Power grid digital twin fault simulation equipment reconstructed from historical data

Abstract

The present application provides a power grid digital twin fault simulation device that reconstructs historical data, and relates to the field of power digital twin technology, including: a fault feature acquisition unit that collects power grid fault feature information when a target power grid fails; a historical data retrieval unit that indexes historical power grid faults based on the power grid fault feature information, and obtains historical power grid faults of the same family and historical power grid fault operation data sets when the same family of historical power grid faults occurs; a historical data reconstruction unit that obtains device maintenance data of multiple power grid devices after the same family of historical power grid faults, reconstructs and generates historical data, and obtains a reconstructed power grid fault operation data set; a digital twin simulation unit that uses the reconstructed power grid fault operation data set to construct a power grid fault twin simulation result. The present application solves the technical problem that the existing power grid fault diagnosis method lacks effective use of historical data, resulting in limited accuracy of fault diagnosis, and significantly improves the accuracy of fault diagnosis and the foresight of fault prevention.

Description

Technical Field

[0001] The present application relates to the field of power digital twin technology, and specifically to a power grid digital twin fault simulation device that reconstructs historical data. Background Art

[0002] Grid fault diagnosis is a crucial component of power system operation, directly impacting grid security, stability, and power supply reliability. As grids expand in size and become more complex, the types and conditions of grid faults are becoming more diverse and complex. Accurate and rapid fault diagnosis has become a key focus of power system research.

[0003] Currently, power grid fault diagnosis primarily relies on real-time monitoring data and analysis of equipment status. This typically involves collecting grid operating data in real time through sensors and monitoring equipment, monitoring the operating status of grid equipment, and locating faults using fault detection algorithms, models, and rules. However, existing fault diagnosis methods have several significant shortcomings. First, fault diagnosis largely relies on real-time data. This means that for certain fault types (such as transient or hidden faults), real-time data may not accurately capture the full picture. Second, given the diverse nature of power grid equipment and the complex fault characteristics, a single monitoring method cannot fully cover all fault modes. Consequently, existing methods often suffer from issues such as incomplete fault feature identification, insufficient fault simulation, and high misdiagnosis rates. Summary of the Invention

[0004] This application provides a digital twin fault simulation device for power grids that reconstructs historical data, which solves the technical problem that existing power grid fault diagnosis methods lack effective utilization of historical data, resulting in limited accuracy of fault diagnosis, and achieves the technical effect of significantly improving the accuracy of fault diagnosis and the foresight of fault prevention.

[0005] In view of the above problems, the present application provides a power grid digital twin fault simulation device for historical data reconstruction, the device comprising: a fault feature acquisition unit, the fault feature acquisition unit being used to collect and obtain power grid fault feature information when a fault is detected in a target power grid; a historical data retrieval unit, the historical data retrieval unit being used to index historical power grid faults according to the power grid fault feature information, obtain historical power grid faults of the same family, and index and obtain a historical power grid fault operation data set when the historical power grid fault of the same family occurs, wherein the historical power grid fault operation data set includes multiple historical fault operation data of multiple power grid devices in the target power grid; a historical data reconstruction unit, the historical data reconstruction unit being used to obtain multiple device maintenance data of the multiple power grid devices in the target power grid after the historical power grid fault of the same family, perform historical data reconstruction on the historical power grid fault operation data set to generate a reconstructed power grid fault operation data set; a digital twin simulation unit, the digital twin simulation unit being used to adopt the reconstructed power grid fault operation data set to construct a power grid fault twin simulation result based on a digital twin.

[0006] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0007] When a fault is detected in the target power grid, the fault feature acquisition unit collects and obtains grid fault feature information. The historical data retrieval unit indexes historical grid faults based on the grid fault feature information, obtains historical grid faults of the same family, and then indexes and obtains a historical grid fault operation dataset at the time the historical grid fault occurred. The fault feature acquisition unit and the historical data retrieval unit combine the real-time feature data at the time of the grid fault with historical data, and retrieves the historical grid fault dataset using a feature indexing method. By combining historical data of similar faults, a more comprehensive and accurate understanding of the fault features can be achieved, providing richer data support for subsequent simulations.

[0008] The historical data reconstruction unit obtains maintenance data for multiple power grid devices in the target power grid after the historical power grid failure in the same family, reconstructs the historical power grid failure operation data set, and generates a reconstructed power grid failure operation data set. The historical data reconstruction unit reconstructs the historical failure operation data to generate a predicted operation data set under the current power grid failure, supplementing missing data and improving the comprehensiveness and accuracy of the fault simulation.

[0009] The digital twin simulation unit adopts the reconstructed power grid fault operation data set and constructs a power grid fault twin simulation result based on the digital twin, simulating a more accurate power grid fault situation, providing real-time simulation results for power grid operation and maintenance personnel, and assisting decision-making.

[0010] In summary, by integrating the aforementioned functional units and combining historical fault data reconstruction with digital twin technology, this application achieves systematic analysis and simulation of power grid faults. This allows for the prediction of power grid behavior under different fault conditions, significantly improving the accuracy of fault diagnosis and the foresight of fault prevention. The application of digital twin technology allows for the testing and optimization of faults in a virtual environment, reducing risks in actual operations and providing a scientific basis for maintenance decisions, thereby improving the efficiency and overall stability of power grid operations.

[0011] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Figure 1 A schematic diagram of the structure of a power grid digital twin fault simulation device that reconstructs historical data provided in an embodiment of the present application.

[0013] Figure 2 A schematic diagram of the process of obtaining a historical power grid fault operation data set in a power grid digital twin fault simulation device for historical data reconstruction provided in an embodiment of the present application.

[0014] Figure 3 A schematic diagram of the process of reconstructing and generating historical data of a historical power grid fault operation data set in a power grid digital twin fault simulation device for historical data reconstruction provided in an embodiment of the present application.

[0015] Explanation of the accompanying symbols: fault feature collection unit 10, historical data retrieval unit 20, historical data reconstruction unit 30, digital twin simulation unit 40. DETAILED DESCRIPTION

[0016] The embodiments of the present application provide a power grid digital twin fault simulation device that reconstructs historical data, and utilizes historical data reconstruction and digital twin technology to simulate power grid faults, thereby solving the technical problem that the existing power grid fault diagnosis methods lack effective utilization of historical data, resulting in limited accuracy of fault diagnosis, and achieving the technical effect of significantly improving the accuracy of fault diagnosis and the foresight of fault prevention.

[0017] like Figure 1 As shown, an embodiment of the present application provides a power grid digital twin fault simulation device for historical data reconstruction, the device comprising:

[0018] The fault feature acquisition unit 10 is used to acquire grid fault feature information when a fault occurs in the target grid.

[0019] Specifically, the target grid refers to the power system or network being monitored and experiencing a fault. This can be a specific distribution network, substation, or a broader power transmission network. Grid fault signature information refers to the key characteristic data of a grid fault, including the fault location and fault parameters.

[0020] During grid operation, monitoring devices continuously collect grid operating parameters. For example, current and voltage sensors and energy meters installed in substations monitor grid load changes and power quality in real time. Temperature and vibration sensors deployed on transmission lines detect early signs of overheating or mechanical damage. When a fault occurs in the target grid, the fault signature acquisition unit 10 connects to sensors, smart meters, protection devices, and other acquisition devices in the grid to acquire real-time characteristic data related to the fault, including the fault location and fault parameters, such as the faulted device and operating parameters such as current, voltage, power, and frequency. This characteristic data can reveal the specific circumstances of the grid fault and provide reliable data support for subsequent fault location and simulation.

[0021] A historical data retrieval unit 20 is used to index historical power grid faults based on the power grid fault characteristic information, obtain historical power grid faults of the same family, and index to obtain a historical power grid fault operation data set when the historical power grid faults of the same family occurred, wherein the historical power grid fault operation data set includes multiple historical fault operation data of multiple power grid devices in the target power grid.

[0022] Specifically, historical grid faults of the same family refer to historical faults with similar characteristics or belonging to the same category as the current grid fault. A historical grid fault operation dataset is a collection of operational data from multiple historical grid faults. This dataset includes information such as the operating status, performance parameters, and response measures of multiple grid devices during the fault period. Grid devices refer to various equipment and devices in the grid system, such as transformers, switches, circuit breakers, distribution lines, and protective equipment.

[0023] The historical data retrieval unit 20 receives the grid fault feature information collected by the fault feature collection unit 10, processes these fault features, and extracts key data points, such as current and voltage fluctuations. Next, based on the current grid fault features, it compares them with the historical fault data stored in the grid database to find historical fault cases with high similarity, and marks them as historical grid faults in the same group. During the query, a match is performed based on multiple dimensions in the fault feature information, such as the fault type, location, and grid load at the time, to ensure that the historical grid faults of the same family with high similarity to the current grid fault are found. Then, the operating status and behavior of each grid device during the fault period when the historical grid faults of the same family occur are extracted and summarized to generate a historical grid fault operation data set. These data sets can include information such as temperature changes of transformers, operation records of circuit breakers, operating current, voltage, power, and the fluctuation amplitude of these parameters.

[0024] For example, in a city's power grid system, a short circuit fault is detected in a distribution network in a certain area. The fault characteristic captured by the fault feature acquisition unit 10 is a momentary overcurrent on a specific line. The historical data retrieval unit 20 searches the database for historical fault records of previous instances of the same momentary overcurrent on the line. If a historical fault record exists in the database for a short circuit on the same line due to a tree branch, then this is considered a historical power grid fault of the same family. The historical fault operation data of the transformer, distribution box, and other devices on the line at that time, such as the transformer's oil temperature and the distribution box's voltage fluctuations, are then indexed and used as a historical power grid fault operation dataset.

[0025] The historical data retrieval unit 20 can efficiently retrieve historical fault data sets similar to the current fault from the historical fault database based on the fault characteristic information collected in real time, providing data support for subsequent fault simulation and diagnosis. Especially when facing complex fault scenarios, through the comparison and analysis of historical data, it can help maintenance personnel better understand the causes and consequences of the fault, and thus make more accurate response measures.

[0026] The historical data reconstruction unit 30 is used to obtain the maintenance data of multiple power grid devices of the target power grid after the historical power grid failure of the same family, reconstruct the historical data of the historical power grid failure operation data set, and obtain a reconstructed power grid failure operation data set.

[0027] Specifically, grid device maintenance data refers to the data recorded during maintenance operations on multiple grid devices in the target power grid after a historical grid fault of the same family occurs. For example, when repairing a faulty transformer, this data includes the repair details, information about replaced components, and post-repair inspection data. This data can reflect the post-fault repair status and post-repair status of the device. The reconstructed grid fault operation dataset is based on the original historical grid fault operation dataset and combines it with device maintenance data to reconstruct the grid operation dataset at the time of the current fault.

[0028] First, multiple device maintenance data for multiple grid devices in the target power grid following a historical grid failure of the same family is obtained from relevant data sources. These data sources can include databases of power grid equipment maintenance management systems. By analyzing the maintenance data from multiple devices in the grid, the actual performance of the devices at the time of the failure can be inferred. For example, if a transformer had been recently overhauled and its maintenance records indicate good health, the reconstructed data may show that the device has a relatively fast recovery capability.

[0029] The historical data reconstruction unit 30 infers and supplements data that cannot be directly measured or is lost based on known historical fault data and device maintenance data. This reconstructed data may include actual device performance (e.g., whether the repaired device can fully recover to normal operation), response time at the time of the fault, and current and voltage changes during the recovery process, generating a reconstructed grid fault operation dataset. This reconstructed grid fault operation dataset can more comprehensively reflect the state changes of grid devices throughout the entire process from the occurrence of the current grid fault to the repair of the fault. This provides a high-quality data source for subsequent digital twin simulations, enabling more accurate simulation of the current grid fault's performance and recovery process, helping grid operations and maintenance personnel develop more scientific emergency response strategies.

[0030] The digital twin simulation unit 40 is used to use the reconstructed power grid fault operation data set to construct a power grid fault twin simulation result based on the digital twin.

[0031] Specifically, the digital twin simulation unit 40 uses digital twin technology to construct a virtual grid model based on information such as the grid topology and equipment parameters. This model not only includes the physical structure of the grid (such as power transmission lines, substation locations, and transformers), but also takes into account the dynamic performance, operating status, and failure modes of various equipment. For example, factors such as transformer operating characteristics and circuit breaker triggering delay are incorporated into the model.

[0032] The reconstructed grid fault operation dataset generated by the historical data reconstruction unit 30 is obtained. This dataset includes information such as the behavior data, performance parameters, and response time of each device in the target grid (such as transformers, circuit breakers, and cables) under the current fault condition. Next, utilizing the characteristics of digital twins, namely, bidirectional data interaction and real-time updates, the reconstructed grid fault operation dataset is input into the grid digital model to simulate the response process of the target grid after the current fault occurs, such as current fluctuations, voltage drops, and equipment switching. Ultimately, a grid fault twin simulation result is constructed. This result is a simulation of the grid fault constructed using digital twin technology based on the reconstructed grid fault operation dataset. It can intuitively demonstrate the various characteristics of the current grid fault, including the various states of the target grid when the fault occurs, such as the location of the fault point, the impact of the fault on the grid voltage and current distribution, and the path of fault propagation.

[0033] By combining real-time data with historical fault data, digital twin models can accurately reflect the dynamic behavior and response of the power grid during a fault. For example, by simulating an overload fault, the digital twin model can demonstrate key milestones in the grid's recovery process, such as equipment recovery time and load adjustments. These simulation results can help grid operators understand the timing and consequences of faults and predict the grid's recovery process, thereby optimizing grid operation and maintenance strategies.

[0034] Furthermore, the fault feature collection unit 10 of the embodiment of the present application is further configured to perform the following steps:

[0035] When a fault occurs in the target power grid, fault location information and fault parameters are collected; and the fault location information and fault parameters are integrated to obtain power grid fault characteristic information.

[0036] Specifically, the fault signature acquisition unit 10 uses two key data points: fault location information and fault parameters to acquire complete grid fault signature information. When a fault occurs in the target grid, sensors and monitoring equipment within the grid detect the fault in real time. These monitoring devices, including current sensors, voltage sensors, and fault indicators, determine the presence of a fault by detecting abnormal changes in parameters such as current and voltage.

[0037] Fault location information is typically obtained through sensor networks or fault location equipment. Precise location devices are installed throughout the power grid (such as transformers, circuit breakers, and transmission lines). These devices can upload real-time information about the fault location via the communication network. For example, in the event of a short circuit, a circuit breaker immediately sends a signal to the system, indicating the specific grid segment or location where the fault occurred. When a fault occurs, the grid's sensor network captures changes in operating parameters, monitoring current, voltage, power, and other operating parameters in real time and recording the relevant data. For example, a short circuit causes a sudden increase in current and a sudden drop in voltage; while equipment overload causes a sustained increase in current, it does not produce the dramatic fluctuations that a short circuit would.

[0038] The fault feature acquisition unit 10 interacts with sensors and monitoring equipment in the target power grid to obtain fault location information and fault parameters when the fault occurs, and uses a data fusion algorithm to integrate the fault location information and fault parameters to generate power grid fault feature information, providing a basis for subsequent fault analysis and simulation.

[0039] Further, such as Figure 2 As shown, the historical data retrieval unit 20 in the embodiment of the present application is further configured to perform the following steps:

[0040] Step P21: Acquire historical fault record data of the target power grid within a preset historical time range.

[0041] Step P22: Based on the power grid fault characteristic information, perform traversal indexing in the historical fault record data to obtain the historical power grid fault characteristic information with the greatest similarity as the same family of historical power grid faults.

[0042] Step P23: Retrieve a plurality of historical fault operation data of a plurality of power grid devices in the target power grid when the same family historical power grid fault occurs, and obtain the historical power grid fault operation data set.

[0043] Specifically, the historical data retrieval unit 20 retrieves historical fault records for the target power grid from the power grid database based on a preset historical time range. This data may include information such as the time, location, equipment, fault type, duration, and affected power grid components. This historical record provides an important reference for analyzing the current fault.

[0044] Next, the currently collected grid fault feature information is used to traverse and index the historical fault record data. During the traversal process, similarity matching algorithms such as Euclidean distance and cosine similarity are used to calculate the similarity between historical fault features and the current fault features. This quantitatively evaluates the similarity between different historical faults and finds the historical fault with the most similar features to the current fault, which is considered as the historical grid fault of the same family.

[0045] After identifying historical grid faults of the same family, the historical fault operation data of the relevant grid devices in the target grid when these faults occurred are retrieved, including data such as current, voltage, temperature, switch status, and repair time of each grid device. A historical grid fault operation dataset is generated, providing rich reference data for the simulation and analysis of the current fault, thereby improving the accuracy of fault simulation.

[0046] Furthermore, the historical data retrieval unit 20 of the embodiment of the present application is further configured to perform the following steps:

[0047] Step P221: Based on the power grid fault characteristic information, perform traversal indexing in the historical fault record data, calculate the fault location similarity between the fault location information in the power grid fault characteristic information and multiple historical fault location information, and calculate the fault parameter similarity between the fault parameters in the power grid fault characteristic information and multiple historical fault parameters.

[0048] Step P222: Perform weighted calculation on the fault location similarity and fault parameter similarity of each historical power grid fault feature information to obtain multiple historical power grid fault similarities.

[0049] Step P223: Select the historical power grid fault feature information with the greatest similarity to the historical power grid fault, and output it to obtain the historical power grid faults of the same family.

[0050] Specifically, the historical data retrieval unit 20 determines the same family of historical power grid faults by calculating the similarity between the current power grid fault feature information and the fault location and fault parameter similarity in the historical fault record data.

[0051] The historical data retrieval unit 20 first searches for the location information of all faults in the historical fault record data based on the fault location information of the current power grid fault, including the power grid equipment, location, and area where the fault occurred. Using methods such as Euclidean distance or cosine similarity, the location information of the current fault is compared with multiple locations in the historical fault record to calculate the similarity of the fault locations. For example, the current fault location coordinates are (x1, y1), and the historical fault location coordinates are (x2, y2). Using the Euclidean distance formula , and calculate the similarity between the two based on their relative positions.

[0052] In addition to location information, the unit also calculates the similarity between the electrical parameters of the current fault (such as current and voltage fluctuations) and the electrical parameters in the historical records. Usually, parameters such as current and voltage are described by statistical methods such as standard deviation, peak value or mean, and then the current fault parameters and historical fault parameters are calculated. Taking voltage as an example, the current fault voltage is V1 and the historical fault voltage is V2. A simple proportional difference method can be used to calculate the similarity. , assuming For multiple fault parameters, their similarities can be calculated separately and then considered comprehensively.

[0053] After obtaining the location similarity and parameter similarity of each historical fault, these two similarities are weighted according to the importance of different fault characteristics to obtain the comprehensive similarity of each historical power grid fault. For example, if the fault location similarity weight is 0.4 and the fault parameter similarity weight is 0.6, and the fault location similarity of a historical fault is 0.8 and the fault parameter similarity is 0.7, then the historical power grid fault similarity = 0.4 × 0.8 + 0.6 × 0.7 = 0.74.

[0054] After calculating the similarity of all historical grid fault signatures, the fault with the highest similarity is selected as the same-family historical grid fault. This is because the characteristics of the same-family historical grid fault are most similar to the current fault, and the relevant processing experience and data are of high reference value for the current grid fault.

[0055] Further, such as Figure 3 As shown, the historical data reconstruction unit 30 in the embodiment of the present application is further configured to perform the following steps:

[0056] Step P31: Acquire maintenance data records of the plurality of power grid devices in the target power grid.

[0057] Step P32: Retrieve, from the maintenance data record, multiple device maintenance data of the multiple power grid devices within the time period after the historical power grid failure of the same family.

[0058] Step P33: Reconstructing and generating historical data of the historical power grid fault operation data set based on the plurality of device maintenance data.

[0059] Specifically, maintenance data records refer to data related to the maintenance, inspection, and upkeep of power grid equipment. These records include equipment inspection time, fault history, repair details, and equipment status, providing detailed information on the health of the equipment and the repair process after a fault occurs. Maintenance data records are typically stored in the device management database within the power grid management system. The historical data reconstruction unit 30 accesses the maintenance data records of each power grid device in the target power grid through an interface, including information related to each device's regular inspection, maintenance, and fault repair.

[0060] After acquiring complete maintenance data records, the historical data reconstruction unit 30 retrieves grid device maintenance data for a certain period of time after the fault, based on the time range of the historical grid fault. This data is closely related to previous historical grid faults of the same family and may contain important information such as maintenance measures taken for that type of fault and changes in device status after repair. For example, if the historical grid fault of the same family is a short circuit fault at a substation, the device maintenance data after the fault occurs will include maintenance and adjustment data for transformers, circuit breakers, and other devices within the substation.

[0061] The historical data of the power grid fault operation dataset is reconstructed based on the maintenance data of multiple devices selected. The reconstructed dataset can reflect the actual performance of the equipment after the current power grid fault, providing a more accurate reference for fault analysis and simulation.

[0062] Furthermore, the historical data reconstruction unit 30 in the embodiment of the present application is further configured to perform the following steps:

[0063] Step P331: Based on the generative adversarial network, train the power grid operation data reconstruction channel.

[0064] Step P332: Input the plurality of device maintenance data and the historical power grid fault operation data set into the power grid operation data reconstruction channel to generate a reconstructed power grid fault operation data set.

[0065] Specifically, the historical data reconstruction process first requires training a power grid operation data reconstruction channel using a generative adversarial network (GAN). This GAN consists of a generator and a discriminator. The generator is responsible for generating reconstructions that are as close to the actual fault data as possible based on historical fault operation data and device maintenance data. The discriminator distinguishes whether the input data comes from a real dataset or is fake data generated by the generator. The discriminator continuously optimizes the generator to improve the authenticity of the data. Through repeated training, the generator and the discriminator continuously engage in adversarial optimization. The generator gradually learns how to generate more realistic power grid fault data, while the discriminator becomes more discerning, accurately distinguishing between real and generated data. Ultimately, the trained GAN outputs the power grid operation data reconstruction channel. This power grid operation data reconstruction channel can be trained using historical power grid fault operation data and device maintenance data to generate more accurate power grid fault operation data.

[0066] The maintenance data of multiple power grid devices and historical power grid fault data are input into a trained power grid operation data reconstruction channel. This channel corrects the behavior of the power grid devices during faults, supplements missing data, and adjusts the recovery of the equipment after the fault based on the maintenance records. This makes the data set more consistent with the actual operation status of the power grid after the fault, and outputs a reconstructed power grid fault operation data set to provide more reliable data support for subsequent power grid fault simulations.

[0067] Furthermore, the historical data reconstruction unit 30 in the embodiment of the present application is further configured to perform the following steps:

[0068] Step P331-1: Based on the fault operation record data of the same power grid, collect multiple sample historical power grid fault operation data sets and multiple sample device maintenance data sets, and obtain power grid fault operation data sets after the same power grid fault occurs under different sample device maintenance data sets, as multiple sample reconstructed power grid fault operation data sets.

[0069] Step P331-2: Based on the generative adversarial network, construct a generator and a discriminator in the power grid operation data reconstruction channel.

[0070] Step P331-3: Use the multiple sample historical power grid fault operation data sets and the multiple sample device maintenance data sets to perform generation supervision training on the generator, use the multiple samples to reconstruct the power grid fault operation data sets, and combine the generation results of the generator to perform discriminant supervision training on the discriminator.

[0071] Step P331-4: Perform supervised training until convergence requirements are met to obtain a grid operation data reconstruction channel.

[0072] Specifically, the historical data reconstruction unit 30 trains the power grid operation data reconstruction channel based on the generative adversarial network in the following process:

[0073] Fault operation records for power grids with the same specifications and configuration as the target power grid are obtained. From this data, grid status information at the time of the fault, as well as post-fault device maintenance information, is extracted for different grid faults. This generates multiple sample historical grid fault operation datasets and multiple sample device maintenance datasets. Furthermore, grid fault operation datasets are extracted for the recurrence of the same grid fault after device maintenance according to the different sample device maintenance datasets, generating multiple sample reconstructed grid fault operation datasets. This sample data provides diverse input for subsequent generative adversarial network training, covering various scenarios and device maintenance situations associated with grid faults.

[0074] A generative adversarial network model is constructed using a deep learning framework. The model includes a generator and a discriminator. The generator is responsible for generating false data based on the input historical power grid fault operation data and device maintenance data, that is, predicting the operating status of the power grid when the fault occurs. The discriminator is responsible for evaluating whether the generated false data is true or consistent with the actual power grid fault operation data set. Exemplarily, the generator can be a model composed of a multi-layer neural network, which receives a sample historical power grid fault operation data set and a sample device maintenance data set as input, and outputs a reconstructed power grid fault operation data set after a series of neuron calculations. The discriminator can also be a neural network, which receives a sample reconstructed power grid fault operation data set and the result generated by the generator as input, extracts and judges the input data, and outputs a probability value representing the authenticity of the data.

[0075] During the training process, the generator undergoes generative supervision training using multiple sample historical power grid fault operation datasets and sample device maintenance datasets. The generator generates prediction results (i.e., fake data) based on the input sample data. After a preset training step, the generator outputs the predicted reconstructed power grid fault operation dataset. Then, multiple samples are used to reconstruct the power grid fault operation dataset, and the discriminator is trained based on the generator's results. The discriminator learns to distinguish between the real sample reconstructed power grid fault operation dataset and the results generated by the generator, adjusting its parameters based on the accuracy of its judgment to improve its discrimination ability. After a preset training step, the discriminator outputs the discrimination results. The generator adjusts its parameters based on the discrimination results and proceeds to the next round of training. The generator and discriminator are trained alternately over multiple rounds, with their parameters continuously adjusted until convergence is met, meaning that the model's loss function reaches a minimum or stops decreasing significantly. At this point, the data generated by the generator is sufficiently realistic and can reconstruct the power grid fault operation data well, making it difficult for the discriminator to distinguish between real and fake data. Output the trained generator and discriminator to obtain the power grid operation data reconstruction channel.

[0076] Furthermore, the digital twin simulation unit 40 of the embodiment of the present application is further configured to perform the following steps:

[0077] Step P41: using the multiple reconstructed fault operation data of the multiple power grid devices in the reconstructed power grid fault operation data set to identify the multiple power grid devices.

[0078] Step P42: Based on the digital twin and according to the identification results, a power grid fault twin model is constructed as a power grid fault twin simulation result.

[0079] Specifically, the reconstructed fault operation data of multiple power grid devices in the reconstructed power grid fault operation dataset is used to identify the multiple power grid devices and determine the specific attributes or status of each power grid device during the fault operation. For example, for a transformer in a substation, the reconstructed fault operation data includes information such as the oil temperature and voltage change rate during the fault period. By identifying the transformer based on this data, it is possible to determine the specific status of the transformer at the time of the fault, such as whether there was a risk of overheating or whether the voltage was unstable.

[0080] The identification process first categorizes different grid devices, such as transformers, circuit breakers, switches, and distribution lines, based on their function and type. The operational characteristics of each device are then extracted from the reconstructed grid fault operation dataset. These characteristics typically include: electrical parameters such as voltage, current, power, and frequency; device state changes such as device start-up, shutdown, switching, tripping, and recovery; and time series data such as device response and recovery times.

[0081] Based on device behavior changes, specific fault types, such as short circuits, overloads, ground faults, and equipment failures, are identified. The state changes of each power grid device during a fault are correlated with key information such as the device type, fault occurrence time, and fault type. This creates associated data and corresponding identification tags for each power grid device in a specific fault scenario. For example, if transformer A experiences a short circuit at a certain point in time, transformer B will activate overload protection. This fault information is mapped to each device. Transformer A is identified as "short circuit fault," while device A's status is recorded as "faulty," along with the fault time (start and end times) and relevant electrical parameters. Finally, the identification results are exported to a standardized format (such as a database, JSON, or XML) for subsequent digital twin simulations. This data, including each device's fault characteristics, device status, and operational sequence, makes power grid fault simulations more realistic and reliable, effectively reflecting the actual performance of the power grid under fault conditions.

[0082] After identifying the fault status of each grid device, a corresponding virtual model is constructed in the digital space. This model, known as the grid fault twin model, accurately reflects the state of each device at the time of the fault. For example, for a grid consisting of multiple substations, transmission lines, and distribution network equipment, after identifying the fault status of each grid device, a corresponding virtual model is constructed in the digital space. If a potential fault is detected in a certain area of ​​a transmission line during the identification process, the digital model corresponding to that transmission line in the grid fault twin model will exhibit characteristics related to the potential fault, such as abnormal resistance or interrupted signal transmission.

[0083] This power grid fault twin model, used as a simulation result, provides grid maintenance personnel with an intuitive virtual scenario that closely resembles the actual fault situation, enabling them to conduct fault analysis, develop repair strategies, and conduct emergency drills. For example, they can simulate different repair measures on this digital twin model and observe changes in the state of the grid devices in the model to select the optimal repair solution. The model can also be used to predict the potential chain reactions of a fault to other parts of the grid, allowing them to take preventative measures in advance.

[0084] In summary, the power grid digital twin fault simulation device reconstructed by historical data provided by the embodiments of the present application has the following technical effects:

[0085] When a fault is detected in the target power grid, the fault feature acquisition unit 10 acquires grid fault feature information. The historical data retrieval unit 20 traverses and indexes the historical fault record data based on the grid fault feature information to obtain the historical grid fault feature information with the greatest degree of similarity as the same family of historical grid faults; and retrieves multiple historical fault operation data of multiple grid devices in the target power grid when the same family of historical grid faults occurs to obtain a historical grid fault operation data set. The fault feature acquisition unit 10 and the historical data retrieval unit 20 combine the real-time feature data at the time of the grid fault with the historical data, and use a feature index-based method to retrieve the historical grid fault data set. By combining similar fault historical data, the fault features can be understood more comprehensively and accurately, providing richer data support for subsequent simulations.

[0086] Historical data reconstruction unit 30 obtains maintenance data from multiple power grid devices in the target power grid after the historical power grid failure of the same family. Using a generative adversarial network, it trains a power grid operation data reconstruction channel. This data and the historical power grid failure operation dataset are input into the power grid operation data reconstruction channel to generate a reconstructed power grid failure operation dataset. Historical data reconstruction unit 30 reconstructs the historical failure operation data to generate a predicted operation dataset for the current power grid failure, supplementing missing data and improving the comprehensiveness and accuracy of fault simulation.

[0087] The digital twin simulation unit 40 uses the reconstructed power grid fault operation data set to construct a power grid fault twin simulation result based on the digital twin, simulates a more accurate power grid fault situation, and provides real-time simulation results for power grid operation and maintenance personnel to assist in decision-making.

[0088] Overall, the embodiment of the present application integrates the above-mentioned functional units, combines the reconstruction of historical fault data with digital twin technology, realizes the systematic analysis and simulation of power grid faults, can predict the behavior of the power grid under different fault conditions, and significantly improves the accuracy of fault diagnosis and the foresight of fault prevention. Through accurate fault simulation, it is possible to better predict the changes in the state of the power grid after the fault occurs, formulate countermeasures in advance, reduce the impact of the fault on the operation of the power grid, and ensure the stable operation of the power grid. At the same time, the application of digital twin technology allows faults to be tested and optimized in a virtual environment, providing more targeted strategies for power grid maintenance, reducing risks in actual operations, and thus improving the efficiency and overall stability of power grid operation.

[0089] The above description of the disclosed embodiments is intended to enable one skilled in the art to implement or use the present application. Various modifications to these embodiments will be readily apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application is not limited to the embodiments shown herein, but is intended to conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A digital twin fault simulation device for power grids reconstructed from historical data, characterized by: The device comprises: A fault feature acquisition unit, configured to acquire grid fault feature information when a fault occurs in the target grid; a historical data retrieval unit, configured to index historical power grid faults based on the power grid fault characteristic information, obtain historical power grid faults of the same family, and obtain a historical power grid fault operation data set at the time when the historical power grid faults of the same family occurred, wherein the historical power grid fault operation data set includes multiple historical fault operation data of multiple power grid devices in the target power grid; a historical data reconstruction unit, configured to obtain maintenance data of a plurality of power grid devices of the target power grid after a historical power grid failure of the same family, and to reconstruct the historical power grid failure operation data set to obtain a reconstructed power grid failure operation data set; A digital twin simulation unit, configured to construct a power grid fault twin simulation result based on the digital twin using the reconstructed power grid fault operation data set; The historical data reconstruction unit is further configured to perform the following steps: Obtaining maintenance data records of the plurality of power grid devices in the target power grid; Retrieving, from the maintenance data record, maintenance data of the plurality of power grid devices within a period of time after the historical power grid fault of the same family; Reconstructing and generating historical data of the historical power grid fault operation data set based on the plurality of device maintenance data; The historical data reconstruction unit is further configured to perform the following steps: Based on the generative adversarial network, the grid operation data is trained to reconstruct the channel; Inputting the plurality of device maintenance data and the historical power grid fault operation data set into the power grid operation data reconstruction channel to generate a reconstructed power grid fault operation data set; The historical data reconstruction unit is further configured to perform the following steps: Based on the fault operation record data of the same power grid, multiple sample historical power grid fault operation data sets and multiple sample device maintenance data sets are collected, and the power grid fault operation data sets after the same power grid fault occurs under different sample device maintenance data sets are obtained as multiple sample reconstructed power grid fault operation data sets; Based on a generative adversarial network, a generator and a discriminator are constructed within the power grid operation data reconstruction channel; Using the multiple sample historical power grid fault operation data sets and the multiple sample device maintenance data sets, the generator is trained for generative supervision, the multiple samples are used to reconstruct the power grid fault operation data sets, and the discriminant is trained for discriminative supervision based on the generation results of the generator; Supervised training is performed until convergence requirements are met to obtain the grid operation data reconstruction channel.

2. The power grid digital twin fault simulation device reconstructed by historical data according to claim 1 is characterized in that: The fault feature collection unit is further configured to perform the following steps: When a fault occurs in the target power grid, the fault location information and fault parameters are collected; The fault location information and fault parameters are integrated to obtain power grid fault characteristic information.

3. The power grid digital twin fault simulation device reconstructed by historical data according to claim 1 is characterized in that: The historical data retrieval unit is further configured to perform the following steps: Obtaining historical fault record data of the target power grid within a preset historical time range; Based on the power grid fault characteristic information, performing traversal indexing in the historical fault record data to obtain the historical power grid fault characteristic information with the greatest similarity as the same family of historical power grid faults; A plurality of historical fault operation data of a plurality of power grid devices in the target power grid when the same family historical power grid fault occurs are retrieved to obtain the historical power grid fault operation data set.

4. The power grid digital twin fault simulation device reconstructed by historical data according to claim 3 is characterized in that: The historical data retrieval unit is further configured to perform the following steps: Based on the power grid fault characteristic information, performing traversal indexing in the historical fault record data, calculating the fault location similarity between the fault location information in the power grid fault characteristic information and multiple historical fault location information, and calculating the fault parameter similarity between the fault parameter in the power grid fault characteristic information and multiple historical fault parameters; Performing weighted calculation on the fault location similarity and fault parameter similarity of each historical power grid fault feature information to obtain multiple historical power grid fault similarities; The historical power grid fault feature information with the greatest similarity to the historical power grid faults is selected and output to obtain the same family of historical power grid faults.

5. The power grid digital twin fault simulation device reconstructed by historical data according to claim 1 is characterized in that: The digital twin simulation unit is further configured to perform the following steps: Using the multiple reconstructed fault operation data of the multiple power grid devices in the reconstructed power grid fault operation data set, marking the multiple power grid devices; Based on the digital twin, a power grid fault twin model is constructed according to the identification results as the power grid fault twin simulation result.

Images