Intelligent substation fault identification method and system based on virtual loop topology

Abstract

The invention discloses an intelligent substation fault identification method and system based on virtual loop topology, and relates to the technical field of substation fault identification, and the method comprises the following steps: obtaining a first model and a second model, and constructing an identification mode which comprises independent identification and combined identification; recognizing a substation fault through the recognition mode, and regulating and controlling substation equipment nodes according to a recognition result; the identification mode is switched by controlling a switching system; the control switching system is obtained by obtaining model identification features and based on random forest construction, and the control switching system comprises verification of identification results and updating of the system. The method is used for solving the problem of balance of misjudgment rate and timeliness caused by topology dynamic change of a traditional single model.

Description

Technical Field

[0001] The present invention relates to the technical field of substation fault identification. More specifically, the present invention relates to an intelligent substation fault identification method and system based on virtual loop topology. Background Art

[0002] With the wide application of intelligent substations in the power system, their virtual loops, as digital communication links connecting secondary equipment, undertake key functions such as relay protection and monitoring signals. Traditional fault identification methods mainly rely on inference systems based on expert rules, which can quickly locate faults through preset logical judgments. However, when facing new fault modes or complex working conditions, due to problems such as lagging rule base updates and rigid inference logics, it is difficult to meet the dynamic change requirements of the power grid. On the other hand, although deep learning-based identification models can mine fault features from historical data, they have limitations such as high training data annotation costs, insufficient real-time performance, and weak scene migration capabilities. Especially in the scenario of multi-source heterogeneous data fusion analysis in intelligent substations, the identification efficiency of a single model is difficult to guarantee.

[0003] The digital transformation of intelligent substations makes virtual loop faults exhibit characteristics of dynamics, concealment, and multi-source. The limitations of traditional methods are further highlighted. Existing technical solutions generally lack in-depth utilization of real-time loop topology features. For example, key indicators such as the network redundancy and signal path complexity of virtual loops are not incorporated into the decision-making basis, resulting in a significant increase in the misjudgment rate in scenarios with high redundancy or dense heterogeneous devices. In addition, most existing systems adopt static strategies to execute fault identification, unable to dynamically adjust the combination of identification strategies according to the historical accuracy of the model, the complexity of real-time working conditions, etc., and lacking a closed-loop optimization mechanism based on repair effects, making it difficult to cope with the challenges brought about by the diversification of equipment types and the complication of communication protocols during the upgrade and transformation of substations.

[0004] For example, a real-time monitoring method and system announced in the invention patent with the announcement number of CN118818970A. The present invention provides a real-time monitoring method and system based on interval observation, belonging to the technical field of intelligent monitoring. The method includes: S1: Obtain the real-time operation parameters of the monitored object, and divide the real-time operation parameters into several intervals according to the preset interval thresholds; S2: Conduct statistical analysis on the real-time operation parameters in each interval to obtain the statistical characteristic values of each interval; S3: Compare the statistical characteristic values of each interval with the preset standard model, and calculate the deviation degree of each interval; S4: Determine whether the operation state of the monitored object is abnormal according to the deviation degree of each interval. If it is abnormal, send out a warning signal; S5: Automatically adjust the control parameters of the monitored object according to the warning signal to make the monitored object return to the normal operation state. This real-time monitoring method fully excavates the statistical characteristics of real-time data, adaptively judges the operation state, and timely adjusts the control parameters, improving the automation degree of monitoring.

[0005] For example, a workstation control mode conversion method and system announced in the invention patent with the announcement number of CN117806246A. The present invention provides a workstation control mode conversion method and system, belonging to the technical field of electrical equipment control. Specifically, it includes collecting historical command data, and intercepting the specified command data according to each control mode of the workstation; performing iterative cross-validation on the specified command data to obtain the credibility score of each specified command data; calculating the association degree score between each specified command data and the corresponding control mode according to the credibility score; combining the association degree score with the configuration information of multiple working modules controlled by the workstation, and performing deep learning to obtain the recognition model of the specified command data and the control mode; according to the input control mode requirements, based on the recognition model, identify the current operation situation of the workstation to obtain the control mode flow data, and switch the control mode of the workstation through the control mode flow data. The present invention can ensure the accuracy of the corresponding control commands and the safety of electrical equipment during the conversion of the workstation control mode.

[0006] In the above disclosed technical solutions, there are at least the following technical problems: Adopting a fixed strategy to execute fault identification cannot dynamically select the optimal model combination according to the real-time scenario, resulting in a high misjudgment rate under complex working conditions; ignoring the dynamic change characteristics of the virtual loop topology structure, and failing to incorporate key features such as loop redundancy and signal path complexity into the decision-making basis; lacking a closed-loop feedback optimization mechanism, unable to continuously update the model strategy through the actual repair effect, and difficult to cope with the challenges brought by new fault modes and the evolution of the power grid structure.

[0007] In view of the above problems, the present invention proposes a solution. Summary of the Invention

[0008] To overcome the above-mentioned defects of the prior art, embodiments of the present invention provide an intelligent substation fault identification method and system based on a virtual loop topology, which solve the balance problem of misjudgment rate and timeliness caused by the dynamic change of topology in a traditional single model through a closed-loop feedback optimization mechanism and a model control switching method.

[0009] To achieve the above object, the present invention provides the following technical solutions: An intelligent substation fault identification method based on a virtual loop topology, comprising the following steps: obtaining a first model and a second model, and constructing an identification mode, the identification mode including independent identification and combined identification; identifying substation faults through the identification mode, and regulating substation equipment nodes according to the identification results; the identification mode is switched through a control switching system; the control switching system is obtained by acquiring model identification features and constructing based on a random forest, and the control switching system includes verification of identification results and update of the system.

[0010] In a preferred embodiment, the model identification features include virtual loop signal data, the accuracy rates of the first model and the second model, and the loop topology complexity obtained through graph theory analysis; the first model includes a rule base model constructed based on expert rules, and the second model includes a machine learning model constructed based on deep learning.

[0011] In a preferred embodiment, the specific acquisition method of the first model is: converting the physical connection of the substation into a virtual loop topology structure; constructing a rule base according to a first rule, the first rule including a topology constraint rule, a combination mapping rule, and a conflict resolution rule; parsing the virtual loop topology structure, and based on the topology state and virtual loop signal data, traversing the fault library through forward chain reasoning to match and output the fault cause.

[0012] In a preferred embodiment, the specific acquisition method of the second model is: obtaining historical fault recording data as training data, and constructing an LSTM layer and a GNN layer; extracting the time series features of the historical fault recording data based on the LSTM layer, and performing topology analysis on the historical fault recording data based on the GNN layer to obtain topology features; fusing the time series features and the topology features to obtain the second model, and outputting the fault cause through a fully connected layer.

[0013] In a preferred embodiment, the recognition mode is switched through a control switching system. The specific method is as follows: Obtain historical fault data as training data, take the highest accuracy rate and the shortest recognition time as decision-making objectives, label the historical fault recognition decision label as the first model independent recognition mode or the second model independent recognition mode, and construct a control switching system based on a random forest. The historical fault data includes model recognition features obtained through simulation, historical fault causes, and historical verification conclusions; obtain virtual loop signal data and loop topology complexity in real time based on the SCADA system and PMU as inputs to the control switching system; output the best recognition mode through the control switching system. When manual intervention occurs, the recognition time constraint is cancelled, and the recognition mode is restricted to the combined recognition mode.

[0014] In a preferred embodiment, the verification of the recognition result and the update of the system are as follows: Based on the recognition result, virtual nodes are supplemented and virtual nodes are replaced in the virtual loop topology of the substation; based on the supplementation and replacement results, a comparison and verification are performed with the preset normal virtual loop topology to determine whether the recognition is correct, and a verification conclusion is obtained in combination with the recognition time; the verification also includes that if the recognition is incorrect for a preset number of consecutive times, the recognition mode is restricted to the first model independent recognition mode, and a warning is given for manual intervention; the model recognition features, recognition results, and verification conclusions are added to the historical fault data to train and update the control switching system.

[0015] In a preferred embodiment, the specific method for obtaining the loop topology complexity is as follows: Based on the graph theory model and the substation SCD file, construct a virtual loop topology relationship graph of device nodes to obtain the number of virtual nodes, the number of signal paths, and the graph density; calculate the path redundancy index of the node according to the network redundancy, quantify the heterogeneity index of the substation device type distribution according to the entropy method, determine the weights of each dimension according to the analytic hierarchy process, perform normalization processing on the path redundancy index and the heterogeneity index, construct a linear weighted comprehensive scoring model, and output the loop topology complexity.

[0016] In a preferred embodiment, the recognition method of the combined recognition mode includes: adding a preset number of neurons to the output layer of the second model, and the neurons correspond to the confidence levels of independent fault causes; select the second model for recognition and output several fault causes with the highest confidence levels; construct a first fault set, and the first fault set includes several fault causes output by the second model; select the first model, traverse the first fault set, match and output the fault causes, and restore the output layer of the second model after the fault causes are output.

[0017] In a preferred embodiment, the regulation of the substation device nodes according to the recognition result is as follows: After determining that the recognition is correct, convert the virtual loop topology structure into the physical connection of the substation, and repair and replace the device nodes of the physical connection of the substation.

[0018] A system for the intelligent substation fault identification method based on virtual circuit topology, including an identification module, a control switching module, a verification and update module, and a regulation and control module: The identification module is used to obtain the first model and the second model, and construct an identification mode, and identify substation faults through the identification mode. The identification mode includes independent identification and combined identification; The control switching module is used to switch the identification mode through the control switching system, and the control switching system is constructed by obtaining model identification features and based on random forests; The verification and update module is used for the verification of the identification result and the update of the system; The regulation and control module is used to regulate the substation equipment nodes according to the identification result.

[0019] The technical effects and advantages of the intelligent substation fault identification method and system based on virtual circuit topology of the present invention: The present invention constructs a fast model by constructing a fault library and integrating expert rules, constructs a slow model based on deep learning, and forms an identification mode independently or in combination, and dynamically selects the optimal identification mode in combination with random forests, thereby realizing the dual improvement of the accuracy and timeliness of virtual circuit fault identification; furthermore, based on the quantitative evaluation of topological complexity, the adaptability in complex topological scenarios is enhanced; through the repair verification feedback loop and continuous iterative update of the model, it helps to construct a self-optimizing fault diagnosis system, effectively solving the problems of high misjudgment rate, fixed strategy and strong manual dependence caused by the topological dynamic change of the traditional single model. Description of the Drawings

[0020] Figure 1 It is a schematic flowchart of the intelligent substation fault identification method based on virtual circuit topology provided by an embodiment of the present invention.

[0021] Figure 2 It is a schematic structural diagram of the system of the intelligent substation fault identification method based on virtual circuit topology provided by an embodiment of the present invention. Detailed Embodiments

[0022] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0023] Embodiment 1, Figure 1 The intelligent substation fault identification method based on virtual circuit topology of the present invention is given, including the following steps: S1, obtain the first model and the second model, and construct an identification mode, where the identification mode includes independent identification and combined identification; S2. Identify substation faults through an identification mode, and regulate substation equipment nodes according to the identification results; the identification mode is switched through a control switching system; the control switching system is obtained by acquiring model identification features and constructing based on a random forest, and the control switching system includes verification of identification results and system update.

[0024] In this embodiment, a fast model is constructed by integrating expert rules in a fault library, a slow model is constructed based on deep learning, and they are formed into an identification mode independently or in combination. The optimal identification mode is dynamically selected in combination with a random forest, thereby achieving a double improvement in the accuracy and timeliness of virtual loop fault identification; furthermore, based on the quantitative evaluation of topological complexity, the adaptability in complex topological scenarios is enhanced; through the repair verification feedback loop and continuous iterative update of the model, it helps to construct a self-optimizing fault diagnosis system, effectively solving the problems of high misjudgment rate, fixed strategies, and strong manual dependence caused by the dynamic change of the traditional single model due to topology.

[0025] S1. Obtain a first model and a second model, and construct an identification mode, where the identification mode includes independent identification and combined identification.

[0026] In this embodiment, the first model includes a rule library model constructed based on expert rules, and the second model includes a machine learning model constructed based on deep learning.

[0027] In this embodiment, the specific acquisition method of the first model is as follows: Convert the physical connection of the substation into a virtual loop topology structure; Construct a rule library according to the first rules, and the first rules include topological constraint rules, combination mapping rules, and conflict resolution rules; Analyze the virtual loop topology structure, and based on the topological state and virtual loop signal data, traverse the fault library through forward chain reasoning to match and output the fault cause.

[0028] In this embodiment, the topological constraint rules include the connectivity of the virtual loop; the combination mapping rules include the logical matching relationship between multi-source signals and fault types; the conflict resolution rules include the arbitration of contradictions in conclusions caused by the triggering of multiple rules.

[0029] It should be noted that the topological constraint rules are used to verify the physical legality of the virtual loop structure. For example, differential protection must form a closed loop with the corresponding CT / PT. If a loop break or illegal bypass is detected, the fault determination is not triggered; the combination mapping rules are used to establish the connection between signal combinations and faults. For example, when the signal combination of "differential protection action + circuit breaker tripping + regional voltage loss" appears, it is mapped to "internal short circuit", while "overcurrent protection action + circuit breaker refusal to operate + voltage dip" points to "circuit breaker failure"; the conflict resolution rules are used to arbitrate the contradictions in conclusions when multiple rules are triggered to ensure the uniqueness of the diagnosis result, such as sorting by priority.

[0030] In this embodiment, the specific method for obtaining the second model is as follows: Obtain historical fault recording data as training data, and construct an LSTM layer and a GNN layer; Extract the time series features of the historical fault recording data based on the LSTM layer, and perform topological analysis on the historical fault recording data based on the GNN layer to obtain topological features; Fuse the time series features and topological features to obtain the second model, and the fully connected layer outputs the fault cause.

[0031] It should be noted that the first model is suitable for rapid fault identification, but it is difficult to handle fault analysis under complex topologies. The second model is suitable for more complex topological scenarios, but the processing time is longer.

[0032] S2. Identify the substation fault through the identification mode, and regulate the substation equipment nodes according to the identification result; the identification mode is switched through the control switching system; the control switching system is obtained by acquiring the model identification features and constructing based on the random forest, and the control switching system includes the verification of the identification result and the update of the system.

[0033] In this embodiment, the model identification features include virtual loop signal data, the accuracy rates of the first model and the second model, and the loop topology complexity obtained according to graph theory analysis.

[0034] In this embodiment, the identification mode is switched through the control switching system. The specific method is as follows: Obtain historical fault data as training data, take the highest accuracy rate and the least identification time as the decision-making objectives, label the historical fault identification decision label as the independent identification mode of the first model or the independent identification mode of the second model, and construct the control switching system based on the random forest. The historical fault data includes the model identification features obtained according to simulation, historical fault causes, and historical verification conclusions; Based on the SCADA system and PMU, real-time obtain the virtual loop signal data and loop topology complexity as the input of the control switching system; Output the best identification mode through the control switching system. When manual intervention occurs, the identification time constraint is cancelled, and the identification mode is restricted to the combined identification mode.

[0035] In this embodiment, the specific method for obtaining the loop topology complexity is as follows: Based on the graph theory model and the substation SCD file, construct the virtual loop topology relationship graph of the equipment nodes to obtain the number of virtual nodes, the number of signal paths, and the graph density; Calculate the path redundancy index of nodes according to the network redundancy, quantify the heterogeneity index of the distribution of substation equipment types according to the entropy method, determine the weights of each dimension according to the analytic hierarchy process, normalize the path redundancy index and the heterogeneity index, construct a linear weighted comprehensive scoring model, and output the loop topology complexity.

[0036] It should be noted that the virtual loop signal data includes protection action signals, equipment status signals, electrical parameter sampling, and control command signals.

[0037] The Supervisory Control and Data Acquisition (SCADA) system is an automated system used for real-time monitoring and control of industrial equipment and infrastructure. It collects data such as temperature, pressure, and flow through remote terminal units (RTUs) or programmable logic controllers (PLCs) distributed at the site and transmits the data to the central control room through a communication network. The SCADA system supports operators to remotely monitor the equipment status, adjust parameters, or trigger control commands, and is widely used in fields such as power, water, oil and gas, transportation, etc., especially suitable for distributed systems that require centralized management. Its characteristics include strong real-time performance, wide coverage, and a user-friendly human-machine interface, but it focuses on steady-state monitoring and has limited response capabilities for dynamic processes.

[0038] The Phasor Measurement Unit (PMU) is the core device of the Wide Area Measurement System (WAMS), mainly used for high-precision dynamic monitoring of the power system. Through the synchronized phasor technology, combined with the GPS clock signal, it collects the amplitude, phase, and frequency of electrical quantities such as bus voltage and line current at high frequencies and synchronously uploads the data to the master station. The PMU can capture system transient processes such as short circuits and oscillations, provide real-time dynamic information for power dispatching, and support power grid stability analysis, fault location, and optimization of control strategies. Compared with the traditional SCADA system, the PMU has higher time synchronization accuracy and data refresh rate, but higher costs, and is mainly applied to the advanced monitoring and protection of high-voltage transmission networks and large power systems.

[0039] Random Forest is an ensemble learning algorithm based on decision trees, which improves the generalization ability of the model by constructing multiple decision trees and integrating their prediction results. Its core advantage lies in effectively reducing the risk of overfitting through bootstrap sampling and random feature selection, while maintaining the efficiency of the model. The model built based on Random Forest has significant lightweight characteristics: the natural hierarchical structure of the decision tree itself has low computational complexity, and the parallel training mechanism further improves the efficiency; the model can achieve stable performance without complex parameter tuning, and the final integrated result can be quickly output through simple voting or averaging without additional in-depth calculation. This lightweight nature makes it perform well in resource-constrained environments, being able to quickly complete training and inference, and achieve high-precision prediction with a small model size, making it an ideal choice that balances performance and efficiency.

[0040] The Substation Configuration Description (SCD) file is the core configuration file of the standard substation automation system, which is used to describe the communication parameters, data models, logical node configurations, and network communication relationships of all intelligent electronic devices (IEDs) in the substation. It contains key information such as the physical structure of the substation, device function definitions, GOOSE / SV communication configurations, dataset mappings, and report control blocks.

[0041] In this embodiment, the recognition method of the combined recognition mode includes: Add a preset number of neurons to the output layer of the second model, and the neurons correspond to the confidence levels of independent fault causes; Select the second model for recognition and output several fault causes with the highest confidence levels; Construct a first fault set, and the first fault set includes several fault causes output by the second model; Select the first model, traverse the first fault set, match and output the fault causes, and restore the output layer of the second model after the fault causes are output.

[0042] In this embodiment, the verification of the recognition result and the update of the system are specifically as follows: Supplement and replace virtual nodes in the virtual loop topology of the substation according to the recognition result; Compare and verify the supplement and replacement results with the preset normal virtual loop topology to determine whether the recognition is correct, and obtain a verification conclusion in combination with the recognition time; the verification also includes that if the recognition is incorrect for a preset number of consecutive times, limit the recognition mode to the independent recognition mode of the first model and alert for manual intervention; Add the model recognition features, recognition results, and verification conclusions to the historical fault data to train and update the control switching system.

[0043] In this embodiment, the verification conclusion includes recognition correctness and recognition time.

[0044] In this embodiment, the regulation of the substation equipment nodes according to the recognition result is specifically as follows: After determining that the recognition is correct, convert the virtual loop topology structure into the physical connection of the substation, and repair and replace the equipment nodes of the physical connection of the substation.

[0045] Embodiment 2 Figure 2 This embodiment provides a system for the intelligent substation fault recognition method based on the virtual loop topology of the present invention, including a recognition module, a control switching module, a verification and update module, and a regulation module: The recognition module is used to obtain the first model and the second model, construct a recognition mode, and recognize substation faults through the recognition mode. The recognition mode includes independent recognition and combined recognition; A control switching module, which is used to switch the recognition mode through a control switching system obtained by acquiring model recognition features and constructing based on a random forest; A verification and update module, which is used for verifying the recognition result and updating the system; A regulation module, which is used to regulate the substation equipment nodes according to the recognition result.

[0046] The above formulas are all dimensionless and take their numerical values for calculation. The formulas are obtained by collecting a large amount of data for software simulation to get a formula closest to the actual situation. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0047] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product.

[0048] Those of ordinary skill in the art can realize that the modules and algorithm steps of each example described in combination with the embodiments disclosed in this article can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of this application.

[0049] In addition, the functional modules in each embodiment of this application can be integrated into one processing module, or each module can exist physically alone, or two or more modules can be integrated into one module.

[0050] The above is only the specific implementation manner of this application, but the protection scope of this application is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed in this application, and all should be covered by the protection scope of this application. Therefore, the protection scope of this application should be subject to the protection scope of the claims.

[0051] Finally: The above is only the preferred embodiment of the present invention and is not used to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. An intelligent substation fault identification method based on virtual loop topology, characterized in that, It includes the following steps: Obtain a first model and a second model, and construct an identification mode, where the identification mode includes independent identification and combined identification; Identify substation faults through the identification mode, and regulate substation equipment nodes according to the identification results; the identification mode is switched through a control switching system; the control switching system is obtained by acquiring model identification features and constructing based on a random forest. The control switching system includes acquiring virtual loop signal data and loop topology complexity as inputs, outputting the best identification mode, and also includes verification of the identification result and update of the system.

2. The intelligent substation fault identification method based on virtual loop topology according to claim 1, characterized in that, The model identification features include virtual loop signal data, the accuracy rates of the first model and the second model, and the loop topology complexity obtained through graph theory analysis; The first model includes a rule base model constructed based on expert rules, and the second model includes a machine learning model constructed based on deep learning.

3. The intelligent substation fault identification method based on virtual loop topology according to claim 2, characterized in that, For the first model, the specific acquisition method is: Convert the physical connection of the substation into a virtual loop topology structure; Construct a rule base according to the first rules, where the first rules include topology constraint rules, combination mapping rules, and conflict resolution rules; Parse the virtual loop topology structure, and based on the topology state and virtual loop signal data, traverse the fault library through forward chain reasoning to match and output the fault cause.

4. The intelligent substation fault identification method based on virtual loop topology according to claim 3, characterized in that, For the second model, the specific acquisition method is: Obtain historical fault recording data as training data, and construct an LSTM layer and a GNN layer; Extract the time series features of the historical fault recording data based on the LSTM layer, and perform topology analysis on the historical fault recording data based on the GNN layer to obtain topology features; Fuse the time series features and topology features to obtain the second model, and the fully connected layer outputs the fault cause.

5. The intelligent substation fault identification method based on virtual loop topology according to claim 4, characterized in that, The identification mode is switched through a control switching system. The specific method is: Obtain historical fault data as training data, take the highest accuracy rate and the least identification time as the decision-making goals, label the historical fault identification decision label as the independent identification mode of the first model or the independent identification mode of the second model, and construct a control switching system based on a random forest. The historical fault data includes the model identification features obtained through simulation, historical fault causes, and historical verification conclusions; Obtain virtual loop signal data and loop topology complexity in real time based on the SCADA system and PMU as inputs to the control switching system; Output the best identification mode through the control switching system. When manual intervention occurs, the identification time constraint is cancelled, and the identification mode is restricted to the combined identification mode.

6. The intelligent substation fault identification method based on virtual loop topology according to claim 5, characterized in that, For the verification of the identification result and the update of the system, specifically: Perform virtual node supplementation and virtual node replacement on the substation virtual loop topology according to the identification result; Compare and verify the supplementation and replacement results with the preset normal virtual loop topology to determine whether the identification is correct, and obtain a verification conclusion in combination with the identification time; the verification also includes that if the identification is incorrect for a preset number of consecutive times, the identification mode is restricted to the independent identification mode of the first model, and a warning is given for manual intervention; Add the model identification features, identification results, and verification conclusions to the historical fault data to train and update the control switching system.

7. The intelligent substation fault identification method based on virtual loop topology according to claim 6, characterized in that, For the loop topology complexity, the specific acquisition method is: Construct a topological relationship graph of virtual circuits for device nodes based on a graph theory model and the SCD file of a substation, and obtain the number of virtual nodes, the number of signal paths, and the graph density; Calculate the path redundancy index of nodes according to the network redundancy, quantify the heterogeneity index of the distribution of substation equipment types according to the entropy method, determine the weights of each dimension according to the analytic hierarchy process, normalize the path redundancy index and the heterogeneity index, construct a linear weighted comprehensive scoring model, and output the loop topology complexity.

8. The intelligent substation fault identification method based on virtual loop topology according to claim 7, characterized in that, The recognition method of the combined recognition mode includes: Add a preset number of neurons to the output layer of the second model, and the neurons correspond to the confidence levels of independent failure causes; Select the second model for recognition and output several failure causes with the highest confidence levels; Construct a first failure set, and the first failure set includes several failure causes output by the second model; Select the first model, traverse the first failure set, match and output the failure causes, and restore the output layer of the second model after the failure causes are output.

9. The intelligent substation fault identification method based on virtual circuit topology according to claim 8, characterized in that, The regulation of substation device nodes according to the recognition result is specifically: After determining that the recognition is correct, convert the virtual circuit topology structure into the physical connection of the substation, and repair and replace the device nodes of the physical connection of the substation.

10. A system using the intelligent substation fault identification method based on virtual circuit topology according to any one of claims 1-9, characterized in that, It includes an identification module, a control switching module, a verification and update module, and a regulation module: The identification module is used to obtain the first model and the second model, construct an identification mode, and identify substation faults through the identification mode. The identification mode includes independent identification and combined identification; The control switching module is used to switch the identification mode through a control switching system, and the control switching system is constructed by obtaining model identification features and based on a random forest; The verification and update module is used for the verification of the identification result and the update of the system; The regulation module is used to regulate substation device nodes according to the recognition result.

Images