Distribution network online fault analysis system based on the Internet of Things

Abstract

The present invention discloses an online fault analysis system for a distribution network based on the Internet of Things, belonging to the technical field of the Internet of Things. The system comprises an Internet of Things perception module, a data acquisition and preprocessing module, a topology modeling and analysis module, a fault analysis module, a fault recovery strategy module, and a communication interaction module. The present invention constructs a topological structure model of the distribution network through the topology modeling and analysis module, and can accurately calculate the fault location and determine the optimal switch operation sequence to isolate the fault area through the fault detection, positioning, and isolation units in the fault analysis module. This precise positioning and isolation helps to reduce the impact range of the fault on the distribution network and shorten the recovery time. The fault analysis module comprises three units: fault detection, positioning, and isolation. Fault positioning is performed through an abnormal score function, fault probability score calculation, and combination with a graph-level representation vector, so that the fault location can be determined quickly and accurately.

Description

Technical Field

[0001] The present invention belongs to the technical field of Internet of Things, and specifically refers to an online fault analysis and judgment system for a distribution network based on the Internet of Things. Background Art

[0002] With the rapid development of smart grid and Internet of Things technologies, distribution network, as a key component of the power grid system, has received increasing attention for its fault location and diagnosis technology. Traditional distribution network fault troubleshooting methods are time-consuming, inefficient, and costly, and are no longer able to meet the needs of modern power systems for rapid fault response.

[0003] However, the existing online fault analysis and judgment system for distribution networks still has certain defects. The existing online fault analysis and judgment system for distribution networks relies on traditional fault detection methods and cannot accurately locate the specific location of the fault, resulting in unnecessary expansion of the power outage range. It relies on a relatively simplified topology model and fails to fully collect or update the geographical location, connection relationship and detailed operating status information of the equipment in the distribution network in real time, resulting in a large difference between the model and the actual distribution network, affecting the accuracy and real-time performance of fault analysis. It may lack sufficient intelligence and flexibility when formulating fault recovery strategies, making it difficult to make the best decision based on the actual situation of the distribution network and the type of fault. Therefore, an online fault analysis and judgment system for distribution networks based on the Internet of Things is proposed. Summary of the Invention

[0004] The purpose of the present invention is to provide an online fault analysis system for distribution network based on the Internet of Things to solve the problems raised in the above background technology.

[0005] To achieve the above-mentioned purpose, the present invention provides the following technical solutions: an online fault analysis and judgment system for distribution network based on the Internet of Things, comprising an Internet of Things perception module, a data acquisition and preprocessing module, a topology modeling and analysis module, a fault analysis module, a fault recovery strategy module, and a communication interaction module;

[0006] The IoT sensing module monitors the state of the power distribution network in real time through the deployed IoT sensors;

[0007] The data acquisition and preprocessing module is used to collect data from the Internet of Things perception module in real time and perform preprocessing;

[0008] The topology modeling and analysis module is used to construct a distribution network topology model and perform topology analysis;

[0009] The fault analysis module monitors the abnormal situation of the distribution network in real time, determines the specific location of the fault, and calculates the optimal solution to isolate the fault area;

[0010] The fault recovery strategy module analyzes the fault analysis results and formulates a fault recovery strategy;

[0011] The communication interaction module is used to provide an intuitive operation interface, allowing operation and maintenance personnel to monitor the status of the distribution network and perform control operations.

[0012] The topology modeling and analysis module collects basic data of the distribution network from the data acquisition and preprocessing module, obtains the location coordinates, geographical distribution information, and line data information of all devices in the distribution network, determines the key nodes in the distribution network based on the collected basic data, defines attributes for each key node, determines the physical connection between nodes based on the actual connection mode of the distribution network, defines attributes for each edge, designs a model of a graph database based on the definitions of nodes and edges, imports the collected data into the graph database, and establishes connection relationships between nodes in the graph database based on the actual connection mode between devices to form a distribution network topology model;

[0013] The status of the distribution network is monitored in real time according to the IoT perception module, and the real-time monitoring data is combined with the constructed distribution network topology model to update the status of the nodes and edges in the distribution network topology model.

[0014] The IoT perception module can monitor the status of the distribution network in real time and continuously through the deployed sensors, including key parameters such as voltage, current, and temperature, to ensure the real-time and accuracy of the data.

[0015] The topology modeling and analysis module performs distribution network topology analysis based on the constructed distribution network topology model and extracts node and edge features from the distribution network topology model. Suppose an undirected graph is G=(V, E), where V represents the node set, E represents the edge set, and each node There are associated node feature vectors , each edge There are associated edge eigenvectors , calculate the new node feature vector of each node i in the l+1 layer through the graph neural network , expressed in the l+1 layer as:

[0016] ,

[0017] In the formula, represents the weight matrix of the lth layer, represents the feature vector of node i, Represents the feature vector of node i's neighbor node j The weighted sum of represents the activation function, N(i) represents the neighbor set of node i, represents the standard factor, Represents the bias vector of the lth layer.

[0018] Among them, according to the obtained new node feature vector, the last layer of new node feature vectors of all nodes are aggregated into the representation vector of the entire graph through pooling. The implementation formula of the graph-level representation vector g is:

[0019] ,

[0020] In the formula, Represents the new node feature vector set of all nodes in the last layer, Represents the new node feature vector of node i in the last layer, represents the pooling function, represents the normalization coefficient, represents the sum of the eigenvectors of all nodes, Indicates the total number of nodes, Represents the new node feature vector of node i in layer L.

[0021] The data acquisition and preprocessing module cleans, filters and standardizes the raw data, while the topology modeling and analysis module uses the graph neural network algorithm to accurately model the topology structure of the distribution network to ensure the accuracy of fault analysis.

[0022] The fault analysis module includes a fault detection unit, a fault location unit and a fault isolation unit;

[0023] The fault detection unit collects data from the data acquisition and preprocessing module in real time to perform fault anomaly detection;

[0024] The fault location unit calculates a fault probability score based on the detection result of the fault detection unit, combines the fault probability score with the graph-level representation vector, and determines the specific location of the fault;

[0025] The fault isolation unit calculates an optimal switch operation sequence according to the specific location of the fault;

[0026] The fault detection unit collects data from the data acquisition and preprocessing module in real time. , define an anomaly score function to measure the degree to which the data at a given time point deviates from the normal range. The implementation formula is:

[0027] ,

[0028] In the formula, Represents a data point at time t The anomaly score, represents a data point at time t, represents the weight of the i-th distribution network state feature, represents the value of the state characteristic of the ith distribution network at time t, represents the mean value of the state characteristics of the ith distribution network, represents the standard deviation of the i-th distribution network state feature, and n represents the number of distribution network state features.

[0029] The fault location unit calculates the fault probability of each node according to the fault anomaly score. There are N nodes, and the probability score of each node j having a fault is , the fault probability score realization formula is:

[0030] ,

[0031] In the formula, Representation node The anomaly score, N represents the total number of nodes, represents the natural exponential function, which is used to convert anomaly scores into positive values ​​and amplify the differences, represents the sum of the exponentials of all node anomaly scores;

[0032] The specific location of the fault is determined based on the obtained fault probability score combined with the graph-level representation vector. The implementation formula is:

[0033] ,

[0034] In the formula, represents the fault score of node j, represents the new node feature vector of node i in the last layer, g represents the graph level representation vector, Represents a similarity function, which calculates the similarity between vectors and scores them according to the fault All nodes are sorted, and the node with the highest score is the specific location of the node failure.

[0035] The fault analysis module includes three units: fault detection, location, and isolation. It locates the fault quickly and accurately by using anomaly score functions, fault probability score calculations, and combining graph-level representation vectors.

[0036] The fault isolation unit calculates the optimal switch operation sequence according to the specific fault location, and the optimal switch state is set as , the optimal isolation implementation formula is:

[0037] ,

[0038] Subject to:

[0039] ,

[0040] ,

[0041] In the formula, represents the switch state connecting node i and node j, Indicates switch The cost of λ is the adjustment parameter. represents the sum of the costs of all switches, represents the sum of the products of the fault score and the switch status, and Indicates that each node can only be connected to one node.

[0042] Among them, the fault recovery strategy module formulates a fault recovery strategy based on the positioning results and optimal isolation results provided by fault analysis, and according to the status of the fault area and non-fault area, executes the fault recovery strategy and monitors the quasi-status of the distribution network in real time, and obtains a graph-level representation vector through pooling aggregation, thereby improving the accuracy and efficiency of fault analysis.

[0043] Among them, the communication interaction module provides an intuitive operation interface, allowing operation and maintenance personnel to monitor the status of the distribution network and perform control operations, generate instructions based on the output of the fault recovery strategy module, send the instructions to the control center, collect distribution network status and recovery progress information, and send status reports to the control center.

[0044] Compared with the prior art, the present invention has the following beneficial effects:

[0045] 1. To address the problems of insufficient real-time performance, difficulty in fault location, and low fault isolation and recovery efficiency in traditional distribution networks, and to improve real-time performance, accuracy, and efficiency, the present invention constructs a distribution network topology model through a topology modeling and analysis module. Furthermore, through the fault detection, location, and isolation units in the fault analysis module, the fault location can be accurately calculated, and the optimal switching operation sequence can be determined to isolate the fault area. This precise location and isolation helps reduce the impact of the fault on the distribution network and shorten the recovery time.

[0046] 2. The present invention uses a topology modeling and analysis module to collect information such as the geographical location, connection relationship, and operating status of each device in the distribution network, construct a topology model that is highly consistent with the actual distribution network, track changes in the status of devices in the distribution network in real time, and automatically update the topology model to ensure the accuracy of fault diagnosis;

[0047] 3. The present invention uses a fault analysis module, including three units: fault detection, location, and isolation. It locates faults through anomaly score functions, fault probability score calculations, and in combination with graph-level representation vectors. This allows for rapid and accurate determination of the fault location. The fault detection unit analyzes collected data in real time and immediately triggers the fault analysis process upon detection of an anomaly, shortening the time interval between fault discovery and analysis. The fault location unit calculates the fault probability score and, in combination with graph-level representation vectors, can accurately locate the specific location of the fault, reducing the possibility of misjudgment and missed judgment. The fault isolation unit calculates the optimal switch operation sequence based on the fault location, achieving rapid isolation of the fault area and preventing the fault from expanding and affecting non-fault areas.

[0048] 4. The present invention can intelligently formulate a fault recovery strategy based on the fault analysis results and the actual situation of the distribution network through the fault recovery strategy module, including steps such as fault isolation, load transfer, and power restoration, to ensure the efficiency and safety of fault recovery. During the fault recovery process, the system can monitor the status of the distribution network in real time, including changes in parameters such as voltage, current, and load, to ensure the smooth implementation of the recovery strategy. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 This is a structural diagram of the online fault analysis and judgment system for the distribution network based on the Internet of Things of the present invention;

[0050] Figure 2 This is an operation flow chart of the topology modeling and analysis module of the distribution network online fault analysis system based on the Internet of Things of the present invention;

[0051] Figure 3 This is an operation flow chart of the fault analysis module of the online fault analysis system for distribution network based on the Internet of Things of the present invention. DETAILED DESCRIPTION

[0052] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0053] Example

[0054] See also Figure 1-Figure 3 As shown, the present invention provides a technical solution: including an Internet of Things perception module, a data acquisition and preprocessing module, a topology modeling and analysis module, a fault analysis module, a fault recovery strategy module, and a communication interaction module;

[0055] The IoT sensing module monitors the state of the power distribution network in real time through the deployed IoT sensors;

[0056] The data acquisition and preprocessing module is used to collect data from the Internet of Things perception module in real time and perform preprocessing;

[0057] The topology modeling and analysis module is used to construct a distribution network topology model and perform topology analysis;

[0058] The fault analysis module monitors the abnormal situation of the distribution network in real time, determines the specific location of the fault, and calculates the optimal solution to isolate the fault area;

[0059] The fault recovery strategy module analyzes the fault analysis results and formulates a fault recovery strategy;

[0060] The communication interaction module is used to provide an intuitive operation interface, allowing operation and maintenance personnel to monitor the status of the distribution network and perform control operations.

[0061] The topology modeling and analysis module collects basic data of the distribution network from the data acquisition and preprocessing module, obtains the location coordinates, geographical distribution information, and line data information of all devices in the distribution network, determines the key nodes in the distribution network based on the collected basic data, defines attributes for each key node, determines the physical connection between nodes based on the actual connection mode of the distribution network, defines attributes for each edge, designs a model of a graph database based on the definitions of nodes and edges, imports the collected data into the graph database, and establishes connection relationships between nodes in the graph database based on the actual connection mode between devices to form a distribution network topology model;

[0062] The status of the distribution network is monitored in real time according to the IoT perception module, and the real-time monitoring data is combined with the constructed distribution network topology model to update the status of the nodes and edges in the distribution network topology model.

[0063] The IoT perception module can monitor the status of the distribution network in real time and continuously through the deployed sensors, including key parameters such as voltage, current, and temperature, to ensure the real-time and accuracy of the data.

[0064] The topology modeling and analysis module performs distribution network topology analysis based on the constructed distribution network topology model and extracts node and edge features from the distribution network topology model. Suppose an undirected graph is G=(V, E), where V represents the node set, E represents the edge set, and each node There are associated node feature vectors , each edge There are associated edge eigenvectors , calculate the new node feature vector of each node i in the l+1 layer through the graph neural network , expressed in the l+1 layer as:

[0065] ,

[0066] In the formula, represents the weight matrix of the lth layer, represents the feature vector of node i, Represents the feature vector of node i's neighbor node j The weighted sum of represents the activation function, N(i) represents the neighbor set of node i, represents the standard factor, Represents the bias vector of the lth layer.

[0067] Among them, according to the obtained new node feature vector, the last layer of new node feature vectors of all nodes are aggregated into the entire graph through pooling, and the graph-level representation vector g is implemented as follows:

[0068] ,

[0069] In the formula, Represents the new node feature vector set of all nodes in the last layer, Represents the new node feature vector of node i in the last layer, represents the pooling function, represents the normalization coefficient, represents the sum of the eigenvectors of all nodes, Indicates the total number of nodes, Represents the new node feature vector of node i in layer L.

[0070] The data acquisition and preprocessing module cleans, filters and standardizes the raw data, while the topology modeling and analysis module uses the graph neural network algorithm to accurately model the topology structure of the distribution network to ensure the accuracy of fault analysis.

[0071] The fault analysis module includes a fault detection unit, a fault location unit and a fault isolation unit;

[0072] The fault detection unit collects data from the data acquisition and preprocessing module in real time to perform fault anomaly detection;

[0073] The fault location unit calculates a fault probability score based on the detection result of the fault detection unit, combines the fault probability score with the graph-level representation vector, and determines the specific location of the fault;

[0074] The fault isolation unit calculates an optimal switch operation sequence according to the specific location of the fault;

[0075] The fault detection unit collects data from the data acquisition and preprocessing module in real time. , define an anomaly score function to measure the degree to which the data at a given time point deviates from the normal range. The implementation formula is:

[0076] ,

[0077] In the formula, Represents a data point at time t The anomaly score, represents a data point at time t, represents the weight of the i-th distribution network state feature, represents the value of the state characteristic of the ith distribution network at time t, represents the mean value of the state characteristics of the ith distribution network, represents the standard deviation of the i-th distribution network state feature, and n represents the number of distribution network state features.

[0078] The fault location unit calculates the fault probability of each node according to the fault anomaly score. There are N nodes, and the probability score of each node j having a fault is , the fault probability score realization formula is:

[0079] ,

[0080] In the formula, Representation node The anomaly score, N represents the total number of nodes, represents the natural exponential function, which is used to convert anomaly scores into positive values ​​and amplify the differences, represents the sum of the exponentials of all node anomaly scores;

[0081] The specific location of the fault is determined based on the obtained fault probability score combined with the graph-level representation vector. The implementation formula is:

[0082] ,

[0083] In the formula, represents the fault score of node j, represents the new node feature vector of node i in the last layer, g represents the graph level representation vector, Represents a similarity function, which calculates the similarity between vectors and scores them according to the fault All nodes are sorted, and the node with the highest score is the specific location of the node failure.

[0084] The fault analysis module includes three units: fault detection, location, and isolation. It locates the fault quickly and accurately by using anomaly score functions, fault probability score calculations, and combining graph-level representation vectors.

[0085] The fault isolation unit calculates the optimal switch operation sequence according to the specific fault location, and the optimal switch state is set as , the optimal isolation implementation formula is:

[0086] ,

[0087] Subject to:

[0088] ,

[0089] ,

[0090] In the formula, represents the switch state connecting node i and node j, Indicates switch The cost of λ is the adjustment parameter. represents the sum of the costs of all switches, represents the sum of the products of the fault score and the switch status, and Indicates that each node can only be connected to one node.

[0091] Among them, the fault recovery strategy module formulates a fault recovery strategy based on the positioning results and optimal isolation results provided by fault analysis, and according to the status of the fault area and non-fault area, executes the fault recovery strategy and monitors the quasi-status of the distribution network in real time, and obtains a graph-level representation vector through pooling aggregation, thereby improving the accuracy and efficiency of fault analysis.

[0092] Among them, the communication interaction module provides an intuitive operation interface, allowing operation and maintenance personnel to monitor the status of the distribution network and perform control operations, generate instructions based on the output of the fault recovery strategy module, send the instructions to the control center, collect distribution network status and recovery progress information, and send status reports to the control center.

[0093] In this example, specifically: the topology modeling and analysis module performs distribution network topology analysis based on the constructed distribution network topology model, and extracts node and edge features from the distribution network topology model; suppose an undirected graph is G=(V, E), where V represents the node set, E represents the edge set, and each node There are associated node feature vectors , each edge There are associated edge eigenvectors , calculate the new node feature vector of each node i in the l+1 layer through the graph neural network , expressed in the l+1 layer as:

[0094] ,

[0095] In the formula, represents the weight matrix of the lth layer, represents the feature vector of node i, Represents the feature vector of node i's neighbor node j The weighted sum of represents the activation function, N(i) represents the neighbor set of node i, represents the standard factor, for , Represents the bias vector of the lth layer.

[0096] In this example, specifically: based on the obtained new node feature vector, the last layer of new node feature vectors of all nodes are aggregated into the entire graph through pooling. The graph-level representation vector g is implemented as follows:

[0097] ,

[0098] In the formula, Represents the new node feature vector set of all nodes in the last layer, Represents the new node feature vector of node i in the last layer, represents the pooling function, represents the normalization coefficient, represents the sum of the eigenvectors of all nodes, Indicates the total number of nodes, Represents the new node feature vector of node i in layer L.

[0099] In this example, specifically: the fault location unit calculates the fault probability of each node according to the fault anomaly score. There are N nodes, and the probability score of each node j failing is , the fault probability score realization formula is:

[0100] ,

[0101] In the formula, represents the abnormal score of node j, N represents the total number of nodes, represents the natural exponential function, which is used to convert anomaly scores into positive values ​​and amplify the differences, represents the sum of the exponentials of all node anomaly scores;

[0102] The specific location of the fault is determined based on the obtained fault probability score combined with the graph-level representation vector. The implementation formula is:

[0103] ,

[0104] In the formula, represents the fault score of node j, represents the new node feature vector of node i in the last layer, g represents the graph level representation vector, Represents a similarity function, which calculates the similarity between vectors and scores them according to the fault All nodes are sorted, and the node with the highest score is the specific location of the node failure.

[0105] In this example, specifically: the fault isolation unit calculates the optimal switch operation sequence according to the specific fault location, and assumes that the optimal switch state is , the optimal isolation implementation formula is:

[0106] ,

[0107] Subject to:

[0108] ,

[0109] ,

[0110] In the formula, represents the switch state connecting node i and node j, Indicates switch The cost of λ is the adjustment parameter. represents the sum of the costs of all switches, represents the sum of the products of the fault score and the switch status, and Indicates that each node can only be connected to one node.

[0111] Working principle: The deployed IoT sensors monitor the status of the distribution network in real time. The data acquisition and pre-processing module collects the data output by the IoT sensing module in real time and performs preliminary processing such as data cleaning, denoising, and compression.

[0112] The topology modeling and analysis module constructs a topology model of the distribution network based on the basic data of the distribution network. Through technologies such as graph databases, the connection relationship between devices is graphically represented. Real-time monitoring data is combined with the topology model to dynamically update the status of nodes and edges. The fault analysis module includes a fault detection unit, a fault location unit, and a fault isolation unit. The fault detection unit uses algorithms such as anomaly score functions to detect anomalies in the data in real time and determine whether a fault has occurred in the distribution network. The fault location unit calculates the fault probability score and determines the specific location of the fault based on the detection results of the fault detection unit, combined with algorithms such as graph neural networks. The fault isolation unit calculates the optimal switching operation sequence based on the fault location to isolate the fault area and prevent the fault from spreading. The fault recovery strategy module formulates and executes the fault recovery strategy based on the location results and optimal isolation results provided by the fault analysis module. It monitors the distribution network status in real time, evaluates the recovery effect, and adjusts the recovery strategy as needed. The communication interaction module provides an intuitive operation interface, enabling operation and maintenance personnel to easily monitor the distribution network status and perform control operations, receive the output instructions of the fault recovery strategy module, and send them to the control center.

[0113] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

[0114] The present invention and its embodiments are described above. This description is not restrictive. The drawings show only one embodiment of the present invention, and the actual structure is not limited thereto. In short, if a person skilled in the art is inspired by this and, without departing from the purpose of the present invention, designs structures and embodiments similar to this technical solution without inventiveness, they shall fall within the scope of protection of the present invention.

Claims

1. The online fault analysis and judgment system for distribution network based on the Internet of Things is characterized by: It includes IoT perception module, data acquisition and preprocessing module, topology modeling and analysis module, fault analysis module, fault recovery strategy module, and communication interaction module; The IoT sensing module monitors the state of the power distribution network in real time through the deployed IoT sensors; The data acquisition and preprocessing module is used to collect data from the Internet of Things perception module in real time and perform preprocessing; The topology modeling and analysis module is used to construct a distribution network topology model and perform topology analysis; The fault analysis module monitors the abnormal situation of the distribution network in real time, determines the specific location of the fault, and calculates the optimal solution to isolate the fault area; The fault recovery strategy module analyzes the fault analysis results and formulates a fault recovery strategy; The communication interaction module is used to provide an intuitive operation interface, allowing operation and maintenance personnel to monitor the status of the distribution network and perform control operations; The fault analysis module includes a fault detection unit, a fault location unit and a fault isolation unit; The fault detection unit collects data from the data acquisition and preprocessing module in real time to perform fault anomaly detection; The fault location unit calculates a fault probability score based on the detection result of the fault detection unit, combines the fault probability score with the graph-level representation vector, and determines the specific location of the fault; The fault isolation unit calculates an optimal switch operation sequence according to the specific location of the fault; The fault detection unit collects data from the data acquisition and preprocessing module in real time. , define an anomaly score function to measure the degree to which the data at a given time point deviates from the normal range. The implementation formula is: ， In the formula, Represents a data point at time t The anomaly score, represents a data point at time t, represents the weight of the i-th distribution network state feature, represents the value of the state characteristic of the ith distribution network at time t, represents the mean value of the state characteristics of the ith distribution network, represents the standard deviation of the i-th distribution network state feature, and n represents the number of distribution network state features.

2. The online fault analysis and judgment system for distribution network based on the Internet of Things according to claim 1 is characterized by: The topology modeling and analysis module collects basic data of the distribution network from the data acquisition and preprocessing module, obtains the location coordinates, geographical distribution information, and line data information of all devices in the distribution network, determines key nodes in the distribution network based on the collected basic data, defines attributes for each key node, determines the physical connection between nodes based on the actual connection mode of the distribution network, defines attributes for each edge, designs a model of a graph database based on the definitions of nodes and edges, imports the collected data into the graph database, and establishes connection relationships between nodes in the graph database based on the actual connection mode between devices to form a distribution network topology model; The status of the distribution network is monitored in real time according to the IoT perception module, and the real-time monitoring data is combined with the constructed distribution network topology model to update the status of the nodes and edges in the distribution network topology model.

3. The online fault analysis and judgment system for distribution network based on the Internet of Things according to claim 2 is characterized in that: The topology modeling and analysis module performs distribution network topology analysis based on the constructed distribution network topology model and extracts node and edge features from the distribution network topology model; suppose an undirected graph is G=(V, E), where V represents the node set, E represents the edge set, and each node There are associated node feature vectors , each edge Associated edge eigenvectors , calculate the new node feature vector of each node i in the l+1 layer through the graph neural network , expressed in the l+1 layer as: ， In the formula, represents the weight matrix of the lth layer, represents the feature vector of node i, Represents the feature vector of node i's neighbor node j The weighted sum of represents the activation function, N(i) represents the neighbor set of node i, represents the standard factor, Represents the bias vector of the lth layer.

4. The online fault analysis and judgment system for distribution network based on the Internet of Things according to claim 3 is characterized by: According to the obtained new node feature vector, the last layer of new node feature vectors of all nodes are aggregated into the entire graph through pooling. The graph-level representation vector g is implemented as follows: ， In the formula, Represents the new node feature vector set of all nodes in the last layer, Represents the new node feature vector of node i in the last layer, represents the pooling function, represents the normalization coefficient, represents the sum of the eigenvectors of all nodes, Indicates the total number of nodes, Represents the new node feature vector of node i in layer L.

5. The online fault analysis and judgment system for distribution network based on the Internet of Things according to claim 1 is characterized in that: The fault location unit calculates the fault probability of each node according to the fault anomaly score. There are N nodes, and the probability score of each node j failing is , the fault probability score realization formula is: , In the formula, Representation node The anomaly score, N represents the total number of nodes, represents the natural exponential function, which is used to convert anomaly scores into positive values ​​and amplify the differences, represents the sum of the exponentials of all node anomaly scores; The specific location of the fault is determined based on the obtained fault probability score combined with the graph-level representation vector. The implementation formula is: ， In the formula, represents the fault score of node j, represents the new node feature vector of node i in the last layer, g represents the graph level representation vector, Represents a similarity function, which calculates the similarity between vectors and scores them according to the fault All nodes are sorted, and the node with the highest score is the specific location of the node failure.

6. The online fault analysis and judgment system for distribution network based on the Internet of Things according to claim 5 is characterized in that: The fault isolation unit calculates the optimal switch operation sequence according to the specific fault location, and the optimal switch state is set to , the optimal isolation implementation formula is: ， Subject to: ， ， In the formula, represents the switch state connecting node i and node j, Indicates switch The cost of λ is the adjustment parameter. represents the sum of the costs of all switches, represents the sum of the products of the fault score and the switch status, and Indicates that each node can only be connected to one node.

7. The online fault analysis and judgment system for distribution network based on the Internet of Things according to claim 1 is characterized in that: The fault recovery strategy module formulates a fault recovery strategy based on the positioning results and optimal isolation results provided by the fault analysis and the status of the fault area and the non-fault area, executes the fault recovery strategy and monitors the distribution network status in real time.

8. The online fault analysis and judgment system for distribution network based on the Internet of Things according to claim 1 is characterized by: The communication interaction module provides an intuitive operation interface, allowing operation and maintenance personnel to monitor the distribution network status and perform control operations, generate instructions based on the output of the fault recovery strategy module, send the instructions to the control center, collect distribution network status and recovery progress information, and send status reports to the control center.

Images