A power communication network maintenance simulation checking method and system based on digital twinning

Abstract

This invention discloses a simulation verification method and system for power communication network maintenance based on digital twins, relating to the field of intelligent operation and maintenance technology for power and communication networks. It solves the problems of existing technologies where maintenance approval relies on manual experience, lacks dynamic simulation methods, and cannot assess the risk of superimposed faults. The key technical points are: converting the pending power communication maintenance application form into a graph operation command; inputting the graph operation command into a pre-constructed digital twin model for topology simulation, outputting the maintenance-state network topology; wherein the digital twin model is constructed from the full topology graph of the power communication network; under the maintenance-state network topology, performing a full-network static connectivity analysis, traversing and simulating the network survivability after a single fault occurs on the remaining critical links, and calculating the maintenance risk index by combining the voltage level and type weight of the carried services; comparing the maintenance risk index with a preset safety threshold, and outputting the simulation verification result.

Description

Technical Field

[0001] This invention relates to the field of intelligent operation and maintenance technology for power grids and communication networks, and more specifically, to a method and system for power and communication network maintenance simulation and verification based on digital twins. Background Technology

[0002] The power communication network is the "nervous system" for the safe and stable operation of the power grid, carrying out key production operations such as relay protection, stability control, and dispatch automation. With the advancement of the construction of new power systems, the maintenance work of communication networks (such as fiber optic cable relocation, equipment board upgrades, and the decommissioning of old equipment) is becoming increasingly frequent.

[0003] Currently, the review and arrangement of power communication maintenance plans mainly suffer from the following pain points:

[0004] Unstructured data barriers: Maintenance requests are usually processed in the form of text work orders, which contain complex descriptions that are difficult for computer systems to understand directly. They rely on dispatchers to read and translate them manually, which is inefficient and prone to errors.

[0005] Lack of dynamic simulation methods: Existing communication management systems typically only display the current network status. When approving maintenance tickets, dispatchers can only rely on experience to judge whether maintenance operations should directly cut off services. However, extreme risks often occur in a "superposition state," where maintenance leads to a reduction in network redundancy (N-1), and if a random fault occurs at this time (Risk N-1), it may cause the three lines of defense of the power grid to fail. The human brain cannot exhaustively calculate such combined risks.

[0006] Lack of advance optimization suggestions: When there are risks in the maintenance plan, the existing handling methods can usually only give a "not pass" conclusion, and cannot intelligently give constructive suggestions on "how to adjust to pass", which leads to repeated coordination of maintenance work and high costs.

[0007] Therefore, there is an urgent need for a method that can connect maintenance order text with network topology and conduct risk simulation in virtual space to achieve pre-emptive quantitative assessment and closed-loop management of maintenance risks. Summary of the Invention

[0008] The purpose of this invention is to provide a method and system for simulating and verifying the maintenance of power communication networks based on digital twins, which solves the problems in the prior art that maintenance approval relies on manual experience, lacks dynamic simulation methods, and cannot assess the risk of superimposed faults.

[0009] The above-mentioned technical objective of the present invention is achieved through the following technical solution:

[0010] In a first aspect, the present invention provides a method for simulating and verifying the maintenance of a power communication network based on digital twins, the method comprising:

[0011] Convert the pending power communication maintenance application form into a graphical operation command;

[0012] The map operation commands are input into a pre-constructed digital twin model for topology deduction, and the maintenance-state network topology is output; wherein, the digital twin model is constructed from the full topology map of the power communication network;

[0013] Under the maintenance network topology, perform a full network static connectivity analysis and traverse the network survivability after a single failure of the remaining critical links. Combine the voltage level and type weight of the carried services to calculate the maintenance risk index.

[0014] The maintenance risk index is compared with the preset safety threshold, and the simulation verification results are output.

[0015] In one implementation, the power communication maintenance application form to be reviewed is converted into a graph operation instruction, including: using a deep learning model to extract entities and identify intents from the power communication maintenance application form to obtain the maintenance object, action type and time range, and converting the maintenance object, action type and time range into a graph operation instruction.

[0016] In one implementation, a deep learning model is used to extract entities and identify intents from power communication maintenance application forms to obtain the maintenance object, action type, and time range. This includes: extracting word vectors from the power communication maintenance application forms and extracting contextual features from the word vectors.

[0017] The context features are constrained and decoded to calculate the score function, and the optimal label sequence is obtained by solving the score function.

[0018] The inspection object, action type, and time range are extracted from the optimal label sequence.

[0019] In one implementation, the construction process of the digital twin model includes:

[0020] Obtain the full topology graph of the power communication network; wherein, the full topology graph of the power communication network includes a set of nodes and a set of edges;

[0021] The full topology of the power communication network is processed using a copy-on-write mechanism to construct the digital twin model.

[0022] In one implementation, a sandbox baseline state is constructed within the maintenance time window, and the approved maintenance plan within the maintenance time window is superimposed on the time axis superposition operator to adjust the sandbox baseline state.

[0023] The local topology of the digital twin model is modified based on the adjusted sandbox baseline state.

[0024] In one implementation, the map operation command is input into a pre-built digital twin model to perform topology deduction and output the maintenance-state network topology, including: using a state setting function to set the state of the virtual network elements or links in the digital twin model to generate the maintenance-state network topology.

[0025] In one implementation scheme, the maintenance risk index is calculated as follows: ,in, The static service interruption value directly caused by maintenance; This is the set of remaining active links in the sandbox; For the first One active link, This represents the probability of failure of the link. For simulated link Conditional risk value after failure; These are the weighting coefficients.

[0026] In one implementation, the formula for calculating the static service interruption value is: ;in, Assigning importance weight to individual business line b. Voltage level factor, B is the business type factor, and B is the set of interrupted businesses.

[0027] In one implementation scheme, the maintenance risk index is compared with a preset safety threshold, and the simulation verification results are output, including:

[0028] If the maintenance risk index is less than the safety threshold, the simulation verification result that has passed the verification will be output.

[0029] If the maintenance risk index is greater than the safety threshold, an objective function with the path hop count and time delay as the minimum is constructed. A graph search algorithm is used to search for completely disjoint routes in the digital twin model. If a path that meets the constraints is found, a preliminary detour path is generated.

[0030] In a second aspect, the present invention provides a power communication network maintenance simulation and verification system based on digital twins, including a memory and a processor;

[0031] A memory for storing computer programs, the computer programs including program instructions;

[0032] A processor is configured to execute the program instructions to cause the electronic device to perform the steps of a digital twin-based power communication network maintenance simulation verification method provided in the first aspect of the present invention.

[0033] Compared with the prior art, the present invention has the following beneficial effects:

[0034] This invention first parses the maintenance request form into a graphical operation instruction; then, it constructs a digital twin model based on the full topology map of the power communication network, introducing a time axis overlay mechanism to evolve the current network topology to a baseline state at a future planned time; based on this, it performs virtual simulation of maintenance operations within the digital twin model, while simultaneously conducting dynamic cascading fault simulation under maintenance conditions; finally, it quantitatively calculates the maintenance risk index based on the power grid business importance matrix, and outputs simulation verification results based on the comparison between the maintenance risk index and a preset safety threshold. This invention solves the problem that traditional maintenance scheduling relies on manual experience and cannot dynamically assess the risk of cascading faults during maintenance, realizing a transformation from static approval to dynamic simulation and effectively avoiding the risk of unplanned power grid business interruptions. Attached Figure Description

[0035] The accompanying drawings, which are included to provide a further understanding of embodiments of the invention and form part of this application, do not constitute a limitation thereof. In the drawings:

[0036] Figure 1 A flowchart of a power communication network maintenance simulation verification method based on digital twin provided in an embodiment of the present invention;

[0037] Figure 2 This is a schematic diagram of the multi-source maintenance intent semantic understanding and instruction conversion logic provided in an embodiment of the present invention;

[0038] Figure 3 A schematic diagram illustrating the calculation logic of the maintenance risk index provided in an embodiment of the present invention;

[0039] Figure 4 This is a schematic diagram illustrating the intelligent strategy recommendation and optimization logic for high-risk maintenance scenarios provided in this embodiment of the invention. Detailed Implementation

[0040] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the embodiments and accompanying drawings. The illustrative embodiments and descriptions of this invention are only for explaining this invention and are not intended to limit this invention.

[0041] It should be noted that the terms "comprising" or "may include" used in the various embodiments of this application indicate the presence of the claimed function, operation, or element, and do not limit the addition of one or more functions, operations, or elements. Furthermore, as used in the various embodiments of this application, the terms "comprising," "having," and their cognates are intended only to indicate a specific feature, number, step, operation, element, component, or combination of the foregoing, and should not be construed as primarily excluding the presence of one or more other features, numbers, steps, operations, elements, components, or combinations of the foregoing, or adding one or more combinations of the foregoing.

[0042] It should be understood that terms such as "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0043] like Figure 1 As shown in the figure, this embodiment of the invention provides a power communication network maintenance simulation verification method based on digital twins, the method including:

[0044] S101 converts the pending power communication maintenance application form into a map operation command.

[0045] like Figure 2 As shown, for unstructured power communication maintenance request forms, this embodiment uses the BERT-BiLSTM-CRF deep learning model to extract entities and identify intents, obtain the maintenance object, action type and time range, and convert the maintenance object, action type and time range into graph operation instructions.

[0046] Specifically, the word vectors of the power communication maintenance application form are extracted using the BERT layer. The word vectors are then input into the BiLSTM (Bidirectional Long Short-Term Memory) layer to extract the contextual features of the word vectors. Finally, the output of the BiLSTM is constrained and decoded using the CRF (Conditional Random Field) layer to calculate the score function Score(X,Y) and solve for the optimal label sequence, from which time, object, and action entity are extracted.

[0047] For example, suppose the received maintenance application form is: "Request to relocate and cut the OPGW optical cable between 500kV Anning Substation and 500kV Renhe Substation from 10:00 to 14:00 on October 1, 2025, during which the optical cable will be interrupted."

[0048] (1) Text vector representation: Input the above text sequence into the BERT pre-training layer, and convert each Chinese character (such as "申", "请", "在") into a 768-dimensional character vector representation.

[0049] (2) Context feature extraction: Input the character vectors into the BiLSTM layer. For time t, the hidden layer output of BiLSTM is composed of the forward hidden layer and the backward hidden layer concatenated as: .

[0050] (3) Optimal label sequence decoding: Use the CRF layer to perform constrained decoding on the output of BiLSTM. The model calculates the score function , and outputs the optimal label sequence .

[0051]

Example of decoding result

[0052] For the above text, the entity label sequence output by the deep learning model is as follows:

[0053] "2025...14:00" [TIME];

[0054] "500kV Anning Substation" (source site);

[0055] "500kV Renhe Substation" (destination site);

[0056] "OPGW Optical Cable" (object type);

[0057] "Interrupted" [ACTION] (action).

[0058] (4) Instruction mapping: According to the above decoding result, automatically generate a standardized graph database operation instruction .

[0059] The generated graph operation instruction is specifically:

[0060] UPDATE EDGE e SET e.status = 'DOWN';

[0061] WHERE e.type = 'OPGW';

[0062] AND e.source = 'Node_Anning_500';

[0063] AND e.target = 'Node_Renhe_500';

[0064] AND time_range = ['2025-10-01 10:00', '2025-10-01 14:00'].

[0065] S102, input the map operation command into the pre-constructed digital twin model to perform topology deduction and output the maintenance state network topology; wherein, the digital twin model is constructed from the full topology map of the power communication network.

[0066] In this embodiment, the process of constructing a time-series digital twin model of the power communication network based on graph theory is as follows:

[0067] First, define the physical topology of the entire network. , where the set of nodes Example: .

[0068] edge set Example: .

[0069] Secondly, establish maintenance time windows. Sandbox baseline state Introducing a time-axis superposition operator Load other approved maintenance sets within this time period: ,in, This is the current network topology. For the approved maintenance assembly, This is the topology change matrix caused by a single maintenance task.

[0070] Finally, a search revealed another approved work order at 11:00 AM on the same day: "Dechang Substation Communication Power Supply Repair (causing node...)" (Offline). Therefore, the initial sandbox The status includes not only the current network topology It also has the superimposed state of "Dechang variable node failure" (i.e. A copy-on-write mechanism is adopted to initialize the sandbox in an isolated memory area, ensuring that the simulation process does not affect the live network data.

[0071] Specifically, the graph manipulation instructions generated in step S101 are executed in the digital twin model. Set the status of virtual network elements or links.

[0072] Finding edges in a digital twin model (Anning-Renhe OPGW), change its status attribute from UP to DOWN. At this time, the network topology state (maintenance state) within the digital twin model is: edge (Anning-Renhe): Disconnect (due to this work order). Node (Dechang Substation): Failed (due to overlapping work orders). Other links: Normal.

[0073] The specific operation is defined as a state setting function: .

[0074] S103, under the maintenance network topology, performs a full network static connectivity analysis and traverses the network survivability after a single failure of the remaining critical links. Combining the voltage level and type weight of the carried services, the maintenance risk index is calculated.

[0075] Specifically, static connectivity analysis of the entire network refers to the analysis of the network topology state (i.e., maintenance state) that only overlays planned maintenance operations. Under the condition of not introducing any sudden failures, the graph routing algorithm is used to calculate the end-to-end route of all services in the network and to evaluate whether the maintenance operation directly causes the primary and backup channels of some services to be interrupted at the same time (i.e., connectivity and direct impact verification in the N-0 state).

[0076] The relationship between the static connectivity analysis of the entire network and the network survivability of the remaining critical links after a single failure in the traversal simulation described below refers to the fact that the result of the static connectivity analysis serves as the baseline state and effective input for the dynamic traversal simulation. On the one hand, the static analysis identifies the "set of remaining active links" that actually carry core services under maintenance. Subsequent traversal simulations only simulate failures on links in this set one by one, thereby eliminating invalid calculations. On the other hand, if the static analysis finds that the services affected by maintenance have successfully switched to backup routes, the subsequent dynamic traversal simulations are specifically used to examine the vulnerability of these "backup routes" to a sudden single failure (i.e., N-1 failure), thereby quantifying the implicit risks caused by the decrease in network redundancy during maintenance.

[0077] Network survivability refers to the ability of a power communication network to automatically find available alternative routes and maintain end-to-end logical connectivity of core power grid services (such as relay protection and stability control) when encountering specific faults such as physical node failure or link breakage, thus avoiding the disruption of services.

[0078] like Figure 3 As shown, the calculation of the maintenance risk index includes static impact calculation. and dynamic N-1 simulation Two parts, including static influence calculation. The process is as follows: edge detection It carries the "Anning-Renhe Line Longitudinal Differential Protection A Network" service. Due to Upon disconnection, the service will automatically switch to the backup route. The system calculation shows the backup route is: Anning. Tongzilin Renhe Result: Business was not interrupted, therefore... .

[0079] Dynamic N-1 Simulation Iterate through the remaining active links in the sandbox to simulate "what would happen if this link also broke." Simulation scenario: Assume an edge... A random failure occurred on the (Anning-Tongzilin) ​​route. The deduced route (Anning-Tongzilin-Renhe) will be interrupted. An attempt is being made to find a third route. Typically, the third route passes through Dechang Substation. However, according to step S102, Dechang Substation is also under maintenance at this time. Conclusion: If Disconnecting will completely block all services (N-2 risk becomes a reality).

[0080] Therefore, the formula for calculating the maintenance risk index is: ,in, The static service interruption value directly caused by maintenance; This is the set of remaining active links in the sandbox; For the first One active link, This represents the probability of failure of the link. For simulated link Conditional risk value after failure; This is the weighting coefficient. The formula for calculating the static service interruption value is: ;in, Assigning importance weight to individual business line b. Voltage level factor, B is the business type factor, and B is the set of interrupted businesses.

[0081] Here, let , For simulated scenarios : (Probability of failure): Because this section is an old optical cable, the value is... . (Business Impact): Involves 500kV protection services, weighting The calculation results of the maintenance risk index are as follows: (Assuming other link risks are negligible). It should be noted that the 0.6 here is the normalized high-risk value.

[0082] S104 compares the maintenance risk index with the preset safety threshold and outputs the simulation verification results.

[0083] Specifically, the maintenance risk index is compared with the preset safety threshold, and the comparison is divided into the following two situations, such as... Figure 4 As shown, firstly, if the maintenance risk index is less than the safety threshold, the simulation verification result that has passed the verification will be output.

[0084] For example, setting a security threshold The calculated maintenance risk index The maintenance application was determined to be "high-risk," and the simulation verification result was "recommendation to reject."

[0085] Another approach is to construct an objective function that minimizes the number of hops and the time delay, and then use a graph search algorithm to search for completely disjoint routes in the digital twin model if the search finds a path that meets the constraints. If a path is found that satisfies the constraints, a preliminary detour path is generated.

[0086] Specifically, construct the objective function to find the optimal preliminary detour path. The objective function is expressed as: Search process: The K-SP algorithm does not perform a search in the sandbox. (Dechang) route.

[0087] Find a path through the low-pressure side station. :peaceful A certain 220kV substation Renhe. Although the number of jumps is relatively high ( (Larger), but can be connected.

[0088] If a path is established in advance The risk index was recalculated and reduced to .

[0089] The final simulation verification result is: "Before performing the optical cable cutover, please establish a temporary protection channel of 'Anning-220kV station-Renhe' as the third line of defense."

[0090] This invention also provides a power communication network maintenance simulation and verification system based on digital twins. It includes a processor, a memory, a communication interface, and at least one communication bus for connecting the processor, memory, and communication interface. The memory includes, but is not limited to, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (PROM), or portable read-only memory (CD-ROM), and is used for related instructions and data.

[0091] The communication interface is used to receive and send data. The processor can be one or more CPUs; if it is a single CPU, it can be a single-core CPU or a multi-core CPU. The processor in the electronic device reads one or more programs stored in the memory and performs the following operations: converting the pending power communication maintenance application form into a topology operation instruction; inputting the topology operation instruction into a pre-built digital twin model for topology deduction and outputting the maintenance-state network topology; wherein the digital twin model is constructed from the full topology map of the power communication network; under the maintenance-state network topology, performing a full-network static connectivity analysis and traversing the simulation of network survivability after a single failure of the remaining critical links, calculating the maintenance risk index by combining the voltage level and type weight of the carried services; comparing the maintenance risk index with a preset safety threshold and outputting the simulation verification result.

[0092] It should be noted that the specific implementation of each operation can be described above. Figure 1 The corresponding description of the method embodiment shown indicates that the digital twin-based power communication network maintenance simulation and verification system can be used to execute a digital twin-based power communication network maintenance simulation and verification method according to the above-described method embodiment of this application, which will not be described in detail here.

[0093] This invention also provides a computer-readable storage medium, which is a memory device in a computer device for storing programs and data. It is understood that the computer-readable storage medium here can include both the built-in storage medium in the computer device and extended storage media supported by the computer device. The computer-readable storage medium provides storage space that stores the operating system of the terminal. Furthermore, the storage space also stores one or more instructions suitable for loading and execution by a processor. These instructions can be one or more computer programs (including program code). It should be noted that the computer-readable storage medium here can be high-speed RAM or non-volatile memory, such as at least one disk storage device. The processor can load and execute one or more instructions stored in the computer-readable storage medium to implement the corresponding steps of the above embodiments regarding a digital twin-based power communication network maintenance simulation verification method. Those skilled in the art should understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0094] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for simulating and verifying the maintenance of a power communication network based on digital twins, characterized in that, The methods include: Convert the pending power communication maintenance application form into a graphical operation command; The map operation commands are input into a pre-constructed digital twin model for topology deduction, and the maintenance-state network topology is output; wherein, the digital twin model is constructed from the full topology map of the power communication network; Under the maintenance network topology, perform a full network static connectivity analysis and traverse the network survivability after a single failure of the remaining critical links. Combine the voltage level and type weight of the carried services to calculate the maintenance risk index. The maintenance risk index is compared with the preset safety threshold, and the simulation verification results are output.

2. The method according to claim 1, characterized in that, The process involves converting pending power communication maintenance application forms into graphical operation instructions, including: using a deep learning model to extract entities and identify intents from the power communication maintenance application forms to obtain the maintenance objects, action types, and time ranges, and then converting the maintenance objects, action types, and time ranges into graphical operation instructions.

3. The method according to claim 2, characterized in that, Using a deep learning model, entity extraction and intent recognition are performed on power communication maintenance application forms to obtain the maintenance object, action type, and time range, including: extracting word vectors from power communication maintenance application forms and extracting contextual features of the word vectors; The context features are constrained and decoded to calculate the score function, and the optimal label sequence is obtained by solving the score function. The inspection object, action type, and time range are extracted from the optimal label sequence.

4. The method according to claim 1, characterized in that, The construction process of the digital twin model includes: Obtain the full topology graph of the power communication network; wherein, the full topology graph of the power communication network includes a set of nodes and a set of edges; The full topology of the power communication network is processed using a copy-on-write mechanism to construct the digital twin model.

5. The method according to claim 4, characterized in that, Construct a sandbox baseline state within the maintenance time window, and adjust the sandbox baseline state by superimposing the approved maintenance plan within the maintenance time window based on the time axis superposition operator. The local topology of the digital twin model is modified based on the adjusted sandbox baseline state.

6. The method according to claim 1, characterized in that, Input the map operation instructions into the pre-built digital twin model to perform topology deduction and output the maintenance state network topology, including: using the state setting function to set the state of the virtual network elements or links in the digital twin model to generate the maintenance state network topology.

7. The method according to claim 1, characterized in that, The formula for calculating the maintenance risk index is: ,in, The static service interruption value directly caused by maintenance; This is the set of remaining active links in the sandbox; For the first One active link, This represents the probability of failure of the link. For simulated link Conditional risk value after failure; These are the weighting coefficients.

8. The method according to claim 7, characterized in that, The formula for calculating the static service interruption value is: ;in, Assigning importance weight to individual business line b. Voltage level factor, B is the business type factor, and B is the set of interrupted businesses.

9. The method according to claim 1, characterized in that, The maintenance risk index is compared with the preset safety threshold, and the simulation verification results are output, including: If the maintenance risk index is less than the safety threshold, the simulation verification result that has passed the verification will be output. If the maintenance risk index is greater than the safety threshold, an objective function with the path hop count and time delay as the minimum is constructed. A graph search algorithm is used to search for completely disjoint routes in the digital twin model. If a path that meets the constraints is found, a preliminary detour path is generated.

10. A power communication network maintenance simulation and verification system based on digital twins, characterized in that, Including memory and processor; A memory for storing computer programs, the computer programs including program instructions; A processor is configured to execute the program instructions to cause the electronic device to perform the steps of a digital twin-based power communication network maintenance simulation verification method as described in any one of claims 1 to 9.

Images