Power grid anti-misoperation method and system

Abstract

The invention discloses an intelligent power grid anti-misoperation system and method, which realize real-time monitoring and intelligent anti-misoperation verification of power grid equipment operation through various intelligent anti-misoperation means such as condition verification, logic verification and normative verification, and overcome the defect that a traditional system depends on artificial experience and fixed rules. The system can dynamically adjust anti-error logic and conditions, performs anti-error verification by utilizing topology analysis and machine learning technologies, has real-time monitoring and operation state recognition functions, and remarkably improves the safety and reliability of power grid operation through the technical innovation of cooperative work of condition verification, logic verification and normative verification modules, and the system has a wide application prospect. Dynamic changes of a power grid are adapted, and stable operation of a power system is guaranteed.

Description

Technical Field

[0001] This invention belongs to the field of smart grid technology, and specifically relates to a method and system for preventing misoperation in smart grids. Background Technology

[0002] With the continuous expansion of power grid scale and the increasing complexity of power system operation, the risk of operational errors in power grid operations is also increasing. Traditional anti-error operating systems mainly rely on human experience and fixed rules, which are inadequate in the face of the increasingly complex and variable operating environment of modern power grids. Because the real-time status and operating conditions of the power grid are dynamically changing, traditional systems cannot respond and adjust in a timely manner, leading to an increased risk of operational errors and posing a significant challenge to the safe operation of the power grid.

[0003] Existing methods for preventing misoperation may face challenges in practical applications due to the complexity of data acquisition and processing. Furthermore, they still rely on authorization and deregulation rules when handling particularly complex operations, increasing operational complexity and the likelihood of manual intervention. For example, the invention patent "Power Grid Misoperation Prevention Method and Operating System Based on Depth-First Search" (Publication No. CN117477758B) still suffers from insufficient theoretical support in risk identification for large-scale power grid topologies. Moreover, it is inefficient in updating equipment status and synchronizing topology models, making it difficult to provide fully real-time equipment status information. These shortcomings limit the system's effectiveness in complex power grid environments.

[0004] Therefore, there is an urgent need for a power grid anti-misoperation method that can adapt to the dynamic changes of the power grid in real time. Summary of the Invention

[0005] Purpose of the invention: The purpose of this invention is to provide a power grid anti-misoperation operating system and method. In view of the shortcomings of existing technologies that rely on human experience and fixed rules and cannot adapt to the dynamic changes of the power grid, this invention introduces a variety of intelligent anti-misoperation methods such as condition verification, logic verification and standard verification to achieve comprehensive and real-time monitoring and anti-misoperation verification of power grid operation.

[0006] Technical solution: This invention provides a method for preventing power grid misoperation, comprising:

[0007] A topology knowledge base is established based on the physical / logical connections of power grid equipment to store the status of power grid equipment and the connection relationships between power grid equipment; power grid operation data is collected to establish a fault prevention knowledge base to store fault prevention rules.

[0008] The topological knowledge base is updated in real time by combining the incremental learning model. The cosine similarity between the new and old data in the topological knowledge base is masked and the mask is added to the loss function of the incremental learning model.

[0009] A topology model is constructed based on a real-time updated topology knowledge base, including modeling power grid equipment as nodes, the connection relationships between power grid equipment as edges, and the operation process as a node linked list.

[0010] The topology model is analyzed based on the aforementioned anti-misoperation knowledge base to determine if any anti-misoperation anomalies exist. If an anomaly is found, an alarm is issued.

[0011] Furthermore, the error prevention rules include:

[0012] Condition-based error prevention rules include operating conditions and status information of power grid equipment;

[0013] Logical error prevention rules include the operational logic relationships and operation sequences of power grid equipment;

[0014] Normative error prevention rules include power grid equipment operation standards and instruction specifications.

[0015] Furthermore, the operation process of the incremental learning method model includes:

[0016] Deep learning models are used to extract feature vector sets of power grid equipment from power grid operation data, including the physical state, operating conditions and connection relationships of the equipment.

[0017] The feature vector set is clustered using the K-means algorithm to obtain old class features divided into several clusters, and the center vector of each cluster is used as the old class prototype in the topological knowledge base.

[0018] Deep learning models are used again to extract new feature vectors of power grid equipment from power grid operation data;

[0019] The old class prototype and the new feature vector are jointly trained, and the training results are updated to the topological knowledge base.

[0020] Furthermore, the feature vector set The old class features {C1, C2, ..., C} are divided into K clusters. k Efficient partitioning of eigenvectors is achieved by minimizing the squared error function.

[0021]

[0022] Where E is the total error, μ i For cluster class C i The mean vector, v belongs to cluster C. i The eigenvectors of the cluster and the formula for calculating the cluster mean vector are:

[0023]

[0024] |C i | is cluster Ci The number of samples in the middle It is cluster C i The sum of all eigenvectors.

[0025] Furthermore, the total loss function for the joint training is:

[0026]

[0027] Where p is the prototype vector of the old class, and v is the feature vector of the new data; ||p|| and ||v|| are the norms of vectors p and v, respectively; L ce It is the cross-entropy loss, L distill It is a distillation loss. and This is the mask for the corresponding loss, where N is the total number of samples; L cosine λ1 and λ2 are the cosine similarity loss, and λ1 and λ2 are the hyperparameters for balancing the loss.

[0028] Furthermore, the masking process for the cosine similarity between the old and new data includes:

[0029] Calculate the cosine similarity between the new data features and the old class prototype:

[0030]

[0031] If the cosine similarity is greater than the threshold σ, then a mask M is added to the distillation loss. distill Otherwise, add a mask M to the cross-entropy loss. ce .

[0032] Furthermore, the topology model contains n nodes, the edges between nodes are undirected, the adjacency matrix M is an n×n square matrix, and the element values ​​in the matrix reflect the connection relationship between nodes; the operational logic relationship and safe operation conditions between nodes are defined.

[0033] Furthermore, the analysis of whether the topology model has any anti-misoperation anomalies includes:

[0034] Based on the conditional error prevention rules of the error prevention knowledge base, the adjacency matrix is ​​used to check whether the current state of each node in the topology model meets its safe operation conditions. If not, an alarm is issued and recorded in the topology knowledge base.

[0035] Based on the error prevention knowledge base, the logic-based error prevention rules represent the operational logic relationship of the device through logic gates, and check whether the operational logic relationship of each node in the topology model is abnormal. If it does not conform, an alarm is issued and recorded in the topology knowledge base.

[0036] Based on the standardized error prevention rules of the error prevention knowledge base, the fault tree analysis method is used to identify whether the node linked list conforms to the standard. If it does not conform, an alarm is issued and recorded in the topology knowledge base.

[0037] Furthermore, the safe operating conditions are analyzed using an adjacency matrix, including:

[0038] Calculate the fully connected matrix T by multiplying the adjacency matrix M by itself:

[0039] T = M n-1

[0040] Self-multiplication of a connected matrix:

[0041] T (2k) =T k ×T k

[0042] Where k represents the series of the connected matrix;

[0043] Calculate each element of the fully connected matrix, and update the old elements at the same and symmetric positions with the calculated new elements. The formula for calculating each element of a certain order connected matrix is:

[0044]

[0045] Among them, t ij Let n be an element in the connected matrix T, where the elements reflect the connection relationships between devices, and n is the dimension of the matrix.

[0046] The present invention also provides a smart grid anti-misoperation operating system, comprising:

[0047] The knowledge base module is used to build a topology knowledge base based on the physical / logical connection relationships of power grid equipment to store the status of power grid equipment and the connection relationships between power grid equipment; it also collects power grid operation data to build a fault prevention knowledge base to store fault prevention rules.

[0048] The incremental learning module is used to update the topological knowledge base in real time by combining the incremental learning model, perform cosine similarity processing on the new and old data of the topological knowledge base, and add the mask to the loss function of the incremental learning model.

[0049] The modeling module is used to build a topology model based on a real-time updated topology knowledge base, including modeling power grid equipment as nodes, modeling the connection relationships between power grid equipment as edges, and modeling the operation process as a node linked list.

[0050] The analysis module is used to analyze whether there are any anti-misoperation anomalies in the topology model in conjunction with the anti-misoperation knowledge base, and to issue an alarm if an anomaly is found.

[0051] Beneficial effects: Compared with the prior art, the present invention has the following significant improvements:

[0052] (1) Abundant theoretical support: By introducing condition verification, logic verification and standard verification rules to prevent errors, and constructing an error prevention knowledge base, this invention provides abundant theoretical support for power grid error prevention operations and overcomes the shortcomings of traditional error prevention operating systems that rely on human experience and fixed rules.

[0053] (2) Timely updates and optimizations: The system described in this invention also has the ability to dynamically expand the anti-misoperation conditions and logic. The focus is on optimizing the real-time updates of the topology knowledge base. The incremental learning method is used to add newly generated data from the power grid system to the topology knowledge base, keeping the topology knowledge base synchronized with the actual power grid in real time.

[0054] (3) Improved analysis efficiency: When performing error prevention verification on the topology model, the present invention uses a fused adjacency matrix. Based on the symmetry of the connected matrix, the values ​​of the elements at the symmetrical positions are updated at the same time as the old elements are updated, which greatly improves the analysis efficiency of error prevention verification on the topology model. Attached Figure Description

[0055] Figure 1 This is a schematic diagram of the power grid's anti-misoperation operation principle. Detailed Implementation

[0056] The present invention will be further explained below with reference to the accompanying drawings and specific embodiments. Figure 1 The diagram shown illustrates the operational principle of the method of this invention.

[0057] Step 1: Establish a topology knowledge base based on the physical and logical connections of power grid equipment. This base stores the connection relationships between power grid equipment and their states and conditions. Feature vectors for each power grid device are extracted from the topology knowledge base. Collect power grid operation data and establish a fault prevention knowledge base. This knowledge base includes conditional, logical, and standardized fault prevention verification rule modules. The power grid operation data sources include power grid model files, operation records, and historical fault data. The collected data is structured and stored in the knowledge base to ensure good queryability and scalability.

[0058] Specifically, the modules in the error prevention knowledge base include:

[0059] (1) Condition-based error prevention verification module: collects the safe operating conditions and status requirements of power grid equipment, including the voltage level, load status, circuit breaker and disconnector status that the equipment must meet before operation, as well as the open or closed status, energized or de-energized status of the equipment during and after operation.

[0060] (2) Logic-based error prevention and verification module: Construct an error prevention knowledge base that stores the operation logic relationships and operation sequences of storage devices, and obtain the operation logic relationships and operation sequences of various devices in the power grid (such as circuit breakers, disconnectors, transformers, etc.). The operation logic relationships record the dependencies and operation priorities between devices, while the operation sequence clarifies the specific steps of each operation process. For example, the load status needs to be confirmed before closing the circuit breaker, and the circuit breaker needs to be disconnected before operating the disconnector.

[0061] (3) Standardized Error Prevention and Verification Module: This module constructs an error prevention knowledge base that stores operation procedures and instruction specifications, acquiring standard procedures and instruction specifications for various power grid operations. The operation procedures record the specific steps and requirements for each operation, ensuring that operators can perform operations in the prescribed order and according to standards. For example, when performing transformer switching operations, the relevant circuit breakers and disconnect switches must be disconnected first. The instruction specifications record the standard format and content requirements of dispatch instructions, ensuring the accuracy and consistency of each instruction.

[0062] Step 2: Combining an incremental learning model, the topology knowledge base is periodically and incrementally updated through machine learning to keep it synchronized with the actual power grid. The incremental learning involves using a deep learning model to extract deep features from the existing topology knowledge base and then using the K-means clustering algorithm to cluster the data in the feature dataset, resulting in an updated topology knowledge base. This method follows a sample-free regularization setting to avoid storing samples and maintain the invariance of the model architecture during training.

[0063] Feature vectors include:

[0064] (1) Physical status of equipment: including voltage level, load status, and start / stop status of equipment.

[0065] (2) Operating conditions of the equipment: such as the current, power, temperature, frequency, etc., which reflect the real-time working status of the equipment.

[0066] (3) Operation records: Record the past operation history of the equipment, such as the start and stop time of the equipment, fault operation, and maintenance records.

[0067] (4) Fault report: Information such as the type of fault that the equipment has experienced in the past, the time of occurrence, the duration, and the recovery time, which is used to describe the reliability of the equipment and potential risk points.

[0068] (5) Equipment connection relationship: The topology between equipment in the power grid, including the connection relationship and interdependence between each equipment and other equipment.

[0069] Use the K-means algorithm to extract the old class features Divide into K clusters {C1, C2, ..., C kEfficient partitioning of eigenvectors is achieved by minimizing the squared error function.

[0070]

[0071] Where E is the total error, reflecting the sum of the distances between all samples within a cluster and their center. μ i For cluster C i The mean vector of represents the center of the cluster. v belongs to cluster C. i The eigenvectors of the cluster. The formula for calculating the cluster mean vector is:

[0072]

[0073] |C i | is cluster C i The number of samples in the middle. It is cluster C i This involves synthesizing all feature vectors. The center vector of each cluster is used as the prototype of the old class in the topological knowledge base, thus avoiding bias caused by random sampling and making the prototype more representative. The selected set of old class prototypes is denoted as {p1, p2, ..., p...}. k The value of K determines the number of prototypes sampled. The larger the value of K, the richer the old class features are retained, and the stronger the model's ability to maintain old tasks.

[0074] The old category features represent the main characteristics exhibited by power grid equipment in previous operation, status, or fault records, and are used to describe the behavior patterns or status of the equipment under different conditions.

[0075] During the incremental learning phase, the feature vectors generated by new grid operation data, operation records, or fault reports as time progresses or the grid operating status changes are denoted as new data features. New data features refer to equipment data collected in the current or latest grid operation, representing the behavioral patterns and status information of the equipment at the current moment or in the recent state.

[0076] The joint training involves combining deep features from selected old class prototypes and new data. Specific steps include:

[0077] (1) Calculate the cosine similarity between the new data features and the old class prototypes in the topological knowledge base, using the following formula:

[0078]

[0079] Where p is the prototype vector of the old class, and v is the feature vector of the new data. ||p|| and ||v|| are the norms of vectors p and v, respectively.

[0080] (2) If the cosine similarity is greater than the threshold σ, then add a mask M to the distillation loss. distillOtherwise, add a mask M to the cross-entropy loss. ce This effectively maintains the characteristics of old classes in the topological knowledge base and enhances the distinction between old and new classes.

[0081] (3) The total loss function is:

[0082]

[0083] Among them, L ce It is cross-entropy loss, which measures the difference between the model's prediction and the actual label. and This is the mask for the corresponding loss, indicating which samples should be included in the corresponding loss calculation. N is the total number of samples. L distill This is distillation loss, used to preserve the model's memory of features from older classes. Its definition is as follows:

[0084]

[0085] in A mask used to indicate the prototype of the j-th old class, p j It is the j-th old class prototype. i It is the feature vector of the i-th new data. cosine This is the cosine similarity loss, also known as cos(θ), used to enhance the distinction between the old and new classes. λ1 and λ2 are hyperparameters that balance the loss, used to control the weights of each loss term.

[0086] By masking the cosine similarity between the prototype and the current data, adding the mask to the loss function, and further weighting the loss function, the difference between the old and new classes is further enhanced.

[0087] Through the methods described above, the topology knowledge base can dynamically adapt to the addition of new data and the maintenance of existing knowledge categories, thereby achieving comprehensive optimization and updating of the operating conditions, logical relationships, and operating specifications of power grid equipment. In this dynamically adaptable environment, the topology knowledge base not only improves the real-time performance and accuracy of data but also provides strong support for the safe and stable operation of the power grid.

[0088] Step 3: Establish a detailed power grid topology model based on the topology knowledge base, perform real-time topology analysis on the power grid topology model, identify equipment connection relationships, and optimize anti-misoperation verification rules and predict the risk of misoperation through machine learning.

[0089] Based on the physical and logical connections of the power grid equipment, a complete topology model is established. Physical connections include the power transmission paths of the equipment, the electrical connections between devices, and the information transmission network. Logical connections include the control relationships, operating sequences, and interdependencies of the devices; that is, how a change in the state of one device affects the state and operation of other devices.

[0090] The model includes nodes of various power grid devices (such as circuit breakers, disconnectors, transformers, etc.) and the connections between them. An adjacency matrix M is used to represent these connections. If the knowledge base contains n nodes with undirected edges between them, then the adjacency matrix M is an n×n square matrix, and the elements in the matrix reflect the connections between the nodes.

[0091] (1) Based on the conditional error prevention rules of the error prevention knowledge base, the adjacency matrix is ​​used to check whether the current state of each node in the topology model meets its safe operation conditions. If not, an alarm is issued and recorded in the topology knowledge base.

[0092] For each device node, define its safe operating conditions, including voltage level, load status, and the status of circuit breakers and disconnectors. Analyze whether the device nodes (the current state of each device) in the power grid model meet their safe operating conditions by calculating the connectivity matrix.

[0093] During topology analysis, the elements in the matrix reflect the connectivity relationships between devices. Further analysis of each node is performed, and a fused adjacency matrix is ​​used to improve analysis efficiency. Specifically, this includes:

[0094] (1.1) Calculate the fully connected matrix by multiplying the adjacency matrix M by itself:

[0095] T = M n-1

[0096] The system dynamically updates and analyzes the current status of the equipment. If safety conditions are not met, the system issues an alert and records it in the knowledge base.

[0097] (1.2) The self-multiplication of the connected matrix is ​​obtained through the adjacency matrix, and its formula is:

[0098] T (2k) =T k ×T k

[0099] Where T represents the connectivity matrix, and k represents the level of the connectivity matrix. For node updates of the connectivity matrix, since this matrix describes the connectivity between nodes in the topological graph...

[0100] (1.3) Calculate each element of the fully connected matrix and use it to update the old elements. Due to the symmetry of the connected matrix, when updating the old elements, the values ​​of elements at symmetrical positions can also be updated simultaneously, improving the algorithm's efficiency. The formula for calculating each element of a connected matrix of a certain order is:

[0101]

[0102] Among them, t ij Let n be an element in the connected matrix T, where n is the matrix dimension.

[0103] The system determines whether the device status meets predetermined safety conditions based on the elements in the matrix. If the device status does not meet the conditions, the system will trigger an alarm and record it in the topology knowledge base; if the conditions are met, the operation is considered safe.

[0104] (2) Based on the error prevention knowledge base, the logical relationship of the device is represented by logic gates. The logical relationship of each node in the topology model is checked to see if it is abnormal. If it does not conform, an alarm is issued and recorded in the topology knowledge base.

[0105] Logical error prevention rules aim to ensure the correctness of the operation sequence by defining the operational logic relationships and priorities between power grid equipment. The system uses logic gates to represent the interdependencies of devices, and each operation step is checked against the logical relationships between devices to ensure that the operation sequence conforms to the predetermined logic. For example, the system checks whether the voltage requirements are met before closing a circuit breaker, or whether the circuit breaker needs to be opened before operating a disconnecting switch. If an incorrect operation sequence is detected, the system will issue an alarm and record the problem in the error prevention knowledge base.

[0106] (3) Based on the normative error prevention rules of the error prevention knowledge base, the fault tree analysis method is used to identify whether the node linked list conforms to the norm. If it does not conform, an alarm is issued and recorded in the topology knowledge base.

[0107] Standardized error prevention rules focus on standard operating procedures and instruction specifications for equipment, ensuring that power grid operations comply with predetermined operating standards during execution. The system verifies whether operations conform to regulations based on standard operating procedures and instruction specifications for power grid operations. Each operating step and instruction format has clear requirements, and the system uses fault tree analysis to identify whether equipment operations conform to specifications. If an operation is found to be non-compliant, the system will issue an alarm and record abnormal information.

Claims

1. A method for preventing power grid misoperation, characterized in that, include: A topology knowledge base is established based on the physical / logical connections of power grid equipment to store the status of power grid equipment and the connection relationships between power grid equipment; Collect power grid operation data and establish a knowledge base for preventing errors to store error prevention rules; The topological knowledge base is updated in real time by combining the incremental learning model. The cosine similarity between the new and old data in the topological knowledge base is masked and the mask is added to the loss function of the incremental learning model. A topology model is constructed based on a real-time updated topology knowledge base, including modeling power grid equipment as nodes, the connection relationships between power grid equipment as edges, and the operation process as a node linked list. The topology model is analyzed based on the aforementioned anti-misoperation knowledge base to determine if any anti-misoperation anomalies exist. If an anomaly is found, an alarm is issued.

2. The power grid anti-misoperation method according to claim 1, characterized in that, The error prevention rules include: Condition-based error prevention rules include operating conditions and status information of power grid equipment; Logical error prevention rules include the operational logic relationships and operation sequences of power grid equipment; Normative error prevention rules include power grid equipment operation standards and instruction specifications.

3. The power grid anti-misoperation method according to claim 2, characterized in that, The operation process of the incremental learning method model includes: Deep learning models are used to extract feature vector sets of power grid equipment from power grid operation data, including the physical state, operating conditions and connection relationships of the equipment. The feature vector set is clustered using the K-means algorithm to obtain old class features divided into several clusters, and the center vector of each cluster is used as the old class prototype in the topological knowledge base. Deep learning models are used again to extract new feature vectors of power grid equipment from power grid operation data; The old class prototype and the new feature vector are jointly trained, and the training results are updated to the topological knowledge base.

4. The power grid anti-misoperation method according to claim 3, characterized in that, The feature vector set The old class features {C1, C2, ..., C} are divided into K clusters. k Efficient partitioning of eigenvectors is achieved by minimizing the squared error function. Where E is the total error, μ i For cluster class C i The mean vector, v belongs to cluster C. i The eigenvectors of the cluster and the formula for calculating the cluster mean vector are: |C i | is cluster C i The number of samples in the middle It is cluster C i The sum of all eigenvectors.

5. The power grid anti-misoperation method according to claim 4, characterized in that, The total loss function for the joint training is: Where p is the prototype vector of the old class, and v is the feature vector of the new data; ||p|| and ||v|| are the norms of vectors p and v, respectively; L ce It is the cross-entropy loss, L distill It is a distillation loss. and This is the mask for the corresponding loss, where N is the total number of samples; L cosine λ1 and λ2 are the cosine similarity loss, and λ1 and λ2 are the hyperparameters for balancing the loss.

6. The power grid anti-misoperation method according to claim 5, characterized in that, The process of masking the cosine similarity between the old and new data includes: Calculate the cosine similarity between the new data features and the old class prototype: If the cosine similarity is greater than the threshold σ, then a mask M is added to the distillation loss. distill Otherwise, add a mask M to the cross-entropy loss. ce .

7. The power grid anti-misoperation method according to claim 6, characterized in that, The topology model contains n nodes, with no direction between the edges. The adjacency matrix M is an n×n square matrix, and the element values ​​in the matrix reflect the connection relationships between the nodes. The operational logic relationships and safe operation conditions between the nodes are defined.

8. The power grid anti-misoperation method according to claim 7, characterized in that, The analysis of whether the topology model has anti-misoperation anomalies includes: Based on the conditional error prevention rules of the error prevention knowledge base, the adjacency matrix is ​​used to check whether the current state of each node in the topology model meets its safe operation conditions. If not, an alarm is issued and recorded in the topology knowledge base. Based on the error prevention knowledge base, the logic-based error prevention rules represent the operational logic relationship of the device through logic gates, and check whether the operational logic relationship of each node in the topology model is abnormal. If it does not conform, an alarm is issued and recorded in the topology knowledge base. Based on the standardized error prevention rules of the error prevention knowledge base, the fault tree analysis method is used to identify whether the node linked list conforms to the standard. If it does not conform, an alarm is issued and recorded in the topology knowledge base.

9. The power grid anti-misoperation method according to claim 8, characterized in that, The safe operating conditions are analyzed using an adjacency matrix, including: Calculate the fully connected matrix T by multiplying the adjacency matrix M by itself: T=M n-1 Self-multiplication of a connected matrix: T (2k) =T k ×T k Where k represents the series of the connected matrix; Calculate each element of the fully connected matrix, and update the old elements at the same and symmetric positions with the calculated new elements. The formula for calculating each element of a certain order connected matrix is: Among them, t ij Let n be an element in the connected matrix T, where the elements reflect the connection relationships between devices, and n is the dimension of the matrix.

10. A smart grid anti-misoperation operating system, characterized in that, include: The knowledge base module is used to build a topology knowledge base based on the physical / logical connection relationships of power grid equipment, in order to store the status of power grid equipment and the connection relationships between power grid equipment; Collect power grid operation data and establish a knowledge base for preventing errors to store error prevention rules; The incremental learning module is used to update the topological knowledge base in real time by combining the incremental learning model, perform cosine similarity processing on the new and old data of the topological knowledge base, and add the mask to the loss function of the incremental learning model. The modeling module is used to build a topology model based on a real-time updated topology knowledge base, including modeling power grid equipment as nodes, modeling the connection relationships between power grid equipment as edges, and modeling the operation process as a node linked list. The analysis module is used to analyze whether there are any anti-misoperation anomalies in the topology model in conjunction with the anti-misoperation knowledge base, and to issue an alarm if an anomaly is found.

Images