A fast power grid restoration method based on self-healing fault diagnosis

Abstract

The present invention discloses a method for rapid power grid recovery under self-healing fault diagnosis, which relates to the technical field related to power supply systems. Based on the power grid topology of the target power grid, a power grid twin model is constructed. The self-healing capability of the power grid twin model is evaluated, and the power grid twin model is hierarchically divided according to the self-healing capability evaluation results to obtain a hierarchical power grid twin model. The fault self-healing strategy is configured from bottom to top according to the hierarchical power grid twin model. The power grid twin model is connected to the target power grid data, the power grid data is collected in real time and updated synchronously, and the pre-built fault diagnosis model is called to perform fault diagnosis to obtain the fault diagnosis result. Strategy matching is performed in the fault self-healing strategy, and the corresponding fault self-healing strategy cluster is extracted to perform self-healing treatment of the power grid fault. The technical problem that the power grid self-healing fault diagnosis and recovery method in the prior art is slow to respond and low in efficiency, resulting in difficulty in improving the fault recovery speed of the power grid and the power supply reliability is solved.

Description

Technical Field

[0001] The present application relates to the technical field related to power supply systems, and in particular to a method for rapid power grid recovery under self-healing fault diagnosis. Background Art

[0002] As modern power grids continue to grow in scale and complexity, ensuring the reliability and continuity of power systems has become a major challenge. Traditional power grids rely primarily on manual intervention and independent system operations to address issues such as equipment failures, line faults, and external interference. This is not only time-consuming and labor-intensive, but also makes it difficult to prevent fault spread or quickly restore power. The emergence of digital twin technology has opened up new possibilities for real-time simulation and analysis of power grids. By constructing a digital twin model of the power grid, the operating status and characteristics of the grid can be simulated in a virtual space, providing a virtual environment for predicting and diagnosing faults. However, existing digital twin power grid models lack a systematic approach to fault diagnosis and self-healing strategy configuration, and are unable to fully utilize the grid's hierarchical structure and self-healing capabilities, resulting in inefficient fault handling and slow grid recovery.

[0003] Therefore, in the prior art, the power grid self-healing fault diagnosis and recovery method has slow response and low efficiency, resulting in technical problems such as difficulty in improving the power grid fault recovery speed and power supply reliability. Summary of the Invention

[0004] This application provides a method for rapid grid recovery under self-healing fault diagnosis, addressing the technical issues of existing grid self-healing fault diagnosis and recovery methods, which are slow to respond and inefficient, resulting in difficulties in improving grid fault recovery speed and power supply reliability. By constructing a digital twin model of the grid, evaluating self-healing capabilities and performing hierarchical divisions, and combining simulation verification of self-healing strategies, rapid diagnosis and precise self-healing of grid faults can be achieved, improving the grid's fault response speed and power supply reliability.

[0005] The present application provides a method for rapid power grid recovery under self-healing fault diagnosis, the method comprising: constructing a power grid twin model based on the power grid topology of the target power grid. Performing a self-healing capability evaluation on the power grid twin model, and hierarchically dividing the power grid twin model according to the self-healing capability evaluation result to obtain a hierarchical power grid twin model, wherein the hierarchical power grid twin model includes multiple hierarchical power grid twin models, and the multiple hierarchical power grid twin models correspond to different fault levels. Performing a fault self-healing strategy configuration from bottom to top according to the hierarchical power grid twin model, wherein the fault self-healing strategy includes a cascaded multi-level fault self-healing strategy cluster. Performing a docking of the power grid twin model with the target power grid data, collecting power grid data in real time and updating it synchronously, and calling a pre-built fault diagnosis model to perform fault diagnosis to obtain a fault diagnosis result, wherein the fault diagnosis result includes the fault location, fault level, and fault category. Based on the fault diagnosis result, performing strategy matching in the fault self-healing strategy, extracting the corresponding fault self-healing strategy cluster, and performing self-healing treatment of the power grid fault.

[0006] In one implementation, a grid twin model is constructed based on the target grid's grid topology. This includes interacting with the target grid's grid management terminal to collect grid device information and device connection information from the target grid. A baseline grid topology is constructed based on the device connection information, and node and edge definitions for the baseline grid topology are performed in conjunction with the grid device information to obtain the grid twin model.

[0007] In an implementation, the self-healing capability of the power grid twin model is evaluated, and the power grid twin model is hierarchically divided according to the self-healing capability evaluation result to obtain a hierarchical power grid twin model, including: traversing the power grid twin model and obtaining multiple control nodes based on the node definition information. According to the self-healing processing capability and self-healing processing range of the multiple control nodes, combined with the preset control hierarchy definition, the node hierarchy of the multiple control nodes is determined. Hierarchical division is performed according to the node hierarchy to obtain multiple hierarchical power grid twin models, wherein each hierarchical power grid twin model includes a control node of the corresponding node hierarchy and the subordinate device nodes.

[0008] In the implementation method, the fault self-healing strategy is configured from the bottom up according to the hierarchical power grid twin model, including: the grid management end of the interactive target power grid calls multiple fault handling plans, and performs self-healing feasibility verification on the multiple fault handling plans, and extracts multiple self-healing handling plans. The multiple self-healing handling plans are processed into statements, and the statement processing results are configured as multiple initial strategy groups, the multiple initial strategy groups correspond to multiple fault categories, and each of the initial strategy groups includes multi-level strategy statements. With the multiple hierarchical power grid twin models as the configuration target, the multiple initial strategy groups are matched and called from the bottom up, and the corresponding output is a multi-level fault self-healing strategy cluster, which is stored as the fault self-healing strategy.

[0009] In an implementation, with multiple hierarchical power grid twin models as configuration targets, multiple initial policy groups are matched and called from the bottom up, with the corresponding output being multiple levels of fault self-healing policy clusters, which are stored as the fault self-healing policies. This includes: calling the first-level power grid twin model of the hierarchical power grid twin model from the bottom up. Parsing and extracting the device identity information of multiple device nodes in the first-level power grid twin model. Based on a device fault knowledge base, the fault feature set of the first-level power grid twin model is defined in combination with the device identity information, wherein the fault feature set includes multiple fault type information and corresponding multiple fault level information. Using the multiple fault type information as a first index constraint and the multiple fault level information as a second index constraint, policy statements are called from the multiple initial policy groups, with the output being a first fault self-healing policy cluster. Traversing the multiple hierarchical power grid twin models of the hierarchical power grid twin model from the bottom up to obtain multiple levels of fault self-healing policy clusters, wherein each level of the fault self-healing policy cluster includes multiple fault self-healing policy clusters. Cascading multiple levels of the fault self-healing strategy clusters to obtain the fault self-healing strategy.

[0010] In an implementation, multiple hierarchical power grid twin models of the hierarchical power grid twin model are traversed from bottom to top to obtain multiple levels of fault self-healing strategy clusters. Thereafter, the method further includes: hierarchically combining the multiple levels of fault self-healing strategy clusters according to the hierarchical relationship of the multiple hierarchical power grid twin models to determine multiple hierarchical groups, wherein the hierarchical groups include hierarchically adjacent sub-level fault self-healing strategy clusters and parent-level fault self-healing strategy clusters. The multiple hierarchical groups are traversed to obtain the intersection strategy of the sub-level fault self-healing strategy clusters and the parent-level fault self-healing strategy clusters, and the strategy of the parent-level fault self-healing strategy cluster is simplified using the intersection strategy.

[0011] In this implementation, a fault self-healing strategy is configured from the bottom up based on the hierarchical power grid twin model. Subsequently, the method further includes obtaining homologous historical fault records and intrinsic historical fault records of the target power grid. Based on the homologous historical fault records, the intrinsic historical fault records are mutated and expanded to generate a fault sample set. Random selection is performed within the fault sample set, and simulation verification of the fault self-healing strategy is performed based on the random selection results.

[0012] In an implementation, the method further includes: obtaining, based on the hierarchical power grid twin model, a plurality of subnet self-healing control units of the plurality of hierarchical power grid twin models, establishing an association mapping between the fault self-healing strategy and the plurality of subnet self-healing control units, and issuing the fault self-healing strategy for local deployment based on the association mapping.

[0013] It is intended to construct a power grid twin model based on the power grid topology of the target power grid through the method for rapid power grid recovery under self-healing fault diagnosis proposed in this application. The self-healing capability of the power grid twin model is evaluated, and the power grid twin model is hierarchically divided according to the self-healing capability evaluation result to obtain a hierarchical power grid twin model, wherein the hierarchical power grid twin model includes multiple hierarchical power grid twin models, and the multiple hierarchical power grid twin models correspond to different fault levels. The fault self-healing strategy is configured from bottom to top according to the hierarchical power grid twin model, wherein the fault self-healing strategy includes a cascaded multi-level fault self-healing strategy cluster. The power grid twin model is connected to the target power grid data, the power grid data is collected in real time and updated synchronously, and the pre-built fault diagnosis model is called to perform fault diagnosis to obtain the fault diagnosis result, wherein the fault diagnosis result includes the fault location, fault level, and fault category. Based on the fault diagnosis result, strategy matching is performed in the fault self-healing strategy, the corresponding fault self-healing strategy cluster is extracted, and the self-healing treatment of the power grid fault is performed. This technology addresses the technical issues of existing grid self-healing fault diagnosis and recovery methods, which suffer from slow response and low efficiency, hindering the improvement of grid fault recovery speed and power supply reliability. By building a digital twin model of the power grid, evaluating self-healing capabilities and stratifying them, combined with simulation verification of self-healing strategies, this technology enables rapid diagnosis and precise self-healing of grid faults, improving grid fault response speed and power supply reliability. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] To more clearly illustrate the technical solutions of the embodiments of the present disclosure, the accompanying drawings of the embodiments of the present disclosure are briefly introduced below. Flowcharts are used in this application to illustrate the operations performed by the systems according to the embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed in precise order. Instead, various steps may be processed in reverse order or simultaneously as needed. Furthermore, other operations may be added to these processes, or one or more operations may be removed from these processes.

[0015] Figure 1 A schematic diagram of a flow chart of a method for rapid power grid recovery under self-healing fault diagnosis provided in an embodiment of the present application;

[0016] Figure 2 A flow chart of a method for rapid grid recovery under self-healing fault diagnosis provided in an embodiment of the present application for constructing a grid twin model. DETAILED DESCRIPTION

[0017] 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.

[0018] In order to make the purpose, technical solutions and advantages of this application clearer, the application will be further described in detail below with reference to the accompanying drawings. The described embodiments should not be regarded as limiting this application. All other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0019] In the following description, reference is made to “some embodiments”, which describes a subset of all possible embodiments, but it will be understood that “some embodiments” may be the same subset or different subsets of all possible embodiments and may be combined with each other without conflict, and the terms “first\second” involved are merely used to distinguish similar objects and do not represent a specific ordering of the 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 is not necessarily limited to those steps or units that are clearly listed, but may include other steps or modules that are not clearly listed or that are inherent to these processes, methods, products, or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which this application belongs. The terms used herein are for the purpose of describing the embodiments of this application only.

[0020] The present application provides a method for rapid power grid recovery under self-healing fault diagnosis, such as Figure 1 As shown, the method includes:

[0021] Based on the grid topology of the target power grid, a grid twin model is constructed. The grid twin model is evaluated for its self-healing capabilities and, based on the self-healing capabilities evaluation results, the grid twin model is hierarchically divided to obtain a hierarchical grid twin model. The hierarchical grid twin model includes multiple hierarchical grid twin models, each corresponding to different fault levels. A fault self-healing strategy is configured from the bottom up based on the hierarchical grid twin model. The fault self-healing strategy includes a cascaded multi-level fault self-healing strategy cluster.

[0022] Based on the target power grid's grid topology (which refers to the connections and layout between various devices in the grid, such as generators, transformers, and transmission lines), a grid twin model is constructed. This grid twin model, also known as a digital twin model, is a virtual representation of the actual grid, capable of simulating the grid's operating status and characteristics in a digital space. Subsequently, the grid twin model undergoes a self-healing capability assessment and, based on the assessment results, is hierarchically divided to obtain a hierarchical grid twin model. The hierarchical grid twin model includes multiple hierarchical grid twin models, each corresponding to different fault levels. Furthermore, fault self-healing strategies are configured from the bottom up based on the hierarchical grid twin model, ensuring that each grid twin model layer has a corresponding fault self-healing strategy. The fault self-healing strategies include cascaded multi-level fault self-healing strategy clusters, which are multiple hierarchical fault self-healing strategy clusters.

[0023] like Figure 2 As shown, the method provided in the embodiment of the present application further includes: interacting with the grid management terminal of the target grid to collect grid equipment information and device connection information of the target grid. Based on the device connection information, a baseline grid topology is constructed, and nodes and edges of the baseline grid topology are defined in combination with the grid equipment information to obtain the grid twin model.

[0024] Based on the target grid's grid topology, a grid twin model is constructed. This includes interacting with the target grid's grid management terminal to collect grid device information and device connection information. The grid device information contains detailed information about all devices in the grid, including device type, parameters, location, and operating status. The device connection information describes the connection relationships between devices in the grid, namely, which devices are connected to each other and how they are connected. Subsequently, a baseline grid topology is constructed based on the device connection information. This baseline grid topology, consisting solely of connection relationships and layout, is constructed based on the device connection information. Node and edge definitions for the baseline grid topology are then performed in conjunction with the grid device information. Detailed attributes, such as device type, parameters, and operating status, are added to each node in the baseline grid topology, and attributes, such as line parameters and switch status, are added to each edge. Furthermore, the model can be further refined based on special devices and connection methods, such as ring networks, parallel lines, and series reactors, to ensure its accuracy and completeness. After the definition is complete, the grid twin model is obtained, allowing the physical structure of the grid and device attributes to be fully mapped and presented in the digital space.

[0025] The method provided in an embodiment of the present application further includes: traversing the power grid twin model and obtaining a plurality of control nodes based on the node definition information. Determining the node hierarchy of the plurality of control nodes based on the self-healing processing capabilities and self-healing processing ranges of the plurality of control nodes, combined with a preset control hierarchy definition. Performing hierarchical division based on the node hierarchy to obtain a plurality of hierarchical power grid twin models, wherein each hierarchical power grid twin model includes a control node of the corresponding node hierarchy and its subordinate device nodes.

[0026] The self-healing capability of the power grid twin model is evaluated, and the power grid twin model is hierarchically divided according to the self-healing capability evaluation result to obtain a hierarchical power grid twin model, including: traversing the power grid twin model, and obtaining multiple control nodes based on the node definition information. The control nodes are power grid nodes with control and management capabilities in the node definition, and can perform self-healing processing operations, such as substation control centers, smart switches, etc. Further, based on the self-healing processing capabilities and self-healing processing ranges of the multiple control nodes, combined with the preset control hierarchy definition, the node hierarchy of the multiple control nodes is determined. The self-healing processing capability is the type of fault that the control node can handle and the severity of the fault that can be handled (such as the ability to control harmonics, the ability to filter output ripple, the ability to correct power factor, etc.). The self-healing processing range is the range of the power grid that the control node can regulate, including the number of devices, geographical range, etc. The control hierarchy is defined as a pre-set correspondence between the hierarchy and the self-healing capability and range. Different self-healing capabilities and ranges correspond to node hierarchies. The higher the node hierarchy, the stronger the corresponding node control range and self-healing capability. Finally, a hierarchical division is performed based on the node hierarchy. That is, a hierarchical division is performed based on the node hierarchy of each control node. The power grid range that can be controlled by the control node is regarded as a whole hierarchy, and multiple hierarchical power grid twin models are obtained. Each hierarchical power grid twin model includes the control node of the corresponding node hierarchy and the device nodes under its jurisdiction.

[0027] The method provided in the embodiment of the present application also includes: the grid management end of the interactive target power grid calls multiple fault handling plans, and performs self-healing feasibility verification on the multiple fault handling plans, and extracts multiple self-healing handling plans. The multiple self-healing handling plans are processed in a sentence-based manner, and the sentence-based processing results are configured as multiple initial policy groups, the multiple initial policy groups correspond to multiple fault categories, and each of the initial policy groups includes multi-level policy statements. With the multiple hierarchical power grid twin models as the configuration target, the multiple initial policy groups are matched and called from the bottom up, and the corresponding output is a multi-level fault self-healing policy cluster, which is stored as the fault self-healing strategy.

[0028] A bottom-up fault self-healing strategy is configured based on the hierarchical power grid twin model, including interacting with the target power grid's grid management terminal, recording historical fault handling solutions on the target power grid's grid management terminal, and invoking multiple historical handling solutions for multiple faults on the grid management terminal, i.e., multiple fault handling plans. Each fault handling plan corresponds to a fault category, and self-healing feasibility verification is performed on each of the multiple fault handling plans. During the self-healing feasibility verification, the operations in the plan are analyzed one by one to determine whether they meet the conditions for automated execution, and the fault handling effect is obtained. If the fault handling effect and operation steps of the fault handling plan meet the conditions for automated execution, the self-healing plan is retained; otherwise, the self-healing plan is discarded, thereby extracting multiple self-healing plans. Furthermore, the multiple self-healing plans are sentence-based, i.e., the natural language descriptions are converted into structured policy statements that can be parsed and executed by the system. The sentence-based processing results are configured as multiple initial policy groups, each of which corresponds to multiple fault categories, and each initial policy group includes multiple levels of policy statements. Furthermore, with multiple hierarchical power grid twin models as configuration targets, multiple initial policy groups are matched and called from bottom to top. According to the corresponding hierarchical levels of the power grid twin models, the policy statements in the initial policy groups are matched to the corresponding models from the lowest level to the highest level, so that each level has a corresponding initial policy group. The corresponding output is a multi-level fault self-healing policy cluster, which is stored as the fault self-healing policy, thereby facilitating the acquisition of the corresponding policy during subsequent fault handling.

[0029] The method provided in the embodiment of the present application also includes: calling the first-level power grid twin model of the hierarchical power grid twin model from bottom to top. Parsing and extracting the device identity information of multiple device nodes in the first-level power grid twin model. Based on the equipment fault knowledge base, the fault feature set of the first-level power grid twin model is defined in combination with the device identity information, wherein the fault feature set includes multiple fault type information and corresponding multiple fault level information. Using multiple fault type information as the first index constraint and multiple fault level information as the second index constraint, policy statement calls are performed in multiple initial policy groups, and the output is a first fault self-healing strategy cluster. Traverse the multiple hierarchical power grid twin models of the hierarchical power grid twin model from bottom to top to obtain multiple levels of the fault self-healing strategy clusters, wherein each level of the fault self-healing strategy cluster includes multiple fault self-healing strategy clusters. Cascade multiple levels of the fault self-healing strategy clusters to obtain the fault self-healing strategy.

[0030] Using multiple hierarchical power grid twin models as configuration targets, a bottom-up matching call is performed on multiple initial policy groups, with the corresponding output being a multi-level fault self-healing policy cluster, which is stored as the fault self-healing policy. The method includes: calling the first-level power grid twin model of the hierarchical power grid twin model from the bottom up. The first-level power grid twin model is the lowest-level power grid twin model, typically corresponding to the device layer, including a single device or a small range of control nodes. Subsequently, device identity information is parsed and extracted for multiple device nodes in the first-level power grid twin model. The device identity information is information that uniquely identifies the device node, including device ID, type, model, location, etc. Furthermore, based on a device fault knowledge base, which contains a database of information such as fault modes, fault types, and fault characteristics for various devices, a fault feature set for the first-level power grid twin model is defined in combination with the device identity information. Specifically, the fault feature set for the first-level power grid twin model is defined by obtaining corresponding fault features from the device fault knowledge base based on the device identity information. The fault feature set includes multiple fault type information and corresponding multiple fault level information. Furthermore, using multiple fault type information as the first index constraint and multiple fault level information as the second index constraint, the policy statement is called in the initial policy group using the fault type information and fault level information to ensure that the subsequently acquired fault policy statement completely matches the fault, thereby outputting the first fault self-healing policy cluster. Finally, the multiple hierarchical power grid twin models of the hierarchical power grid twin model are traversed from bottom to top to obtain multiple levels of fault self-healing policy clusters, with the fault self-healing policy clusters corresponding to the hierarchical power grid twin models. Each level of the fault self-healing policy cluster includes multiple fault self-healing policy clusters.

[0031] The method provided in an embodiment of the present application further includes: hierarchically combining the multiple levels of fault self-healing strategy clusters based on the hierarchical relationship of the multiple hierarchical power grid twin models to determine multiple hierarchical groups, wherein the hierarchical groups include hierarchically adjacent child-level fault self-healing strategy clusters and parent-level fault self-healing strategy clusters. The multiple hierarchical groups are traversed to obtain an intersection strategy of the child-level fault self-healing strategy clusters and the parent-level fault self-healing strategy clusters, and the strategy of the parent-level fault self-healing strategy cluster is simplified using the intersection strategy.

[0032] The method further comprises: traversing the multiple hierarchical power grid twin models of the hierarchical power grid twin model from bottom to top to obtain the multiple fault self-healing strategy clusters. The method further comprises: hierarchically combining the multiple fault self-healing strategy clusters according to the hierarchical relationship of the multiple hierarchical power grid twin models to determine multiple hierarchical groups, each of which is composed of an original hierarchical level and adjacent sub-levels and adjacent parent levels of the original hierarchical level. The hierarchical group includes hierarchically adjacent sub-level fault self-healing strategy clusters and parent-level fault self-healing strategy clusters. The method traverses the multiple hierarchical groups to obtain the intersection strategy of the sub-level fault self-healing strategy clusters and the parent-level fault self-healing strategy clusters, and uses the intersection strategy to simplify the strategy of the parent-level fault self-healing strategy cluster. That is, by eliminating the corresponding strategies in the original hierarchical fault self-healing strategy cluster and the sub-level fault self-healing strategy cluster for the strategies that intersect the fault self-healing strategy clusters in the hierarchical group, retaining the strategies of the parent-level fault self-healing strategy cluster, and thus completing the simplification of the strategies of the parent-level fault self-healing strategy cluster.

[0033] The method provided in an embodiment of the present application further includes: obtaining homologous historical fault records and intrinsic historical fault records of the target power grid; mutating and expanding the intrinsic historical fault records based on the homologous historical fault records to generate a fault sample set; randomly selecting from the fault sample set, and performing simulation verification of the fault self-healing strategy based on the random selection results.

[0034] The fault self-healing strategy is configured from the bottom up according to the hierarchical power grid twin model, and then further includes: obtaining the same-source historical fault records and intrinsic historical fault records of the target power grid, wherein the same-source historical fault records are historical fault data of other power grids with the same topology and equipment type as the target power grid. The intrinsic historical fault records are the historical fault data of the target power grid itself, including various fault information that occurred in the past. Based on the same-source historical fault records, the intrinsic historical fault records are mutated and expanded, that is, the intrinsic historical fault records are mutated and expanded according to the same-source historical fault records to generate more fault samples to enrich the fault sample set and generate a fault sample set. Random selection is performed on the fault sample set, and simulation verification of the fault self-healing strategy is performed based on the random selection results.

[0035] The grid twin model is connected to the target grid data, grid data is collected and updated in real time, and a pre-built fault diagnosis model is used to perform fault diagnosis and obtain fault diagnosis results, including fault location, fault level, and fault category. Based on the fault diagnosis results, a strategy is matched with the fault self-healing strategy, and the corresponding fault self-healing strategy cluster is extracted to perform self-healing treatment of the grid fault.

[0036] The grid twin model is connected to the target grid data, grid data is collected and updated in real time, and real-time operational data from the actual grid is input into the grid twin model to reflect the current grid status. A pre-built fault diagnosis model is then invoked for fault diagnosis. This fault diagnosis model, built based on machine learning or an expert system, can quickly identify the location, level, and category of grid faults in the grid twin model. A fault diagnosis result is obtained, including the fault location, level, and category. Finally, based on the fault diagnosis result, the fault location, level, and category are determined. A strategy matching is performed within the fault self-healing strategy, and the corresponding fault self-healing strategy cluster is extracted to perform self-healing of the grid fault. This solves the technical problem of slow response and low efficiency in prior art grid self-healing fault diagnosis and recovery methods, which hinders improvements in grid fault recovery speed and power supply reliability. By constructing a grid digital twin model, evaluating self-healing capabilities, and performing hierarchical divisions, combined with simulation verification of self-healing strategies, rapid diagnosis and accurate self-healing of grid faults are achieved, improving grid fault response speed and power supply reliability.

[0037] The method provided in an embodiment of the present application further includes: obtaining, based on the hierarchical power grid twin model, a plurality of subnet self-healing control units of the plurality of hierarchical power grid twin models, establishing an association mapping between the fault self-healing strategy and the plurality of subnet self-healing control units, and issuing the fault self-healing strategy for local deployment based on the association mapping.

[0038] According to the hierarchical power grid twin model, multiple subnet self-healing control units of multiple hierarchical power grid twin models are obtained, and the subnet self-healing control units are self-healing control units that are lower than the hierarchical power grid twin model level and are associated with the hierarchical power grid twin model. Further, an association mapping is established between the fault self-healing strategy and the multiple subnet self-healing control units, that is, a corresponding relationship is established between the fault self-healing strategy and the subnet self-healing control units, which strategies are executed by which control units, and the fault self-healing strategy is issued according to the association mapping for local deployment, that is, the fault self-healing strategy is sent to the corresponding subnet self-healing control unit through the communication network, and the fault self-healing strategy is configured and implemented in the on-site control unit to enable it to have the ability to execute the strategy.

[0039] The technical solution provided by the embodiments of the present invention constructs a power grid twin model based on the target power grid topology. The self-healing capability of the power grid twin model is evaluated and, based on the evaluation results, the model is hierarchically divided to obtain a hierarchical power grid twin model. Fault self-healing strategies are configured from the bottom up based on the hierarchical power grid twin model. The power grid twin model is connected to the target power grid data, collecting and updating grid data in real time. A pre-built fault diagnosis model is then invoked to perform fault diagnosis and obtain fault diagnosis results. Based on the fault diagnosis results, a strategy is matched against the fault self-healing strategies, and the corresponding fault self-healing strategy cluster is extracted to perform self-healing of the power grid fault. This solves the technical problem of slow response and low efficiency in prior art power grid self-healing fault diagnosis and recovery methods, which hinders improvements in power grid fault recovery speed and power supply reliability. By constructing a power grid digital twin model, evaluating self-healing capabilities, and performing hierarchical division, combined with simulation verification of self-healing strategies, rapid diagnosis and accurate self-healing of power grid faults are achieved, improving the power grid's fault response speed and power supply reliability.

[0040] Note that the above are only preferred embodiments of the present invention and the technical principles employed. Those skilled in the art will understand that the present invention is not limited to the specific embodiments described herein, and that various obvious changes, readjustments, and substitutions can be made by those skilled in the art without departing from the scope of protection of the present invention. Therefore, although the present invention has been described in detail through the above embodiments, the present invention is not limited to the above embodiments and may include many other equivalent embodiments without departing from the concept of the present invention. The scope of the present invention is determined by the scope of the appended claims.

Claims

1. A method for rapid power grid recovery under self-healing fault diagnosis, characterized in that: The method comprises: Build a grid twin model based on the grid topology of the target grid; Performing a self-healing capability evaluation on the power grid twin model, and hierarchically dividing the power grid twin model according to the self-healing capability evaluation result to obtain a hierarchical power grid twin model, wherein the hierarchical power grid twin model includes multiple hierarchical power grid twin models, and the multiple hierarchical power grid twin models correspond to different fault levels; Performing fault self-healing strategy configuration from bottom to top according to the hierarchical power grid twin model, wherein the fault self-healing strategy includes a cascaded multi-level fault self-healing strategy cluster; Interfacing the power grid twin model with target power grid data, collecting and updating power grid data in real time, and calling a pre-built fault diagnosis model to perform fault diagnosis and obtain fault diagnosis results, wherein the fault diagnosis results include fault location, fault level, and fault category; Based on the fault diagnosis result, strategy matching is performed in the fault self-healing strategy, and the corresponding fault self-healing strategy cluster is extracted to perform self-healing treatment of the power grid fault.

2. The method for rapid power grid recovery under self-healing fault diagnosis according to claim 1, characterized in that: Based on the grid topology of the target grid, a grid twin model is constructed, including: Interact with the grid management terminal of the target grid to collect grid equipment information and equipment connection information of the target grid; A baseline power grid topology is constructed based on the device connection information, and node definition and edge definition of the baseline power grid topology are performed in combination with the power grid device information to obtain the power grid twin model.

3. The method for rapid power grid recovery under self-healing fault diagnosis according to claim 2, characterized in that: Performing a self-healing capability evaluation on the power grid twin model, and hierarchically dividing the power grid twin model according to the self-healing capability evaluation result to obtain a hierarchical power grid twin model, including: Traversing the power grid twin model, and acquiring a plurality of control nodes based on node definition information; Determining the node levels of the plurality of control nodes according to the self-healing processing capabilities and self-healing processing ranges of the plurality of control nodes in combination with a preset control level definition; Hierarchical division is performed according to the node level to obtain multiple hierarchical power grid twin models, wherein each hierarchical power grid twin model includes a control node of the corresponding node level and the subordinate device nodes.

4. The method for rapid power grid recovery under self-healing fault diagnosis according to claim 3, characterized in that: The fault self-healing strategy is configured from the bottom up based on the hierarchical power grid twin model, including: The grid management terminal of the interactive target grid calls multiple fault handling plans, verifies the feasibility of self-healing of the multiple fault handling plans, and extracts multiple self-healing handling plans; Performing statement processing on the plurality of self-healing disposal plans, and configuring the statement processing results as a plurality of initial policy groups, wherein the plurality of initial policy groups correspond to a plurality of fault categories, and each of the initial policy groups includes a multi-level policy statement; Taking multiple hierarchical power grid twin models as configuration targets, multiple initial strategy groups are matched and called from bottom to top, and the corresponding output is a multi-level fault self-healing strategy cluster, which is stored as the fault self-healing strategy.

5. The method for rapid power grid recovery under self-healing fault diagnosis according to claim 4, characterized in that: Taking the multiple hierarchical power grid twin models as configuration targets, matching and calling the multiple initial strategy groups from bottom to top, and outputting the corresponding multi-level fault self-healing strategy clusters, which are stored as the fault self-healing strategies, include: Calling the first-level power grid twin model of the hierarchical power grid twin model from bottom to top; Parsing and extracting device identity information of multiple device nodes in the first-level power grid twin model; Based on the device fault knowledge base and in combination with the device identity information, a fault feature set of the first-level power grid twin model is defined, wherein the fault feature set includes multiple fault type information and corresponding multiple fault level information; Using the plurality of fault type information as a first index constraint and the plurality of fault level information as a second index constraint, calling a policy statement in the plurality of initial policy groups, and outputting a first fault self-healing policy cluster; Traversing the multiple hierarchical power grid twin models of the hierarchical power grid twin model from bottom to top to obtain multiple levels of fault self-healing strategy clusters, wherein each level of the fault self-healing strategy cluster includes multiple fault self-healing strategy sub-clusters; Cascading multiple levels of the fault self-healing strategy clusters to obtain the fault self-healing strategy.

6. The method for rapid power grid recovery under self-healing fault diagnosis according to claim 5, characterized in that: Traversing the multiple hierarchical power grid twin models of the hierarchical power grid twin model from bottom to top to obtain the multi-level fault self-healing strategy clusters, and then further comprising: According to the hierarchical relationship of the multiple hierarchical power grid twin models, the multiple levels of fault self-healing strategy clusters are hierarchically combined to determine multiple hierarchical groups, wherein the hierarchical groups include hierarchically adjacent child-level fault self-healing strategy clusters and parent-level fault self-healing strategy clusters; Traversing the plurality of hierarchical groups, obtaining an intersection strategy of the child-level fault self-healing strategy cluster and the parent-level fault self-healing strategy cluster, and simplifying the strategy of the parent-level fault self-healing strategy cluster using the intersection strategy.

7. The method for rapid power grid recovery under self-healing fault diagnosis according to claim 1, characterized in that: The fault self-healing strategy is configured from the bottom up according to the hierarchical power grid twin model, and then the following is included: Obtain the target power grid's historical fault records and intrinsic historical fault records; mutating and expanding the intrinsic historical fault records based on the same-source historical fault records to generate a fault sample set; Random selection is performed on the fault sample set, and simulation verification of the fault self-healing strategy is performed based on the random selection result.

8. The method for rapid power grid recovery under self-healing fault diagnosis according to claim 1, characterized in that: The method further comprises: According to the hierarchical power grid twin model, a plurality of subnet self-healing control units of the hierarchical power grid twin model are obtained; An association mapping between the fault self-healing strategy and the plurality of subnet self-healing control units is established, and the fault self-healing strategy is issued according to the association mapping for local deployment.

Images