation model, evaluating the health status of the transformer internal insulation through the trained evaluation model, and outputting a target insulation health index; An early warning mechanism is triggered according to the target insulation health index.
2. The transformer internal insulation breakdown risk assessment method according to claim 1, It is characterized in that Before acquiring the transformer electric field data collected in real time, the method further includes: Obtain transformer electric field data as sample data; Inputting the sample data into a preset model for training to obtain a preliminary evaluation model; The preliminary evaluation model is used to cross-validate historical electric field data to obtain the trained evaluation model.
3. The transformer internal insulation breakdown risk assessment method according to claim 2, It is characterized in that Before obtaining the transformer electric field data as sample data, the method further includes: The layout plan of the electric field sensors is determined based on the theoretical distribution of the electric field inside the transformer.
4. The transformer internal insulation breakdown risk assessment method according to claim 3, It is characterized in that Determining the layout of the electric field sensors according to the theoretical distribution of the electric field inside the transformer includes but is not limited to the following steps: Spatial distribution analysis, identification of key monitoring points, determination of sensor quantity and type, optimal layout strategy, installation and calibration, testing and adjustment.
5. The transformer internal insulation breakdown risk assessment method according to claim 4, It is characterized in that The spatial distribution analysis specifically includes: By building a three-dimensional model inside the transformer and simulating the electric field distribution, the area where the electric field intensity changes most significantly is determined; The identification of key monitoring points specifically includes: Based on the simulation results of the electric field distribution, the key area with the largest electric field intensity gradient is identified; The number and type of sensors are determined, including: Determine the number and type of sensors required based on monitoring requirements and the specific size and configuration of the transformer; The optimal layout strategy specifically includes: An optimization algorithm is used to determine the best placement of the required sensors, wherein the optimization algorithm aims to cover all key monitoring points and maximize overall monitoring efficiency and accuracy.
6. The transformer internal insulation breakdown risk assessment method according to claim 2, It is characterized in that The preset model adopts a convolutional neural network structure optimized by a residual learning mechanism; The specific structure of the preset model includes: Input layer, first convolutional layer, first residual block, second residual block, third residual block, fourth residual block, fully connected layer and output layer; Each grid of the input layer represents the electric field strength at a measurement point.
7. The transformer internal insulation breakdown risk assessment method according to claim 1, It is characterized in that The triggering of the early warning mechanism according to the target insulation health index includes: The target insulation health index is compared with an index threshold; when the target insulation health index is lower than the index threshold, the early warning mechanism is triggered to provide data-driven maintenance suggestions to support operation and maintenance decisions.
8. A transformer internal insulation breakdown risk assessment device, It is characterized in that include: A data acquisition module, used to acquire transformer electric field data collected in real time; An evaluation module, used to input the transformer electric field data into a trained evaluation model, evaluate the health status of the transformer internal insulation through the trained evaluation model, and output a target insulation health index; The early warning module is used to trigger an early warning mechanism according to the target insulation health index.
9. An electronic device, It is characterized in that The method comprises a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of the method according to any one of claims 1 to 7.
10. A computer-readable storage medium storing a computer program, It is characterized in that When the computer program is executed by a processor, the processor is caused to perform the steps of the method according to any one of claims 1 to 7.
