Power grid security control method and system based on twin sandbox

Abstract

The present invention discloses a power grid security management and control method and system based on a twin sandbox, which relates to the technical field related to the distribution system. The method includes: evaluating the power grid structure of the power grid, establishing a data synchronization node according to the evaluation result of the power grid structure, and building a digital twin model of the power grid. Configure a sensor network, perform key parameter monitoring, establish a real-time shared data set, and synchronize it to the digital twin model to generate a first safety warning result. Obtain the power grid tasks of the power grid and deploy local sub-models. Synchronize the real-time shared data set, activate the local sub-model to identify local anomalies, and establish a second safety warning result. Authenticate and identify the first safety warning result and the second safety warning result, and establish a comprehensive safety management and control plan. It solves the technical problem that the power grid monitoring and management methods in the prior art have poor response capabilities for real-time monitoring and abnormal warning, resulting in the difficulty in improving the safety and stability of power grid operation.

Description

Technical Field

[0001] The present application relates to fields related to power distribution system technology, and specifically to a power grid security management and control method and system based on a twin sandbox. Background Art

[0002] With the complexity and expansion of modern power systems, the safety and stability of power grids are facing unprecedented challenges. Traditional power grid monitoring and management methods have poor response capabilities for real-time monitoring and abnormal warnings when facing frequent load changes and equipment failures. The power grid is often in an abnormal load operation state, making it difficult to improve the safety and stability of power grid operation.

[0003] Therefore, in the prior art, the power grid monitoring and management methods have poor response capabilities for real-time monitoring and abnormal warning, resulting in technical problems such as difficulty in improving the safety and stability of power grid operation. Summary of the invention

[0004] This application solves the technical problem that the power grid monitoring and management methods in the prior art have poor response capabilities for real-time monitoring and abnormal warning, which makes it difficult to improve the safety and stability of power grid operation by providing a power grid safety management method and system based on a twin sandbox. The power grid safety management method based on the twin sandbox, by introducing a digital twin model, a sensor network, and an association analysis network, can timely capture abnormal conditions in the power grid and generate safety warnings, improve the response capabilities of monitoring and abnormal warnings, facilitate technical personnel to quickly conduct safety management of the power grid, and ensure the safety and stability of power grid operation.

[0005] The present application provides a power grid security management and control method based on a twin sandbox, the method comprising: performing a power grid structure evaluation on the power grid, and establishing a data synchronization node according to the power grid structure evaluation result; building a digital twin model of the power grid according to component information in the power grid, the digital twin model being established in a sandbox environment, and the component information including specification information, status information, and historical maintenance information of the component; configuring a sensor network at the data synchronization node, performing key parameter monitoring of the corresponding data synchronization node through the sensor network, establishing a real-time shared data set, synchronizing the real-time shared data set to the digital twin model, executing data update of the digital twin model, and generating a first safety warning result; obtaining a power grid task of the power grid, performing data synchronization node aggregation based on the power grid task, deploying a local sub-model with the aggregation result, and embedding the local sub-model into the digital twin model; after the real-time shared data set is synchronized to the digital twin model, activating the dynamic window of the local sub-model to extract data stream features, performing local anomaly identification of the feature extraction result with the local sub-model, and establishing a second safety warning result; authenticating and identifying the first safety warning result and the second safety warning result, and establishing a comprehensive safety management and control plan.

[0006] In the implementation method, the dynamic window of the activated local sub-model is used to extract data flow features, and also includes: establishing a policy space mapped to the local sub-model, the policy space is the range space of the dynamic window; obtaining historical data based on the digital twin model, and setting the initial window size of the dynamic window of each local sub-model with the historical data; obtaining the node target in each local sub-model, and establishing the utility function of the local sub-model based on the node target; performing utility evaluation of the real-time shared data set through the utility function, and establishing a utility influence coefficient; updating the initial window size in the policy space through the utility influence coefficient, and establishing a dynamic window according to the update result.

[0007] The utility function is as follows: ;in, Characterization The utility function of a local sub-model, Characterization The total number of data synchronization nodes in the local sub-model, Represents any data synchronization node, Characterization The weight of the data synchronization node, Characterized in the Nodes at time The utility function of a real-time shared dataset is established by accuracy and timeliness, Characterized in the Nodes at time Real-time shared data, For the The data sharing influence coefficient of each node is Characterization Nodes at time The utility function of historical data, For the Nodes at time Historical data, For the The environmental impact weight of each node, For in time The response time, Characterization Nodes at time The prediction accuracy of Characterization The target influence weight of each node is Characterization Nodes at time The node objective utility function of Characterization Nodes at time Target data. In the implementation method, the grid structure evaluation of the power grid, the establishment of the data synchronization node according to the grid structure evaluation result, also includes: establishing the centrality index of the node, the centrality index includes the degree centrality index, the closeness centrality index, and the betweenness centrality index; taking each component in the power grid as a basic node, the centrality evaluation of the basic node is performed through the centrality index, and the first node quantity influence is established according to the evaluation result; obtaining the attribute information of the component, and generating the second node quantity influence based on the attribute information; establishing the data synchronization node according to the first node quantity influence, the second node quantity influence and the basic node. In the implementation method, the authentication and identification of the first security warning result and the second security warning result also includes: inputting the first security warning result and the second security warning result into the association analysis network, performing the authentication support of the first security warning result and the second security warning result; outputting the security warning result whose authentication support meets the preset threshold as the authentication security warning result, and outputting the security warning result whose authentication support does not meet the preset threshold as the independent security warning result; completing the authentication and identification according to the authentication security warning result and the independent security warning result. In the implementation method, the establishment of a comprehensive safety control plan also includes: configuring a control verification cycle based on the certified safety warning results and the independent safety warning results; after the comprehensive safety control plan is established, the safety control verification of the power grid is performed through the control verification cycle to establish a safety control verification result; and the execution of the comprehensive safety control plan is corrected based on the safety control verification result.

[0008] In the implementation method, the historical safety impact characteristics and time decay penalty factors of the power grid are established based on the historical maintenance information; a safety concern database is established through the historical safety impact characteristics and time decay penalty factors; real-time shared data sets are jointly identified through the safety concern database to establish a joint identification result; and the comprehensive safety management and control plan is revised based on the joint identification result.

[0009] This application also provides a power grid security management and control system based on twin sandbox, including:

[0010] A data node acquisition module is used to evaluate the power grid structure and establish a data synchronization node based on the evaluation results of the power grid structure;

[0011] A twin model acquisition module, used to build a digital twin model of the power grid according to the component information in the power grid, wherein the digital twin model is established in a sandbox environment, and the component information includes specification information, status information, and historical maintenance information of the component; a first early warning module, used to configure a sensor network at the data synchronization node, perform key parameter monitoring of the corresponding data synchronization node through the sensor network, establish a real-time shared data set, synchronize the real-time shared data set to the digital twin model, execute data update of the digital twin model, and generate a first safety early warning result;

[0012] A local sub-model construction module is used to obtain power grid tasks of the power grid, perform data synchronization node aggregation based on the power grid tasks, deploy a local sub-model based on the aggregation results, and embed the local sub-model into the digital twin model;

[0013] A second warning module is used to activate the dynamic window of the local sub-model to extract data stream features after the real-time shared data set is synchronized to the digital twin model, and to identify local anomalies of the feature extraction results with the local sub-model to establish a second safety warning result;

[0014] The security management and control module is used to authenticate and identify the first security warning result and the second security warning result and establish a comprehensive security management and control plan.

[0015] The power grid security management and control method and system based on the twin sandbox proposed in this application is proposed. By evaluating the power grid structure, a data synchronization node is established according to the evaluation results of the power grid structure, and a digital twin model of the power grid is built. Configure the sensor network, perform key parameter monitoring, establish a real-time shared data set, and synchronize it to the digital twin model to generate a first safety warning result. Obtain the power grid tasks of the power grid and deploy local sub-models. Synchronize the real-time shared data set, activate the local sub-model to identify local anomalies, and establish a second safety warning result. Authenticate and identify the first safety warning result and the second safety warning result, and establish a comprehensive safety management and control plan. It solves the technical problem that the power grid monitoring and management methods in the prior art have poor response capabilities for real-time monitoring and abnormal warning, resulting in difficulty in improving the safety and stability of power grid operation. The power grid security management and control method based on the twin sandbox, by introducing digital twin models, sensor networks, and correlation analysis networks, can realize the timely capture of abnormal situations in the power grid and generate safety warnings, improve the response capabilities of monitoring and abnormal warnings, and facilitate technical personnel to quickly conduct safety management of the power grid, ensuring the safety and stability of power grid operation. Description of the drawings

[0016] In order to more clearly illustrate the technical solution of the embodiment of the present disclosure, the accompanying drawings of the embodiment of the present disclosure will be briefly introduced below. A flow chart is used in the present application to illustrate the operations performed by the system according to the embodiment of the present application. It should be understood that the preceding or following operations are not necessarily performed accurately in order. On the contrary, various steps can be processed in reverse order or simultaneously as needed. At the same time, other operations can also be added to these processes, or one or more operations can be removed from these processes.

[0017] Figure 1 A schematic diagram of a process flow of a power grid security management and control method based on a twin sandbox provided in an embodiment of the present application;

[0018] Figure 2 A schematic diagram of the structure of a power grid security management and control system based on a twin sandbox provided in an embodiment of the present application.

[0019] Explanation of the accompanying drawings: data node acquisition module 11, twin model acquisition module 12, first early warning module 13, local sub-model construction module 14, second early warning module 15, safety management and control module 16. DETAILED DESCRIPTION

[0020] 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.

[0021] In order to make the purpose, technical scheme and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings. The described embodiments should not be regarded as limiting the present application. All other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of the present application. In the following description, "some embodiments" are involved, which describe a subset of all possible embodiments, but it can be understood that "some embodiments" can be the same subset or different subsets of all possible embodiments, and can be combined with each other without conflict. The term "first\second" involved is only to distinguish similar objects and does not represent a specific ordering for objects. The terms "including" and "having" and any variations are intended to cover non-exclusive inclusions. For example, a process, method, system, product or server that includes a series of steps or units does not have to be limited to those steps or units that are clearly listed, but may include other steps or modules that are not clearly listed or inherent to these processes, methods, products or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those generally understood by technicians in the technical field of the present application. The terms used herein are only for the purpose of describing the embodiments of the present application.

[0022] The present application embodiment provides a power grid security management method and system based on a twin sandbox, such as Figure 1 As shown, the method includes:

[0023] The power grid structure is evaluated for the power grid, and a data synchronization node is established according to the evaluation result of the power grid structure; a digital twin model of the power grid is built according to the component information in the power grid, and the digital twin model is established in a sandbox environment, and the component information includes the specification information, status information, and historical maintenance information of the component; a sensor network is configured at the data synchronization node, and the key parameter monitoring of the corresponding data synchronization node is performed through the sensor network, a real-time shared data set is established, and the real-time shared data set is synchronized to the digital twin model, and the data update of the digital twin model is performed to generate a first safety warning result; the power grid structure is evaluated for the power grid, the importance of each basic node in the power grid structure is evaluated, and a data synchronization node is established according to the evaluation result of the power grid structure. Subsequently, after the data synchronization node is established, a digital twin model of the power grid is built according to the component information in the power grid. The digital twin model is a virtual presentation of the real-time status of the power grid in the sandbox environment. Among them, the information of the power grid components includes: equipment specification information, historical maintenance information, and status information. The equipment specification information includes the design specification parameters of the equipment, such as the capacity of the transformer, the voltage level of the line, etc. The status information is the current real-time status, such as the current load current, the status of the switch, etc. With this information, a complete digital twin model of the power grid is created in the sandbox environment. The model can simulate the state of the power grid under different conditions through operating parameters. Furthermore, a sensor network is configured in the data synchronization node to monitor the key parameters of the power grid and establish a real-time shared data set. The real-time shared data set data is transmitted to the digital twin model, and the data update of the digital twin model is executed to realize the simulation operation, obtain the equipment operation parameters in the simulation operation, and determine whether the operation parameters of each equipment exceed the design parameters of the corresponding equipment. If exceeded, the simulation operation is abnormal, and the first safety warning result is generated based on the abnormal equipment and abnormal parameters.

[0024] The method provided in the embodiment of the present application also includes: establishing a centrality index of a node, wherein the centrality index includes a degree centrality index, a closeness centrality index, and an betweenness centrality index; taking each component in the power grid as a basic node, evaluating the centrality of the basic node through the centrality index, and establishing a first node quantity influence according to the evaluation result; obtaining the attribute information of the component, generating a second node quantity influence based on the attribute information; and establishing a data synchronization node according to the first node quantity influence, the second node quantity influence, and the basic node. The power grid structure evaluation of the power grid, and establishing a data synchronization node according to the power grid structure evaluation result, also include: the goal of the power grid structure evaluation is to identify key nodes in the power grid, and the nodes play an important role in power transmission and distribution. By identifying these nodes, the state of the power grid can be more effectively monitored and its safety and stability can be improved. By establishing a centrality index of a node, the centrality index includes a degree centrality index, a closeness centrality index, and an betweenness centrality index. The degree centrality index is a measure of the number of nodes directly connected to a node. Nodes with high degree centrality are usually key points with many connection lines and responsible for a large amount of power transmission. The closeness centrality index is a measure of the average shortest path distance between a node and all other nodes. Nodes with high proximity centrality are usually able to affect the entire network more quickly and are the core nodes for information or power transmission in the power grid. The betweenness centrality index measures how many shortest paths a node plays an intermediary role in other nodes. Nodes with high betweenness centrality are usually "bridges" for power flow in the power grid, and their failures will have widespread impacts.

[0025] Each component in the power grid (such as a substation, a generator, a transmission line, etc.) is taken as a basic node, and the degree centrality, closeness centrality and betweenness centrality of each basic node are calculated by the centrality index, closeness centrality index and betweenness centrality index of the centrality index, wherein the calculation of degree centrality, closeness centrality and betweenness centrality is a prior art and will not be described in detail here. The centrality evaluation of the basic node is performed, specifically the standardized degree centrality, closeness centrality and betweenness centrality, and then, according to the needs of the actual application, weights are assigned to different centrality indicators, and the comprehensive centrality calculation is performed based on the weights and standardized parameters to obtain the centrality evaluation result, and the first node quantity influence is established according to the evaluation result, and the first node quantity influence is the importance of the current component. Subsequently, the attribute information of the component is obtained, and the second node quantity influence is generated based on the attribute information, and the attribute information includes the load capacity of the component, the component category, etc. The corresponding importance parameter is obtained according to the attribute information, wherein different load capacity presets correspond to different importance parameters, and different component categories presets correspond to different importance parameters, and the two are standardized and added to obtain the second node quantity influence. Finally, a data synchronization node is established according to the influence of the first node quantity, the second node quantity and the basic node. The data synchronization node is a basic node whose sum of the first node quantity influence and the second node quantity influence is greater than or equal to the preset node quantity influence threshold. The preset node quantity influence threshold is the lowest threshold of the influence degree. When it is greater than or equal to the threshold, the influence degree of the corresponding node is higher and the node is more important. Therefore, establishing a data synchronization node at the corresponding node has the best effect.

[0026] Obtain the power grid tasks of the power grid, perform data synchronization node aggregation based on the power grid tasks, deploy a local sub-model based on the aggregation results, and embed the local sub-model into the digital twin model; after the real-time shared data set is synchronized to the digital twin model, activate the dynamic window of the local sub-model to extract data stream features, use the local sub-model to perform local anomaly identification of the feature extraction results, and establish a second safety warning result; authenticate and identify the first safety warning result and the second safety warning result, and establish a comprehensive safety management and control plan.

[0027] The power grid task of the power grid is obtained. The power grid task is a pre-assigned power grid task such as new load adjustment, equipment maintenance, etc. The power grid task also includes specific execution equipment requirements. Data synchronization node aggregation is performed based on the power grid task, that is, according to the specific requirements of the power grid task, the relevant data synchronization nodes containing the power grid task execution equipment are aggregated. Further, a local sub-model is deployed at the corresponding aggregation node based on the aggregation result, each local sub-model is connected to the node in the aggregation result, a dedicated local sub-model is constructed for each power grid task, and the local sub-model is embedded in the digital twin model. The local sub-model is used to identify anomalies based on the data of the aggregated data synchronization node to determine whether the data synchronization node is abnormal. Among them, the local sub-model is constructed based on the recurrent neural network model, and the historical power grid operation data in the past period of time is extracted from the power grid operation database for model training and verification. The historical power grid operation data includes the operation parameters collected by the sensor network and the corresponding equipment abnormal parameters and normal parameter identifiers. The historical power grid operation data is divided into a training set, a validation set and a test set, and the recurrent neural network model is trained based on the training set, the validation set and the test set, and the model performance is tested through the validation set data, and the model parameters are adjusted to improve the accuracy. Finally, the test set is used to complete the model test until the local sub-model is obtained when the output accuracy of the model meets the preset requirements. After the real-time shared data set is synchronized to the digital twin model, the dynamic window of the local sub-model is activated to extract data stream features, and the local abnormal parameters of the feature extraction results are identified by the local sub-model to establish a second safety warning result. The first safety warning result and the second safety warning result are authenticated and identified, and a comprehensive safety management and control plan is established. The technical problem that the power grid monitoring and management methods in the prior art have poor response capabilities for real-time monitoring and abnormal warning, resulting in the difficulty in improving the safety and stability of power grid operation is solved. The power grid safety management and control method based on the twin sandbox, by introducing digital twin models, sensor networks, and association analysis networks, can timely capture abnormal situations in the power grid and generate safety warnings, improve the response capabilities of monitoring and abnormal warnings, facilitate technical personnel to quickly conduct safety management of the power grid, and ensure the safety and stability of power grid operation.

[0028] The method provided in an embodiment of the present application also includes: establishing a policy space mapped to a local sub-model, wherein the policy space is a range space of a dynamic window; acquiring historical data based on the digital twin model, and setting an initial window size of a dynamic window of each local sub-model using the historical data; acquiring node targets within each local sub-model, and establishing a utility function of the local sub-model based on the node targets; performing utility evaluation of a real-time shared data set through the utility function, and establishing a utility impact coefficient; updating the initial window size in the policy space through the utility impact coefficient, and establishing a dynamic window based on the update result.

[0029] The dynamic window of the activated local sub-model is used to extract data stream features, and further includes: establishing a strategy space mapped to the local sub-model, the strategy space is the range space of the dynamic window, that is, the time series data generated by the data synchronization node, and extracting the time series range of the time series data based on the strategy space, so that the local sub-model can quickly analyze the data. The establishment of the strategy space is specifically to obtain historical data based on the digital twin model, and the historical data is the data collected and generated by the historical time node of the digital twin model. The initial window size of the dynamic window of each local sub-model is set with the historical data, and the initial window size is the time window range of the historical data. Further, the node target in each local sub-model is obtained, and the node target is the target to be achieved by the power grid task, such as the load reaching a certain parameter value, and the utility function of the local sub-model is established based on the node target. Further, the utility evaluation of the real-time shared data set is performed through the utility function, and the calculation result of the utility function is obtained to obtain the utility impact coefficient. Finally, the initial window size is updated in the strategy space through the utility impact coefficient, and the initial window size that meets the utility impact coefficient ratio is obtained, and the dynamic window is established according to the update result. The utility function is as follows: ;in, Characterization The utility function of a local sub-model, Characterization The total number of data synchronization nodes in the local sub-model, Represents any data synchronization node, Characterization The weight of the data synchronization node, Characterized in the Nodes at time The utility function of a real-time shared dataset is established by accuracy and timeliness, Characterized in the Nodes at time Real-time shared data, For the The data sharing influence coefficient of each node is Characterization Nodes at time The utility function of historical data, For the Nodes at time Historical data, For the The environmental impact weight of each node, For in time The response time, Characterization Nodes at time The prediction accuracy of Characterization The target influence weight of each node is Characterization Nodes at time The node objective utility function of Characterization Nodes at time The method provided in the embodiment of the present application also includes: inputting the first security warning result and the second security warning result into the association analysis network, performing authentication support for the first security warning result and the second security warning result; outputting the security warning result whose authentication support degree meets the preset threshold as an authentication security warning result, and outputting the security warning result whose authentication support degree does not meet the preset threshold as an independent security warning result; completing authentication identification according to the authentication security warning result and the independent security warning result.

[0030] The authentication and identification of the first security warning result and the second security warning result also includes: inputting the first security warning result and the second security warning result into the association analysis network, and performing authentication support for the first security warning result and the second security warning result. The association analysis network is a system for analyzing and evaluating the relationship between security warning results. When constructing, the warning results in the historical data are first collected, and the co-occurrence in different events is analyzed, that is, the co-occurrence situation. Then, the proportion of the co-occurrence of abnormal parameters 1 and 2 in the warning results to the parameter occurrence is extracted to obtain the support of abnormal parameters 1 and 2. Further, the support between each parameter in the first security warning result and the second security warning result is obtained. The security warning result whose authentication support meets the preset threshold is output as the authentication security warning result, and the security warning result whose authentication support does not meet the preset threshold is output as an independent security warning result, and the preset threshold is the predicted support parameter. Finally, the authentication and identification is completed according to the authentication security warning result and the independent security warning result.

[0031] A control verification cycle is configured based on the certified safety warning results and the independent safety warning results; after a comprehensive safety control plan is established, the safety control verification of the power grid is carried out through the control verification cycle to establish a safety control verification result; and the execution of the comprehensive safety control plan is corrected based on the safety control verification result.

[0032] The establishment of a comprehensive safety control plan also includes: configuring a control verification cycle based on the certified safety warning results and the independent safety warning results, wherein the control verification cycle is determined based on the number of parameter warnings in the certified safety warning results, and the greater the number, the shorter the corresponding control verification cycle. Further, the establishment of a comprehensive safety control plan is carried out based on the obtained certified safety warning results and the independent safety warning results. The comprehensive safety control plan can be set up by professional and technical personnel according to the warning results, or a plan library can be built by building a corresponding relationship between the comprehensive safety control plan and the certified safety warning results and the independent safety warning results. Obtain a comprehensive safety control plan based on the plan library. The safety control verification of the power grid is carried out through the control verification cycle, and a safety control verification result is established, that is, within the set verification cycle, the effectiveness of the safety control plan is verified through field testing, and a safety control verification result is established, wherein the safety control verification result contains the verified normal parameters and abnormal parameters. Finally, the execution of the comprehensive safety control plan is corrected based on the plan library or professional and technical personnel through the safety control verification result.

[0033] The method provided in the embodiment of the present application also includes: establishing historical safety impact characteristics and time decay penalty factors of the power grid based on the historical maintenance information; establishing a safety concern database through the historical safety impact characteristics and time decay penalty factors; performing joint identification of real-time shared data sets through the safety concern database to establish a joint identification result; and revising the comprehensive safety management and control plan based on the joint identification result.

[0034] Based on the historical maintenance information, the historical safety impact characteristics and time-series decay penalty factors of the power grid are established. The time-series decay penalty factor is a coefficient used to evaluate the decay of the impact intensity of historical events over time. As time goes by, the impact of historical events on the current safety status gradually weakens, and the corresponding time-series decay penalty factor is smaller. The historical safety impact characteristics are key factors affecting the safety of the power grid extracted by analyzing the historical maintenance records and security event logs of the power grid. Specifically, by analyzing the historical maintenance information, key features affecting the safety of the power grid, such as fault causes, fault parameters, fault frequencies, etc., are extracted. According to different equipment types, the features are classified and sorted for subsequent analysis. A security concern database is established through the historical safety impact characteristics and the time-series decay penalty factor. Further, the real-time shared data set is jointly identified through the security concern database, and the parameters in the real-time shared data set are input into the security concern database to retrieve the fault parameters of the included equipment, obtain the fault parameters contained in the real-time shared data set, and obtain the corresponding faulty equipment, and establish a joint identification result. The comprehensive safety control plan is corrected with the joint identification result, that is, the joint identification result is transmitted to professional and technical personnel to correct the comprehensive safety control plan, thereby ensuring the stable operation of the power grid. For example, in a certain power grid area, historical maintenance records show that a substation has experienced multiple overload failures, and the frequency of equipment updates at the substation is low. When the real-time data set shows that the load of the substation is close to the critical value again, that is, when the historical fault load of the substation in the safety concern database is close to the real-time data set, an early warning is immediately triggered through joint identification, and the operation and maintenance personnel are guided to take load diversion measures first.

[0035] In the above, refer to Figure 1 The power grid security management method based on the twin sandbox according to the embodiment of the present invention is described in detail. Figure 2 A power grid security management and control system based on a twin sandbox according to an embodiment of the present invention is described.

[0036] According to the power grid security management and control system based on the twin sandbox according to the embodiment of the present invention, the technical problem that the power grid monitoring and management methods in the prior art have poor response capabilities for real-time monitoring and abnormal warning, resulting in the difficulty in improving the safety and stability of power grid operation is solved. The power grid security management and control method based on the twin sandbox, by introducing digital twin models, sensor networks, and association analysis networks, can timely capture abnormal conditions in the power grid and generate safety warnings, thereby improving the response capabilities of monitoring and abnormal warnings, facilitating technical personnel to quickly conduct safety management and control of the power grid, and ensuring the safety and stability of power grid operation. The power grid security management and control system based on the twin sandbox includes: a data node acquisition module 11, a twin model acquisition module 12, a first warning module 13, a local sub-model construction module 14, a second warning module 15, and a security management and control module 16.

[0037] The data node acquisition module 11 is used to evaluate the power grid structure and establish a data synchronization node according to the evaluation results of the power grid structure;

[0038] A twin model acquisition module 12 is used to build a digital twin model of the power grid based on component information in the power grid, wherein the digital twin model is established in a sandbox environment, and the component information includes specification information, status information, and historical maintenance information of the component;

[0039] A first warning module 13 is used to configure a sensor network at the data synchronization node, perform key parameter monitoring of the corresponding data synchronization node through the sensor network, establish a real-time shared data set, synchronize the real-time shared data set to the digital twin model, perform data update of the digital twin model, and generate a first safety warning result;

[0040] A local sub-model construction module 14 is used to obtain a power grid task of the power grid, perform data synchronization node aggregation based on the power grid task, deploy a local sub-model based on the aggregation result, and embed the local sub-model into the digital twin model;

[0041] The second warning module 15 is used to activate the dynamic window of the local sub-model to extract data stream features after the real-time shared data set is synchronized to the digital twin model, and to identify local anomalies of the feature extraction results with the local sub-model to establish a second safety warning result;

[0042] The security management and control module 16 is used to authenticate and identify the first security warning result and the second security warning result, and establish a comprehensive security management and control plan.

[0043] Below, the specific configuration of the second early warning module 15 will be described in detail. The second early warning module 15 may further include: the dynamic window of the activated local sub-model performs data stream feature extraction, and also includes: establishing a strategy space mapped to the local sub-model, the strategy space is the range space of the dynamic window; obtaining historical data based on the digital twin model, and setting the initial window size of the dynamic window of each local sub-model with the historical data; obtaining the node target in each local sub-model, and establishing the utility function of the local sub-model based on the node target; performing utility evaluation of the real-time shared data set through the utility function, and establishing a utility influence coefficient; updating the initial window size in the strategy space through the utility influence coefficient, and establishing a dynamic window according to the update result. The utility function is as follows: ;in, Characterization The utility function of a local sub-model, Characterization The total number of data synchronization nodes in the local sub-model, Represents any data synchronization node, Characterization The weight of the data synchronization node, Characterized in the Nodes at time The utility function of a real-time shared dataset is established by accuracy and timeliness, Characterized in the Nodes at time Real-time shared data, For the The data sharing influence coefficient of each node is Characterization Nodes at time The utility function of historical data, For the Nodes at time Historical data, For the The environmental impact weight of each node, For in time The response time, Characterization Nodes at time The prediction accuracy of Characterization The target influence weight of each node is Characterization Nodes at time The node objective utility function of Characterization Nodes at time target data. Below, the specific configuration of the data node acquisition module 11 will be described in detail. The data node acquisition module 11 may further include: the power grid structure evaluation of the power grid, establishing a data synchronization node according to the power grid structure evaluation result, and also include: establishing a centrality index of the node, the centrality index includes a degree centrality index, a closeness centrality index, and an betweenness centrality index; taking each component in the power grid as a basic node, performing a centrality evaluation of the basic node through the centrality index, and establishing a first node quantity influence according to the evaluation result; obtaining the attribute information of the component, and generating a second node quantity influence based on the attribute information; establishing a data synchronization node according to the first node quantity influence, the second node quantity influence and the basic node.

[0044] The specific configuration of the security control module 16 will be described in detail below. The security control module 16 further includes: the authentication and identification of the first security warning result and the second security warning result, and also includes: inputting the first security warning result and the second security warning result into the association analysis network, performing authentication support for the first security warning result and the second security warning result; outputting the security warning result whose authentication support meets the preset threshold as an authentication security warning result, and outputting the security warning result whose authentication support does not meet the preset threshold as an independent security warning result; completing authentication and identification according to the authentication security warning result and the independent security warning result.

[0045] The specific configuration of the security control module 16 will be described in detail below. The security control module 16 further includes: the establishment of a comprehensive security control plan, further including: configuring a control verification cycle based on the certified security warning results and the independent security warning results; after the comprehensive security control plan is established, the security control verification of the power grid is performed through the control verification cycle to establish a security control verification result; and the execution of the comprehensive security control plan is modified according to the security control verification result.

[0046] The specific configuration of the safety control module 16 will be described in detail below. The safety control module 16 further includes: establishing the historical safety impact characteristics and time-series attenuation penalty factors of the power grid based on the historical maintenance information; establishing a safety concern database through the historical safety impact characteristics and time-series attenuation penalty factors; performing joint identification of real-time shared data sets through the safety concern database to establish a joint identification result; and revising the comprehensive safety control plan with the joint identification result.

[0047] The power grid security management and control system based on the twin sandbox provided in an embodiment of the present invention can execute the power grid security management and control method based on the twin sandbox provided in any embodiment of the present invention, and has functional modules and beneficial effects corresponding to the execution method.

[0048] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, any number of different modules may be used and run on the user terminal and / or server, and the various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of the present invention.

[0049] The above specific implementations do not constitute a limitation on the protection scope of this application. It should be understood by those skilled in the art that various modifications, combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of this application should be included in the protection scope of this application.

Claims

1. A power grid security management and control method based on twin sandbox, characterized in that: The method comprises: Conduct grid structure evaluation on the power grid and establish data synchronization nodes based on the grid structure evaluation results; Building a digital twin model of the power grid based on component information in the power grid, wherein the digital twin model is built in a sandbox environment, and the component information includes specification information, status information, and historical maintenance information of the component; configuring a sensor network at the data synchronization node, performing key parameter monitoring of the corresponding data synchronization node through the sensor network, establishing a real-time shared data set, synchronizing the real-time shared data set to the digital twin model, performing data update of the digital twin model, and generating a first safety warning result; Acquire power grid tasks of the power grid, perform data synchronization node aggregation based on the power grid tasks, deploy a local sub-model based on the aggregation results, and embed the local sub-model into the digital twin model; After the real-time shared data set is synchronized to the digital twin model, the dynamic window of the local sub-model is activated to extract data stream features, and the local sub-model is used to identify local anomalies of the feature extraction results to establish a second safety warning result; Authenticate and identify the first security warning result and the second security warning result, and establish a comprehensive security management and control plan; The activating the dynamic window of the local sub-model to extract data stream features also includes: Establishing a strategy space mapped to the local sub-model, wherein the strategy space is a range space of the dynamic window; Acquire historical data based on the digital twin model, and set the initial window size of the dynamic window of each local sub-model with the historical data; Obtaining node targets in each local sub-model, and establishing a utility function of the local sub-model based on the node targets; Performing utility evaluation of the real-time shared data set through the utility function and establishing a utility impact coefficient; Updating the initial window size in the strategy space by using the utility impact coefficient, and establishing a dynamic window according to the update result; The utility function is as follows: Among them, U j (t) represents the utility function of the jth local sub-model, N j represents the total number of data synchronization nodes in the jth local sub-model, i represents any data synchronization node, α i Characterizes the weight of the i-th data synchronization node, f i (x i (t)) represents the utility function of the real-time shared dataset at the i-th node at time t, which is established by accuracy and timeliness, x i (t) represents the real-time shared data at the i-th node at time t, β i is the data sharing influence coefficient of the i-th node, g i (y i (t)) represents the historical data utility function of the i-th node at time t, y i (t) is the historical data of the ith node at time t, γ i is the environmental impact weight of the ith node, r i (t) is the response time at time t, p i (t) represents the prediction accuracy of the i-th node at time t, δ i Characterizes the target influence weight of the i-th node, k i (w i (t)) represents the node objective utility function of the i-th node at time t, w i (t) represents the target data of the i-th node at time t.

2. The power grid security management and control method based on twin sandbox according to claim 1, characterized in that: The performing of a grid structure evaluation on the grid and establishing a data synchronization node according to the grid structure evaluation result also includes: Establishing centrality indexes of nodes, wherein the centrality indexes include degree centrality index, closeness centrality index, and betweenness centrality index; Taking each component in the power grid as a basic node, evaluating the centrality of the basic node by using the centrality index, and establishing the first node quantity influence according to the evaluation result; Acquire attribute information of the component, and generate a second node quantity influence based on the attribute information; A data synchronization node is established according to the first node quantity impact, the second node quantity impact and the basic node.

3. The power grid security management and control method based on twin sandbox according to claim 1, characterized in that: The authenticating and identifying the first security warning result and the second security warning result also includes: Inputting the first security warning result and the second security warning result into the association analysis network, and performing authentication support for the first security warning result and the second security warning result; The security warning result of the authentication support degree meeting the preset threshold is output as the authentication security warning result, and the security warning result of the authentication support degree not meeting the preset threshold is output as the independent security warning result; Complete certification identification based on certification security warning results and independent security warning results.

4. The power grid security management and control method based on twin sandbox according to claim 3 is characterized in that: The establishment of a comprehensive safety management and control program also includes: Configure the control verification cycle based on the certification security warning results and independent security warning results; After the comprehensive safety control plan is established, the safety control verification of the power grid is carried out through the control verification cycle, and the safety control verification result is established; The implementation of the comprehensive security control plan is modified based on the security control verification results.

5. The power grid security management and control method based on twin sandbox according to claim 1, characterized in that: The method further comprises: Establishing historical safety impact characteristics and time-series attenuation penalty factors of the power grid based on the historical maintenance information; Establishing a safety concern database through the historical safety impact characteristics and the time-series attenuation penalty factor; Perform joint identification of the real-time shared data set through the security concern database, and establish a joint identification result; The comprehensive security management and control plan is revised based on the joint identification results.

6. The power grid security management and control system based on twin sandbox is characterized by: The system is used to execute the method according to any one of claims 1 to 5, and the system comprises: A data node acquisition module is used to evaluate the power grid structure and establish a data synchronization node based on the evaluation results of the power grid structure; A twin model acquisition module is used to build a digital twin model of the power grid based on component information in the power grid, wherein the digital twin model is established in a sandbox environment, and the component information includes specification information, status information, and historical maintenance information of the component; A first early warning module is used to configure a sensor network at the data synchronization node, perform key parameter monitoring of the corresponding data synchronization node through the sensor network, establish a real-time shared data set, synchronize the real-time shared data set to the digital twin model, perform data update of the digital twin model, and generate a first safety early warning result; A local sub-model construction module is used to obtain power grid tasks of the power grid, perform data synchronization node aggregation based on the power grid tasks, deploy a local sub-model based on the aggregation results, and embed the local sub-model into the digital twin model; A second warning module is used to activate the dynamic window of the local sub-model to extract data stream features after the real-time shared data set is synchronized to the digital twin model, and to identify local anomalies of the feature extraction results with the local sub-model to establish a second safety warning result; The security management and control module is used to authenticate and identify the first security warning result and the second security warning result and establish a comprehensive security management and control plan.

Images