Digital twin data traceability method based on blockchain

Abstract

The present invention discloses a digital twin data tracing method based on blockchain, which belongs to the field of information processing and management system, wherein the method includes: sensing and collecting monitoring information of multiple power grid locations in the power grid, uploading it to the twin blockchain through the power grid nodes of multiple power grid locations, and performing matching analysis; based on digital twins, using multiple monitoring information respectively, constructing multiple twin power grid models; performing local power grid stability analysis and overall power grid stability analysis on multiple twin model blocks, marking multiple twin model blocks, and adding them to the twin data blockchain; when a fault occurs, obtaining the fault location, local fault scale and overall fault scale, tracing the source in the twin data blockchain, obtaining a traceable twin power grid model set, and performing power grid fault troubleshooting as a traceability result. The present application solves the technical problem that the accuracy of power grid monitoring data and the efficiency of fault tracing in the prior art are low, thereby increasing the difficulty of power grid maintenance and management.

Description

Technical Field

[0001] The present invention relates to the field of information processing and management systems, and in particular to a digital twin data tracing method based on blockchain. Background Art

[0002] With the continuous growth of global electricity demand and the expansion of power grid scale, the complexity and management difficulty of power systems are also increasing. Modern power grids not only need to transmit electricity stably and reliably, but also need to have the ability to respond quickly and adapt to emergencies. In order to improve the management efficiency of power grids and their ability to respond to emergencies, digital twin technology and blockchain technology are gradually entering the application field of power systems.

[0003] At present, the power grid monitoring system mainly relies on distributed sensing devices, which can collect monitoring information from various locations of the power grid in real time. However, due to the huge amount of data and its dispersion, how to ensure the authenticity and integrity of the data has become a major challenge. At the same time, the existing power grid management system often has problems such as slow response speed and inaccurate fault location when dealing with faults and abnormal situations, which seriously affects the stable operation and maintenance efficiency of the power grid. Summary of the invention

[0004] This application provides a blockchain-based digital twin data tracing method to solve the technical problems in the prior art of low accuracy of power grid monitoring data and low efficiency of fault tracing, which leads to increased difficulty in power grid maintenance and management.

[0005] In view of the above problems, this application provides a digital twin data tracing method based on blockchain.

[0006] The present application provides a blockchain-based digital twin data tracing method, which includes sensing and collecting monitoring information of multiple power grid locations in the power grid based on a distributed sensing device array deployed in the power grid to obtain multiple monitoring information; uploading the multiple monitoring information to the twin blockchain through the power grid nodes of the multiple power grid locations, and performing matching degree analysis on the multiple monitoring information respectively to obtain multiple monitoring matching degrees; in the twin blockchain, based on the digital twin, the multiple monitoring information is respectively used to construct multiple twin power grid models to form multiple twin model blocks; in the twin blockchain, local power grid stability analysis and overall power grid stability analysis are performed on the multiple twin model blocks to obtain multiple local stability sets and multiple overall stabilities, and the multiple twin model blocks are identified in combination with the multiple monitoring matching degrees, and added to the twin data blockchain; when a fault occurs in the power grid, the fault location, local fault scale and overall fault scale are obtained, and traced in the twin data blockchain to obtain a traceable twin power grid model set, and the power grid fault is troubleshooted as a traceability result, wherein the traceability twin power grid models are sorted according to the monitoring matching degree in the traceability result.

[0007] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0008] Due to the use of a distributed sensing device array deployed in the power grid, the monitoring information of multiple power grid locations in the power grid is sensed and collected, and uploaded to the twin blockchain through the power grid nodes at multiple power grid locations, the matching degree analysis of the multiple monitoring information is performed respectively, and the technical solution of multiple monitoring matching degrees is obtained, which solves the technical problems of low accuracy of power grid monitoring data and low efficiency of fault tracing. In the twin blockchain, based on the digital twin technology, the multiple monitoring information is respectively used to construct multiple twin power grid models, form multiple twin model blocks, and local and overall power grid stability analysis is performed on these blocks to obtain multiple local stability sets and multiple overall stabilities. Combined with the multiple monitoring matching degrees, multiple twin model blocks are identified and added to the twin data blockchain. When a fault occurs in the power grid, the fault location, local fault scale and overall fault scale are obtained, and the traceability analysis is performed to obtain a traceability twin power grid model set, and sorted by monitoring matching degree. This technical solution achieves the technical effect of improving the accuracy of power grid monitoring data and fault handling efficiency.

[0009] 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

[0010] Figure 1 A flowchart of a blockchain-based digital twin data tracing method is provided for an embodiment of the present application.

[0011] Figure 2 A flowchart for obtaining multiple monitoring matching degrees in a blockchain-based digital twin data tracing method is provided for an embodiment of the present application. DETAILED DESCRIPTION

[0012] The overall idea of ​​the technical solution provided by this application is as follows:

[0013] The embodiment of the present application provides a digital twin data tracing method based on blockchain. By deploying distributed sensing devices in the power grid, monitoring information of each power grid location is collected in real time. These monitoring information are uploaded to the twin blockchain through the power grid node, and matching analysis is performed to ensure the accuracy of the data. Using digital twin technology, multiple twin power grid models are constructed based on these monitoring information, and local and overall stability analysis is performed. In this way, the operating status of the power grid can be monitored in real time, the fault location and scale can be quickly obtained when a fault occurs, and the source can be traced in the twin blockchain, and the analysis results can be sorted, ultimately improving the efficiency of power grid fault troubleshooting and processing.

[0014] After introducing the basic principles of the present application, various non-limiting implementation methods of the present application will be specifically described below in conjunction with the drawings in the specification.

[0015] Examples, such as Figure 1 As shown, the embodiment of the present application provides a digital twin data tracing method based on blockchain, the method comprising:

[0016] Step S100: Based on a distributed sensing device array deployed in the power grid, sensing and collecting monitoring information of multiple power grid locations in the power grid to obtain multiple monitoring information.

[0017] Specifically, a distributed sensing device array refers to a series of sensors or sensing devices deployed at different locations in the power grid, which can collect data in a distributed manner. For example, these devices can be temperature sensors, voltage sensors, current sensors, etc., which are installed at key locations such as transformers, transmission lines, and distribution cabinets to monitor the operating status of these locations. The power grid location refers to a specific location or node in the power system, such as a substation, transmission line, distribution box, etc. These locations are key points in the operation of the power grid. Monitoring information refers to the data collected by the sensing device, including parameters related to the operating status of the power grid, such as voltage, current, temperature, humidity, etc.

[0018] In modern power grids, in order to achieve real-time monitoring and management of the entire power system, a large number of sensors need to be deployed at different locations in the power grid. These sensors form a distributed sensing device array. First, select some key power grid locations, such as substations, transmission lines, and distribution boxes, and install various types of sensors at these locations. For example, temperature sensors are installed in substations to monitor the temperature of transformers, voltage sensors are installed on transmission lines to monitor voltage levels, and current sensors are installed in distribution boxes to monitor current flow.

[0019] These sensors transmit the collected data to a centralized data processing system via wireless or wired means. In this process, the sensing devices collect monitoring information from various grid locations in real time, such as temperature data of transformers, voltage data of transmission lines, and current data of distribution boxes. After these monitoring information are transmitted to the data processing system, the system will summarize and analyze these data to generate multiple monitoring information sets.

[0020] For example, the temperature sensor in the substation can collect temperature data every minute, the voltage sensor in the transmission line can monitor and transmit voltage data in real time, and the current sensor in the distribution box can monitor the changes in current to monitor the operating status of the power grid.

[0021] Through the deployment of this distributed sensing device array, comprehensive monitoring of the entire power grid can be achieved, ensuring the stability of the power grid under normal operation, and being able to promptly detect and handle abnormal situations in the power grid. For example, when the voltage data of a certain section of the transmission line suddenly drops, the system can quickly locate the problem by analyzing the data of other monitoring points, and take timely measures to prevent a larger failure of the power system.

[0022] Step S200: Upload the multiple monitoring information to the twin blockchain through the grid nodes at the multiple grid locations, and perform matching analysis on the multiple monitoring information respectively to obtain multiple monitoring matching degrees.

[0023] Specifically, a grid node refers to a data aggregation point or relay point in the power grid, through which the monitoring information collected by the sensing device can be uploaded and processed. A grid node can be a substation, distribution box, etc., which is responsible for data collection and transmission. Twin Blockchain is a digital twin system based on blockchain technology, which is used to store and manage the monitoring information of the power grid. Blockchain technology ensures the immutability and transparency of data, while digital twin technology creates a virtual model of the power grid for analysis and prediction. Matching analysis refers to the similarity or consistency analysis of monitoring information to determine whether the data between different monitoring points matches. The higher the matching degree, the better the consistency of the data and the higher the credibility.

[0024] First, the sensing devices located at various grid locations (such as substations, distribution boxes, etc.) transmit the monitored data to the corresponding grid nodes. After receiving the monitoring information, these grid nodes upload the data to the twin blockchain through the network. The monitoring information uploaded to the twin blockchain will then undergo a matching analysis. The purpose of the matching analysis is to determine the consistency between different monitoring information in order to evaluate the credibility of the data.

[0025] Through this step, a comprehensive matching analysis can be performed on the monitoring information of the power grid, generating multiple monitoring matching results. These results can identify data anomalies and ensure the accuracy and reliability of the monitoring information.

[0026] Step S300: In the twin blockchain, based on the digital twin, the multiple monitoring information are respectively used to construct multiple twin power grid models to form multiple twin model blocks.

[0027] Specifically, digital twin refers to the creation of a virtual model corresponding to a physical system (such as a power grid) that reflects the state and behavior of the physical system through real-time data updates and simulations.

[0028] First, the distributed sensing devices in the power grid will collect monitoring information from various locations in real time. This monitoring information is uploaded to the twin blockchain through the power grid nodes. Next, based on the monitoring information uploaded to the twin blockchain, the system will create multiple twin power grid models. Digital twin technology generates a virtual model corresponding to the physical power grid by mapping the monitoring information of the physical power grid into the virtual model. These twin power grid models are organized into multiple twin model blocks and stored in the blockchain.

[0029] This step generates twin power grid models based on the monitoring information of the power grid, and organizes these models into blocks and stores them in the twin blockchain. This not only ensures the security and transparency of the data, but also reflects the operating status of the power grid in real time, facilitating further analysis and prediction.

[0030] Step S400: In the twin blockchain, local power grid stability analysis and overall power grid stability analysis are performed on the multiple twin model blocks to obtain multiple local stability sets and multiple overall stabilities. The multiple twin model blocks are identified in combination with the multiple monitoring matching degrees and added to the twin data blockchain.

[0031] Specifically, local grid stability analysis is performed on each twin model block. The stability of a specific location is evaluated by analyzing the monitoring data at that location. The data of all twin model blocks are combined to perform an overall stability analysis of the entire grid.

[0032] Based on the results of local and overall grid stability analysis, multiple local stability sets and overall stability sets are generated. The twin model blocks are identified in combination with the monitoring matching degree. The monitoring matching degree evaluates the consistency of data between different monitoring points to determine the credibility of the data. The identified twin model blocks and stability sets are stored in the twin data blockchain.

[0033] This step not only ensures the security and transparency of the data, but also reflects the operating status of the power grid in real time, facilitating further analysis and optimization.

[0034] Step S500: When a fault occurs in the power grid, the fault location, local fault scale and overall fault scale are obtained, and traceability is performed within the twin data blockchain to obtain a set of traceability twin power grid models, which are used as traceability results for power grid fault troubleshooting, wherein the traceability twin power grid models are sorted according to the monitoring matching degree within the traceability results.

[0035] Specifically, when a fault occurs in the power grid, the location of the fault, the scale of the local fault, and the scale of the overall fault are obtained. The fault location refers to the specific location of the power grid where the fault occurs. The scale of the local fault refers to the scope and severity of the impact of the fault at that specific location, and the overall fault scale refers to the scope and severity of the impact of the fault on the entire power grid system.

[0036] Next, the fault is traced in the twin data blockchain. The traceability process uses the monitoring data and digital twin model stored in the blockchain to track the cause and propagation path of the fault.

[0037] Then, the obtained traceability twin power grid model set is used as the traceability result for power grid fault troubleshooting. The system will sort these traceability twin power grid models according to the monitoring matching degree. Models with high monitoring matching degree indicate good data consistency and high credibility, and these models are displayed first.

[0038] Through this step, fault tracing and troubleshooting can be carried out efficiently, improving the maintenance and operation efficiency of the power grid.

[0039] Further, such as Figure 2 As shown, the multiple monitoring information are uploaded to the twin blockchain through the grid nodes of the multiple grid locations, and the multiple monitoring information are respectively analyzed for matching to obtain multiple monitoring matching degrees, including: constructing multiple grid nodes in the multiple grid locations; constructing the twin blockchain based on the multiple grid nodes; uploading the multiple monitoring information to the twin blockchain through the multiple grid nodes; and performing matching analysis on the multiple monitoring information in the twin blockchain to obtain multiple monitoring matching degrees.

[0040] Specifically, multiple grid nodes are first built at multiple key locations of the power grid (such as substations, distribution boxes, transmission lines, etc.). These grid nodes are responsible for collecting monitoring information from the distributed sensing device array. Among them, the first step in building a grid node is to deploy sensing devices, such as installing temperature sensors in substations, voltage sensors on transmission lines, and current sensors in distribution boxes. These sensors transmit the collected data to the corresponding grid nodes through wired or wireless networks. For example, the control center of a substation can act as a grid node to receive and process data from various sensors in the substation.

[0041] Next, a twin blockchain system is built based on these power grid nodes. The twin blockchain uses blockchain technology to ensure the immutability and transparency of data, and uses digital twin technology to create a virtual model of the power grid. The key to this step is to connect each power grid node to the blockchain network so that each node can upload monitoring data to the blockchain.

[0042] When the monitoring information of multiple power grid locations is uploaded to the twin blockchain through the power grid node, the monitoring information is analyzed for matching. Matching analysis aims to evaluate the consistency between different monitoring information to determine the credibility of the data. For example, if the temperature sensor data of two adjacent locations is very different, it means that the data is highly matched and credible; otherwise, the match is low and further inspection is required.

[0043] Matching analysis can be implemented using a matching analyzer. For example, using Python's machine learning library Scikit-learn, a model can be trained to analyze and calculate the matching of temperature, voltage, and current data. First, a large amount of historical monitoring data is collected, annotated, and a sample monitoring information set is created. Then, a matching analyzer is trained based on these sample data to calculate the matching of new monitoring information.

[0044] Through this step, the monitoring information of multiple power grid locations is uploaded to the twin blockchain, and the data is analyzed for matching, which can ensure the accuracy and reliability of the power grid monitoring data and provide strong guarantees for the stable operation of the power grid.

[0045] Furthermore, matching degree analysis is performed on the multiple monitoring information respectively to obtain multiple monitoring matching degrees, including: based on the historical perception data of the distributed sensing device array, multiple sample monitoring information sets are collected, and according to the matching degree between the individual sample monitoring information and other sample detection information in each sample monitoring information set, a sample monitoring matching degree set and a sample individual monitoring information set are labeled and obtained; a monitoring information matching degree analyzer is trained using the sample individual monitoring information set, multiple sample monitoring information sets and sample monitoring matching degree set; based on the monitoring information matching degree analyzer, the multiple monitoring information are respectively combined with other monitoring information for matching degree analysis to obtain multiple monitoring matching degrees.

[0046] Specifically, based on the historical sensing data of the distributed sensing device array, multiple sets of sample monitoring information are collected. These data include historical monitoring information of different power grid locations, such as temperature, voltage, current, etc.

[0047] According to the matching degree between the individual sample monitoring information and other sample monitoring information in each sample monitoring information set, labeling is performed to obtain the sample monitoring matching set and the sample individual monitoring information set. The purpose of labeling is to assign a matching label to each sample data set to indicate the degree of consistency between the sample data.

[0048] Use the labeled sample individual monitoring information set, multiple sample monitoring information sets and sample monitoring matching sets to train a monitoring information matching analyzer. Commonly used models can be machine learning algorithms, such as decision trees, random forests, support vector machines (SVMs) or neural networks.

[0049] For newly collected monitoring data, use the trained monitoring information matching analyzer to perform matching analysis. Compare the data of each monitoring point with the data of other monitoring points to calculate the matching degree. The matching degree refers to the similarity or consistency analysis of the monitoring information to determine whether the data between different monitoring points match. The higher the matching degree, the better the consistency of the data and the higher the credibility. Based on the results of the matching analysis, it is possible to identify which data has high consistency (high credibility) and which data has anomalies and needs further inspection. Data with a high matching degree indicates good consistency and high credibility between monitoring information; data with a low matching degree indicates a monitoring equipment failure or abnormality.

[0050] Through this step, the matching degree analysis of the power grid monitoring data can be realized, the accuracy and reliability of the data can be ensured, and the operation efficiency and maintenance effect of the power grid monitoring system can be improved.

[0051] Furthermore, in the twin blockchain, based on the digital twin, the multiple monitoring information are respectively used to construct multiple twin power grid models to form multiple twin model blocks, including: according to the historical monitoring data of the multiple power grid locations, training multiple twin model generation branches for generating twin power grid models based on the monitoring information of the multiple power grid locations; combining the multiple twin model generation branches to obtain a twin model generator, and configuring it in the twin blockchain; inputting the multiple monitoring information into the multiple twin model generation branches corresponding to the twin model generator respectively, to generate multiple twin power grid models; according to the timestamp information of the multiple monitoring information, and generating block IDs for the multiple twin power grid models, combining the multiple twin power grid models to form multiple twin model blocks.

[0052] Specifically, the twin model generation branch refers to each sub-model trained based on the monitoring information of different power grid locations for generating the twin power grid model. The twin model generator is a system that combines multiple twin model generation branches to generate a complete twin power grid model. The block ID is a unique identifier used to identify and manage different data blocks in the blockchain.

[0053] First, based on the historical monitoring data of different locations of the power grid, multiple twin model generation branches are trained. Each branch corresponds to a power grid location, and the corresponding twin power grid model is generated based on the monitoring information of the location.

[0054] Combine multiple twin model generation branches to build a twin model generator and configure it in the twin blockchain. The twin model generator is responsible for inputting real-time monitoring information into the corresponding twin model generation branch to generate a twin power grid model.

[0055] The real-time monitoring information is input into multiple twin model generation branches in the twin model generator to generate multiple twin power grid models. These twin power grid models reflect the current operating status and behavior of the power grid. For example, if the temperature sensor of substation A records 30°C, the voltage of transmission line B is 220V, and the current of distribution box C is 50A, these data will be input into the corresponding twin model generation branch to generate a twin power grid model. These models can reflect the operating status of the power grid in real time, which helps to monitor and optimize the performance of the power grid.

[0056] According to the timestamp information of the monitoring information, a block ID is generated for each twin power grid model, and these models are combined to form multiple twin model blocks. For example, the twin power grid models of substation A, transmission line B, and distribution box C will generate their own block IDs and store them in the blockchain. In this way, each block will contain detailed timestamp information and monitoring data to ensure that the data will not be tampered with during the recording process, and facilitate subsequent traceability and verification.

[0057] Through this step, the system can use multiple monitoring information to build twin power grid models based on digital twin technology in the twin blockchain, and organize these models into twin model blocks for storage and management. This method not only ensures the security and transparency of data, but also reflects the operating status of the power grid in real time, facilitating further analysis and prediction.

[0058] Furthermore, based on the historical monitoring data of the multiple power grid locations, multiple twin model generation branches are trained to generate twin power grid models based on the monitoring information of the multiple power grid locations, including: collecting multiple sample monitoring information sets based on the historical monitoring data of the multiple power grid locations, and obtaining a sample twin power grid model set based on digital twin construction; splitting the multiple sample monitoring information sets according to the multiple power grid locations to obtain a first sample monitoring information set for a first power grid location; constructing a first twin model generation branch for the first power grid location based on a generative adversarial network, the first twin model generation branch including a first generator and a first adversary; using the first sample monitoring information set and the sample twin power grid model set, supervised training is performed on the first twin model generation branch until convergence; and continuing to train multiple twin model generation branches for other multiple power grid locations.

[0059] Specifically, the Generative Adversarial Network (GAN) is a deep learning model consisting of a generator and an adversary. The generator generates data, and the adversary discriminates the generated data, improving the quality of the generated data through mutual confrontation. The generator is a part of the generative adversarial network that is responsible for generating new data. The adversary is another part of the generative adversarial network that is responsible for evaluating the difference between the generated data and the real data.

[0060] First, based on the historical monitoring data of different locations of the power grid, multiple sample monitoring information sets are collected to form multiple sample monitoring information sets. The collected sample monitoring information sets are split according to the location of the power grid. For example, the data of substation A is used as the first sample monitoring information set, the data of transmission line B is used as the second sample monitoring information set, and so on. In this way, each power grid location has a corresponding sample data set, which is convenient for subsequent model training.

[0061] For each grid location, a generative adversarial network (GAN) is constructed based on its sample monitoring information set. GAN consists of a generator and an adversary. The generator is responsible for generating new data, and the adversary evaluates the difference between the generated data and the real data.

[0062] Using the first sample monitoring information set and the sample twin power grid model set, supervised training is performed on the twin model generation branch at each power grid location until the model converges. The supervised training process is that the generator continuously generates new data, the adversary continuously evaluates this data, and continuously adjusts the parameters of the generator and adversary through feedback to improve the quality of the generated data.

[0063] Repeat the above process for other grid locations and continue to train multiple twin model generation branches. Each grid location has a corresponding twin model generation branch. In this way, multiple high-quality twin grid models can be generated for the entire grid.

[0064] For example, suppose there are three locations in the power grid: substation A, transmission line B, and distribution box C. First, collect historical monitoring data for these three locations. Then, split these data into three sets of sample monitoring information respectively. Next, build a GAN for each location based on these sample data. By training the first GAN with the data of substation A, the generator generates monitoring data similar to substation A, the adversary evaluates the authenticity of the generated data, and converges the model through supervised training. Finally, repeat this process for transmission line B and distribution box C, and finally generate multiple high-quality twin power grid models.

[0065] Through this step, the system can generate multiple twin model generation branches based on the monitoring information of different power grid locations in the twin blockchain, and use these branches to generate high-quality twin power grid models. These models can reflect the real operating status of the power grid, help monitor and predict the operation of the power grid in real time, and improve the stability and reliability of the power grid.

[0066] Furthermore, in the twin blockchain, local power grid stability analysis and overall power grid stability analysis are performed on the multiple twin model blocks to obtain multiple local stability sets and multiple overall stabilities, including: training local stability analysis branches and overall stability analysis branches based on historical monitoring data of the power grid; obtaining a twin power grid model analyzer by combining the local stability analysis branches and the overall stability analysis branches, and configuring it in the twin blockchain; based on the twin power grid model analyzer, local power grid stability analysis and overall power grid stability analysis are performed on the multiple twin model blocks to obtain multiple local stability sets and multiple overall stabilities.

[0067] Specifically, first, based on the historical monitoring data of the power grid, the local stability analysis branch and the overall stability analysis branch are trained separately. These branch models can be trained based on data from different locations in order to accurately assess the stability of specific locations and the entire power grid.

[0068] Combining the local stability analysis branch and the overall stability analysis branch, a twin power grid model analyzer is constructed and configured in the twin blockchain. The twin power grid model analyzer is responsible for performing stability analysis on the twin model blocks uploaded to the blockchain.

[0069] Based on the twin power grid model analyzer, local power grid stability analysis and overall power grid stability analysis are performed on multiple twin model blocks. Local stability analysis focuses on the data at a specific location to evaluate the stability of that location; overall stability analysis integrates the data at all locations to evaluate the stability of the entire power grid.

[0070] Based on the analysis results, multiple local stability sets and overall stability are generated. These sets reflect the stability of the power grid at different levels. The analysis results (including local stability sets and overall stability) are stored in the twin data blockchain. This ensures the immutability and transparency of the data and facilitates subsequent query and analysis.

[0071] Through the above steps, the system can train the analysis model based on historical monitoring data in the twin blockchain, conduct comprehensive local and overall power grid stability analysis, and store the results in the blockchain. This not only ensures the security and transparency of the data, but also reflects the operating status of the power grid in real time, facilitating further analysis and optimization.

[0072] Furthermore, according to the historical monitoring data of the power grid, local stability analysis branches and overall stability analysis branches are trained, including: according to the historical monitoring data of the power grid, a sample twin power grid model set is obtained based on the digital twin construction, and according to the deviation between the monitoring information of the multiple power grid locations in each sample twin power grid model and the standard monitoring information, multiple sample local stability sets are marked and obtained; according to the deviation between each sample twin power grid model and the standard twin power grid model, a sample overall stability set is marked and obtained; the sample twin power grid model set and multiple sample local stability sets are used to train a local stability analysis branch; the sample twin power grid model set and the sample overall stability set are used to train an overall stability analysis branch.

[0073] Specifically, first, based on the historical monitoring data of the power grid, a set of sample twin power grid models is constructed using digital twin technology. By comparing the deviation between the monitoring information in the sample twin power grid model and the standard monitoring information, the sample local stability set is obtained by annotation. Standard monitoring information is usually ideal or expected monitoring data. The smaller the deviation, the higher the stability. For example, the actual temperature data of substation A is 30°C, and the standard temperature data is 28°C, then the deviation is 2°C.

[0074] By comparing the deviation of each sample twin power grid model with the standard twin power grid model, the overall stability set of the samples is obtained. For example, if the deviation of the actual operation data of the entire power grid from the standard model is small, the overall stability is high.

[0075] Using a sample twin power grid model set and multiple sample local stability sets, the local stability analysis branch is trained. This branch model focuses on analyzing the stability of a specific location of the power grid. Specifically, historical monitoring data of a specific power grid location (such as a substation or transmission line) is collected, which includes key parameters such as temperature, voltage, and current. Then, a sample twin power grid model is constructed based on this data using digital twin technology. Next, the monitoring information in the sample twin power grid model is compared with the standard monitoring information, and the deviation is calculated and annotated to obtain the sample local stability set. Using these annotated data as a training set, a suitable machine learning algorithm (such as random forest or neural network) is selected for model training. The model gradually optimizes parameters through multiple iterations, enabling it to accurately predict and evaluate the stability of a specific location. Finally, the verified and tuned local stability analysis branch model can be used for real-time monitoring and early warning.

[0076] Using the sample twin power grid model set and the sample overall stability set, the overall stability analysis branch is trained. This branch model is used to analyze the comprehensive stability of the entire power grid. Specifically, historical monitoring data of the entire power grid is collected, including parameters such as temperature, voltage, and current at all key locations (such as substations, transmission lines, and distribution boxes). Based on these data, a sample twin power grid model is constructed using digital twin technology. Then, each sample twin power grid model is compared with the standard twin power grid model, and the overall deviation is calculated and annotated to obtain the sample overall stability set. Using these annotated data as a training set, a suitable machine learning algorithm (such as support vector machine, random forest, or deep learning model) is selected for training. Through multiple iterations and optimized parameters, the model can comprehensively analyze the stability of the entire power grid. Finally, the verified and tuned overall stability analysis branch model can be used for real-time monitoring and stability evaluation of the entire network to improve the safety and reliability of the power grid.

[0077] Through this step, local and overall stability analysis branches can be constructed and trained to achieve comprehensive monitoring and analysis of the power grid operation status. These analysis results will help to timely identify and deal with potential problems in the power grid and improve the reliability and security of the power grid.

[0078] Furthermore, when a fault occurs in the power grid, the fault location, local fault scale and overall fault scale are obtained, and traceability is performed in the twin data blockchain to obtain a set of traceable twin power grid models, and power grid fault troubleshooting is performed as the traceability result, including: when a fault occurs in the power grid, the fault location, local fault scale and overall fault scale are obtained; the local fault scale and the overall fault scale are converted into local stability and overall stability to obtain local stability of the fault and overall stability of the fault, wherein a sample local fault scale set and a sample local stability set are collected, and a fault local stability comparison table is constructed to perform local stability conversion; according to the local stability of the fault and the overall stability of the fault, traceability is performed in the twin data blockchain to obtain several traceable twin power grid models with the same local stability and overall stability at the same power grid location, and output as a traceable twin power grid model set; according to the monitoring matching degree of several traceable twin power grid models in the traceable twin power grid model set, several traceable twin power grid models are sorted in order from large to small to obtain traceability results and perform power grid fault troubleshooting.

[0079] Specifically, first, the system will obtain the location of the fault, the local fault scale, and the overall fault scale. For example, if the fault location is substation A, the local fault scale includes the degree of damage to the equipment in substation A, and the overall fault scale affects multiple substations and transmission lines near substation A.

[0080] Next, the system converts the local fault scale and the overall fault scale into local stability and overall stability. This step is achieved by collecting sample local fault scale sets and sample local stability sets, and constructing a fault local stability comparison table. Specifically, historical monitoring data and fault records of key locations in the power grid are collected to establish sample local fault scale sets and sample local stability sets. Based on these data, a fault local stability comparison table is constructed to clearly mark the stability levels corresponding to different fault scales. For example, minor equipment faults correspond to high stability, and serious equipment faults correspond to low stability. Similarly, the overall fault scale generates a sample overall stability set and a comparison table by analyzing the historical fault impact range and stability changes of the entire power grid. Through these comparison tables, the actual fault scale is converted into the corresponding local and overall stability indicators when a fault occurs, which is convenient for further analysis and processing. For example, the degree of equipment damage in substation A is converted into low stability through the comparison table, and the overall fault scale affecting the adjacent area is also converted into a lower overall stability.

[0081] Then, the system traces the source in the twin data blockchain to find the twin power grid model that matches the local stability and overall stability of the fault. The traceability process uses the monitoring data and twin models stored in the blockchain to identify models with the same grid location and the same stability, and generate a set of traceable twin power grid models.

[0082] Subsequently, the models are sorted according to the monitoring matching degree of each model in the traceability twin power grid model set. Models with high matching degree indicate good data consistency and high credibility, and the system will give priority to displaying these models. For example, if the monitoring data of substation A and adjacent substation B have a high matching degree, the system will give priority to displaying the twin models of these two substations, helping managers to find the source of the fault and the scope of impact more quickly.

[0083] For example, assuming that a fault occurs in substation A in the power grid, the system first obtains the fault location (substation A), the local fault scale (the severity of equipment damage in substation A) and the overall fault scale (affecting multiple substations and transmission lines near substation A). Then, the system converts the local fault scale into the local stability of the fault through the fault local stability comparison table. For example, the main equipment in substation A is damaged, resulting in a decrease in its local stability. Next, the system traces the source in the twin data blockchain to find a twin power grid model with the same location as substation A and the same local stability and overall stability. The system obtains the historical monitoring data of substation A, analyzes the abnormal fluctuations in these data, and finds the source of the fault. Assuming that the historical data shows that substation A has experienced abnormal voltage fluctuations many times under high temperature conditions, these abnormal models will be included in the traceability twin power grid model set. Finally, the system sorts these traceability twin power grid models according to the monitoring matching degree, and gives priority to displaying models with high matching degree. Assuming that the temperature and voltage data of substation A and the adjacent substation B have a high matching degree, these models will be displayed first to help managers quickly identify the source of the fault and take corresponding repair measures.

[0084] Through this step, the system can efficiently trace and troubleshoot faults, improving the maintenance and operation efficiency of the power grid.

[0085] In summary, the digital twin data tracing method based on blockchain provided in the embodiment of the present application has the following technical effects:

[0086] 1. By deploying distributed sensing devices in the power grid, monitoring information of multiple power grid locations is collected in real time and uploaded to the twin blockchain for matching analysis and stability analysis. A virtual model of the power grid is constructed using digital twin technology to achieve comprehensive monitoring of the power grid status. When a power grid fails, traceability analysis is performed based on the blockchain to provide accurate troubleshooting results. This method improves the accuracy and credibility of power grid monitoring data, enhances fault detection and analysis capabilities, and ensures the stable operation of the power grid.

[0087] 2. Multiple twin power grid models are built based on digital twin technology in the twin blockchain, and these models are organized into blocks to ensure the immutability and transparency of the data. By training and configuring the twin model generator, the system can generate a virtual model that reflects the operating status of the power grid in real time, providing an accurate and dynamic tool for monitoring and management of the power grid. This method improves the management and utilization efficiency of power grid data and ensures the reliability of power grid operation.

[0088] 3. Perform local and overall grid stability analysis within the twin blockchain, generate stability sets, and identify twin model blocks in combination with monitoring matching, ensuring the comprehensiveness and accuracy of data analysis. This approach enables the system to monitor grid stability in real time, identify potential problems, and take timely measures, thereby improving the reliability and safety of the grid.

[0089] 4. In the event of a power grid failure, the fault location and scale are obtained, stability conversion is performed, and traceability analysis is combined to find the source of the fault. This process uses historical data and models in the twin data blockchain to provide an efficient troubleshooting method. The system can quickly locate the fault point and assess its impact range, and provide the most reliable traceability results by monitoring the matching degree sorting, which significantly improves the speed and accuracy of fault handling and ensures the timely recovery and stable operation of the power grid.

[0090] Any step of the method described above can be stored as a computer instruction or program in an unlimited computer memory, and can be called and recognized by an unlimited computer processor to implement any method in the embodiments of the present application, without any unnecessary restrictions.

[0091] Furthermore, the first or second mentioned above may not only represent an order relationship, but may also represent a specific concept, and / or refer to multiple elements that can be selected individually or in full. Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application intends to include these modifications and variations.

Claims

1. The digital twin data tracing method based on blockchain is characterized by: The method comprises: Based on the distributed sensing device array deployed in the power grid, the monitoring information of multiple power grid locations in the power grid is sensed and collected to obtain multiple monitoring information; Upload the multiple monitoring information to the twin blockchain through the grid nodes at the multiple grid locations, and perform matching degree analysis on the multiple monitoring information to obtain multiple monitoring matching degrees; In the twin blockchain, based on the digital twin, the plurality of monitoring information are respectively used to construct a plurality of twin power grid models to form a plurality of twin model blocks; In the twin blockchain, local power grid stability analysis and overall power grid stability analysis are performed on the multiple twin model blocks to obtain multiple local stability sets and multiple overall stabilities, and the multiple twin model blocks are identified in combination with the multiple monitoring matching degrees and added to the twin data blockchain; When a fault occurs in the power grid, the fault location, local fault scale and overall fault scale are obtained, and traceability is performed in the twin data blockchain to obtain a set of traceable twin power grid models, which are used as traceability results to troubleshoot power grid faults, wherein the traceability twin power grid models are sorted according to the monitoring matching degree in the traceability results; Among them, the multiple monitoring information are uploaded to the twin blockchain through the grid nodes of the multiple grid locations, and the multiple monitoring information are respectively analyzed for matching degree to obtain multiple monitoring matching degrees, including: constructing a plurality of grid nodes within the plurality of grid locations; Based on the multiple power grid nodes, construct the twin blockchain; Uploading the plurality of monitoring information to the twin blockchain through the plurality of power grid nodes; In the twin blockchain, the plurality of monitoring information are respectively analyzed for matching, to obtain a plurality of monitoring matching degrees; The matching degree analysis is performed on the plurality of monitoring information respectively to obtain a plurality of monitoring matching degrees, including: Based on the historical sensing data of the distributed sensing device array, multiple sample monitoring information sets are collected, and according to the matching degree between the individual sample monitoring information and other sample detection information in each sample monitoring information set, a sample monitoring matching degree set and a sample individual monitoring information set are annotated and obtained; Using the sample individual monitoring information set, multiple sample monitoring information sets and sample monitoring matching degree set, training a monitoring information matching degree analyzer; Based on the monitoring information matching degree analyzer, matching degree analysis is performed on the multiple monitoring information in combination with other monitoring information to obtain multiple monitoring matching degrees.

2. The digital twin data tracing method based on blockchain according to claim 1 is characterized in that: In the twin blockchain, based on the digital twin, the plurality of monitoring information are respectively used to construct a plurality of twin power grid models to form a plurality of twin model blocks, including: According to the historical monitoring data of the plurality of power grid locations, training a plurality of twin model generation branches for generating a twin power grid model based on the monitoring information of the plurality of power grid locations respectively; Combining the multiple twin model generation branches, obtaining a twin model generator, and configuring it in the twin blockchain; Inputting the plurality of monitoring information into the plurality of twin model generation branches corresponding to the twin model generator respectively, to generate a plurality of twin power grid models; According to the timestamp information of the multiple monitoring information, block IDs are generated for the multiple twin power grid models, and the multiple twin power grid models are combined to form multiple twin model blocks.

3. The digital twin data tracing method based on blockchain according to claim 2 is characterized in that: According to the historical monitoring data of the multiple power grid locations, training is performed to generate multiple twin model generation branches of a twin power grid model based on the monitoring information of the multiple power grid locations, including: Collecting multiple sample monitoring information sets based on historical monitoring data of the multiple power grid locations, and obtaining a sample twin power grid model set based on digital twin construction; Splitting the multiple sample monitoring information sets according to the multiple power grid locations to obtain a first sample monitoring information set at a first power grid location; Based on a generative adversarial network, construct a first twin model generation branch of the first power grid location, wherein the first twin model generation branch includes a first generator and a first adversary; Using the first sample monitoring information set and the sample twin power grid model set, supervised training is performed on the first twin model generation branch until convergence; Continue training multiple twin models at multiple other grid locations to generate branches.

4. The digital twin data tracing method based on blockchain according to claim 1 is characterized in that: In the twin blockchain, local power grid stability analysis and overall power grid stability analysis are performed on the multiple twin model blocks to obtain multiple local stability sets and multiple overall stabilities, including: training a local stability analysis branch and an overall stability analysis branch according to historical monitoring data of the power grid; Combining the local stability analysis branch and the overall stability analysis branch, a twin power grid model analyzer is obtained and configured in the twin blockchain; Based on the twin power grid model analyzer, local power grid stability analysis and overall power grid stability analysis are performed on the multiple twin model blocks to obtain multiple local stability sets and multiple overall stabilities.

5. The digital twin data tracing method based on blockchain according to claim 4 is characterized in that: According to the historical monitoring data of the power grid, a local stability analysis branch and an overall stability analysis branch are trained, including: According to the historical monitoring data of the power grid, a set of sample twin power grid models is obtained based on the digital twin construction, and according to the deviation between the monitoring information of the multiple power grid locations in each sample twin power grid model and the standard monitoring information, multiple sample local stability sets are annotated and obtained; According to the deviation between each sample twin power grid model and the standard twin power grid model, the overall stability set of the samples is obtained by annotation; Using the sample twin power grid model set and multiple sample local stability sets, training to obtain a local stability analysis branch; The sample twin power grid model set and the sample overall stability set are used to train and obtain an overall stability analysis branch.

6. The digital twin data tracing method based on blockchain according to claim 4 is characterized in that: When a fault occurs in the power grid, the fault location, local fault scale and overall fault scale are obtained, and traceability is performed in the twin data blockchain to obtain a set of traceable twin power grid models, and the power grid fault is investigated as a traceability result, including: When a fault occurs in the power grid, obtaining the fault location, local fault scale and overall fault scale; Performing local stability and overall stability conversion on the local fault scale and the overall fault scale to obtain fault local stability and fault overall stability, wherein a sample local fault scale set and a sample local stability set are collected, and a fault local stability comparison table is constructed to perform local stability conversion; According to the local stability of the fault and the overall stability of the fault, the source is traced in the twin data blockchain to obtain several traceable twin power grid models with the same local stability and overall stability at the same power grid location, and the output is a set of traceable twin power grid models; According to the monitoring matching degree of several traceability twin power grid models in the traceability twin power grid model set, the several traceability twin power grid models are sorted in order from large to small, the traceability results are obtained, and power grid fault troubleshooting is carried out.

Images