Method and system for detecting safe operation of power distribution equipment

Abstract

The invention relates to the field of electric power distribution, in particular to an electric power distribution equipment safe operation detection method and system, and the method comprises the steps: constructing and training a graph neural network related to electric power distribution equipment; nodes of the graph neural network represent each device, and operation data of each power distribution device is collected; the edges represent the electrical connection relationship between the devices; the input is the operation data and the electrical connection relation of the equipment, and the output is the fault probability distribution of each equipment for each fault type; screening nodes according to the output fault probability distribution to obtain predicted fault nodes, calculating the migration influence of the predicted fault nodes on other nodes under the condition of each fault type, and adjusting the fault probability of the node according to the migration influence of each node to obtain the predicted fault probability. The method has the effect of improving the accuracy and reliability of fault prediction.

Description

Technical Field

[0001] The present application relates to the field of electric power distribution, and in particular to a method and system for detecting the safe operation of electric power distribution equipment. Background Art

[0002] Power distribution equipment refers to various devices used for power distribution and control, transmitting electricity from power plants or substations to end users (such as homes, factories, and commercial buildings). Power distribution equipment encompasses every aspect of power distribution, control, protection, monitoring, and intelligent management. They play a critical role in the power system, ensuring the safe, reliable, and efficient distribution of electricity. Graph Neural Networks (GNNs), as an emerging machine learning method, are particularly well-suited for problems with complex relationships and structured data. Therefore, they can be applied to the safe operation monitoring of power distribution equipment, such as determining whether a device is in a fault state (normal / abnormal). Each node corresponds to a class probability distribution.

[0003] However, when a device fails, it not only affects its own operational status but also has a cascading impact on adjacent or related devices through electrical connections. For example, if a node fails, it will affect related nodes, thereby changing the failure probability of other related nodes. This impact can cause the failure probability to propagate and spread among devices. Existing technologies cannot fully reflect the impact of interactions between devices on overall system security, and may therefore underestimate or overestimate the actual failure risk of certain devices. Summary of the Invention

[0004] In order to solve the above-mentioned technical problem of being unable to fully reflect the impact of the interaction between devices on the overall system safety, the present application provides a method and system for detecting the safe operation of power distribution equipment.

[0005] In the first aspect, the present application provides a method for detecting the safe operation of electric power distribution equipment, which adopts the following technical scheme: a method for detecting the safe operation of electric power distribution equipment, comprising the steps of: constructing and training a graph neural network for electric power distribution equipment; the nodes of the graph neural network represent each device, and the operating data of each distribution device is collected; the edges represent the electrical connection relationship between the devices; the input is the operating data and electrical connection relationship of the equipment, and the output is the fault probability distribution of each device for each fault type; according to the output fault probability distribution, the nodes are filtered to obtain predicted fault nodes, the migration impact of the predicted fault nodes on other nodes under each fault type is calculated, and the failure probability of the node is adjusted according to the migration impact of each node to obtain the predicted failure probability; the method for calculating the migration impact is: adaptively setting the node influence order according to the size of the graph neural network; according to the historical data set in the graph neural network training process, the similarity between each node in each fault type is calculated, and the similarity is normalized as the edge weight; the migration impact is obtained according to the edge weight of the edge between nodes, the failure probability of the fault node, and the influence order, the edge weight, the failure probability and the migration impact are positively proportional, and the influence order is inversely proportional to the migration impact.

[0006] Optionally, the node influence order is adaptively set according to the size of the graph neural network, including: rounding the product of the ratio of the node order of the node to the maximum node order of the graph neural network and a preset fixed order as the node influence order.

[0007] Optionally, the similarity is calculated by constructing a fault label sequence for each node, counting the number of time points at which faults occur in both fault label sequences, and taking the ratio of the number of time points to the total number of time points with fault labels as the similarity.

[0008] Optionally, the migration impact is calculated as: ;in, Representation node For Node The impact of migration, Representation node With node The edge weight between the edges, Representation node The order of influence, Representation node There is a fault probability of failure; Indicates An exponential function with base .

[0009] Optionally, the migration impact is calculated as: ;in, Representation node For Node The impact of migration, Representation node With node The edge weight between the edges, Representation node The order of influence, Representation node There is a fault probability of failure.

[0010] Optionally, nodes are filtered according to the output fault probability distribution to obtain predicted fault nodes, including setting a fault probability threshold for each fault type. When the failure probability corresponding to any fault type in the fault probability distribution is greater than the failure probability threshold, the node is determined to be a predicted fault node.

[0011] Optionally, nodes are filtered according to the output fault probability distribution to obtain predicted fault nodes, and abnormal nodes are identified using an anomaly detection algorithm as predicted fault nodes.

[0012] Optionally, in adjusting the failure probability of each node according to the migration impact of the node to obtain the predicted failure probability, the product of the migration impact and the failure probability is used as the predicted failure probability.

[0013] Optionally, the operating data of the device includes current, voltage or power.

[0014] In a second aspect, the present application provides a system for detecting safe operation of power distribution equipment, which adopts the following technical solutions: A power distribution equipment safe operation detection system includes: a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the power distribution equipment safe operation detection method described above is implemented.

[0015] The beneficial effect is that the above-mentioned power distribution equipment safe operation detection method is generated into a computer program and stored in a memory so as to be loaded and executed by a processor, thereby making a system based on the memory and the processor for easy use.

[0016] This application has the following technical effects: 1. This application not only focuses on the operating status and failure probability distribution of individual devices, but also fully considers the electrical connections between devices and their cascading effects. By combining migration effects with failure probabilities to adjust predicted failure probabilities, it effectively captures the dynamic interactions between devices, significantly improving the accuracy and reliability of fault prediction and avoiding underestimation or overestimation of failure risk caused by ignoring interactions between devices.

[0017] 2. The migration impact calculation method combines the size of the graph neural network, the node similarity in the historical dataset, and the characteristics of the faulty nodes to collaboratively reflect the interactions between power distribution equipment and the fault propagation patterns. Specifically, the node influence order is adaptively set according to the size of the graph neural network to ensure that high-order nodes have a larger influence range. The similarity between nodes in the historical dataset is calculated and normalized into edge weights to quantify the strength of the association between nodes. Finally, the migration impact is calculated by combining the edge weight, fault probability, and influence order. The edge weight and fault probability are directly proportional to the migration impact, while the influence order is inversely proportional to the migration impact. This accurately captures the propagation effect of faults between devices, improving the accuracy of fault prediction and the ability to assess system safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 This is a method flow chart of a method for detecting safe operation of electric power distribution equipment in an embodiment of the present application.

[0019] Figure 2 This is a flow chart of a method for calculating migration impact in a method for detecting safe operation of electric power distribution equipment in an embodiment of the present application. DETAILED DESCRIPTION

[0020] The present application discloses a method for detecting safe operation of power distribution equipment, referring to Figure 1 , including steps S1 and S2, specifically as follows: S1: Build and train a graph neural network for power distribution equipment.

[0021] In a graph neural network (GNN), nodes represent individual devices, collecting operational data from each power distribution device. Edges represent the electrical connections between devices. Each node represents a power distribution device (such as a transformer or circuit breaker). The node's feature vector contains the device's operational data, such as current, voltage, or power. Each edge represents the electrical connection between two devices.

[0022] For example, a power distribution system may include: transformer A, circuit breaker B, switch C, and switch D. For each device, operating data such as current, voltage, or power is collected. The electrical connections between devices can be represented by an adjacency matrix. The device connection data is an adjacency matrix. For example, assume that the power distribution system includes the following devices and their connection relationships: transformer A is connected to circuit breaker B and switch D, circuit breaker B is connected to switch C, switch C has no further connections, and switch D is only connected to transformer A. The adjacency matrix can be represented as: , the first line :Transformer A is connected to circuit breaker B and switch D. Second row :Circuit breaker B is connected to transformer A and switch C. Third row :Switch C is only connected to circuit breaker B. Row 4 : Switch D is only connected to transformer A.

[0023] The graph neural network is trained using historical data sets, which are composed of two-dimensional sequences. For example, for current data, the historical data set is the current value corresponding to each moment. Each current value is set with a label, which is 1 or 0. 1 indicates a fault and 0 indicates normal. For example, transformer A: (Fault at moment 3). The construction and training of graph neural networks are both existing technologies. The input of a graph neural network is the operating data and electrical connection relationships of the equipment, and the output is the failure probability distribution for each equipment for each fault type. For example, the output might be Transformer A: Failure probability 0.9.

[0024] S2: Filter nodes based on the output failure probability distribution to obtain predicted failure nodes, calculate the migration impact of the predicted failure nodes on other nodes under each failure type, and adjust the failure probability of each node based on the migration impact to obtain the predicted failure probability.

[0025] In one embodiment, the method for screening nodes is: setting a fault probability threshold for each fault type, and when the fault probability corresponding to any fault type in the fault probability distribution is greater than the fault probability threshold, the node is determined to be a predicted fault node.

[0026] For example, the fault threshold is 0.5, the fault probability of transformer A is 0.9, and the node (transformer A) is a faulty node. The setting of the fault threshold can be adjusted according to the actual application scenario and will not be repeated here.

[0027] In other embodiments, an abnormality detection algorithm can also be used to identify abnormal nodes as predicted fault nodes. Common abnormality detection algorithms include Isolation Forest, Local Outlier Factor (LOF), etc., and the prior art will not be described in detail. Through the abnormality detection algorithm, nodes that are significantly different from the normal operating mode can be detected. Taking the Isolation Forest algorithm as an example, the current value of transformer A is In the figure, the current value (25) at the third moment obviously deviates from the normal range and is marked as abnormal. The node (transformer A) is a fault node.

[0028] Reference Figure 2 The migration impact calculation method includes steps S20 to S22: S20: Adaptively set the node influence order according to the size of the graph neural network.

[0029] In one embodiment, the node influence order is obtained by rounding the product of the ratio of the node order to the maximum node order of the graph neural network and the preset fixed order.

[0030] Node order refers to the total number of nodes a node can indirectly influence through connections, including direct neighbors and their neighbors. For example, the node order of a node a refers to the total number of nodes reachable from that node via edges (including itself). For example, if node a is directly connected to node b, and node b is connected to node c, then the node order of node a is 3 (including node a, node b, and node c). The maximum node order refers to the maximum order of all nodes in a graph neural network.

[0031] A large maximum node order indicates that the graph contains some highly connected nodes, resulting in a large overall graph size. A small maximum node order indicates that the connectivity between nodes in the graph is weak, resulting in a small overall graph size. By adaptively setting the node influence order based on the size of the graph neural network, subsequent steps can more accurately quantify the scope and extent of fault or anomaly propagation between nodes.

[0032] S21: Based on the historical data set during the graph neural network training process, the similarity between each node in each fault type is calculated, and the similarity is normalized as the edge weight.

[0033] In one embodiment, the similarity can be calculated by constructing a fault label sequence for each node, counting the number of time points at which faults occur in both fault label sequences, and using the ratio of the number of time points to the total number of time points at which fault labels occur as the similarity. For example, if the fault label of sequence 1 is 1 at time 3 and the fault label of sequence 2 is 1 at time 3, then time 3 is counted in the number of time points at which faults occur. If the fault label of sequence 1 is 0 at time 4 and the fault label of sequence 3 is 1 at time 4, then time 4 is not counted in the number of time points at which faults occur. Normalization can use the existing standard normalization method, and the prior art will not be described in detail.

[0034] S22: The migration impact is obtained based on the edge weight of the edge between nodes, the failure probability of the faulty node, and the influence order. The edge weight and failure probability are directly proportional to the migration impact, while the influence order is inversely proportional to the migration impact.

[0035] In one embodiment, the migration impact is calculated as follows: ;in, Representation node For Node The impact of migration, Representation node With node The edge weight between the nodes is larger. For Node The stronger the migration effect, the weaker the migration effect.

[0036] Representation node The influence order of the node If the node The influence range of is large, then it has a large impact on a single node. The migration impact will be reduced (because the total impact is spread over more nodes). Representation node There is a fault The greater the failure probability, the greater the impact on other nodes; Indicates This embodiment involves exponential operations, emphasizing the exponential decay effect, and is suitable for describing situations where the impact rapidly decreases with distance, and is applicable to scenarios where the impact rapidly decreases with distance (such as energy attenuation in signal propagation).

[0037] In other embodiments, the calculation formula for migration impact may also be: ;in, Representation node For Node The impact of migration, Representation node With node The edge weight between the edges, Representation node The order of influence, Representation node There is a fault This provides another mathematical expression for the impact of migration. This formula is simple to calculate, has low computational cost, and emphasizes linear distribution logic, making it suitable for describing situations where the impact is evenly distributed across multiple nodes. You can choose the migration impact calculation formula based on your application needs.

[0038] The method for adjusting the failure probability of each node according to the migration impact of the node to obtain the predicted failure probability is: taking the product of the migration impact and the failure probability as the predicted failure probability.

[0039] An embodiment of the present application also discloses a power distribution equipment safe operation detection system, including a processor and a memory, the memory storing computer program instructions, and when the computer program instructions are executed by the processor, the power distribution equipment safe operation detection method according to the present application is implemented.

[0040] The above system also includes other components well known to those skilled in the art, such as a communication bus and a communication interface. The configuration and functions of these components are known in the art and will not be described in detail here.

[0041] In this application, the aforementioned memory can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. For example, a computer-readable storage medium can be any suitable magnetic storage medium or magneto-optical storage medium, such as resistive random access memory, dynamic random access memory, static random access memory, etc., or any other medium that can be used to store the required information and can be accessed by an application, module, or both. Any such computer storage medium can be part of the device or accessible or connectable to the device.

[0042] The above are all preferred embodiments of the present application, and are not intended to limit the scope of protection of the present application. Therefore, any equivalent changes made based on the structure, shape, and principle of the present application should be included in the scope of protection of the present application.

Claims

1. A method for detecting safe operation of power distribution equipment, characterized in that: Including steps: Build and train a graph neural network for power distribution equipment. The nodes of the graph neural network represent each device and collect operating data from each distribution device. The edges represent the electrical connections between devices. The input is the operating data and electrical connection relationship of the device, and the output is the failure probability distribution of each device for each fault type. Filter nodes based on the output failure probability distribution to obtain predicted failure nodes, calculate the migration impact of predicted failure nodes on other nodes under each failure type, and adjust the failure probability of each node based on its migration impact to obtain the predicted failure probability; The method for calculating migration impact is as follows: adaptively set the node influence order according to the size of the graph neural network; calculate the similarity between nodes in each fault type based on the historical data set during the graph neural network training process, and normalize the similarity as the edge weight; obtain the migration impact based on the edge weight of the edge between nodes, the failure probability of the fault node, and the influence order. The edge weight and failure probability are directly proportional to the migration impact, and the influence order is inversely proportional to the migration impact.

2. The method for detecting safe operation of electric power distribution equipment according to claim 1, characterized in that: The node influence order is adaptively set according to the size of the graph neural network, including: taking the product of the ratio of the node order of the node to the maximum node order of the graph neural network and the preset fixed order as the node influence order.

3. The method for detecting safe operation of electric power distribution equipment according to claim 1, characterized in that: The similarity is calculated as follows: construct a fault label sequence for each node, count the number of time points where both fault label sequences have faults, and use the ratio of the number of time points to the total number of time points with fault labels as the similarity.

4. The method for detecting safe operation of electric power distribution equipment according to claim 1, characterized in that: The migration impact is calculated as: ;in, Representation node For Node The impact of migration, Representation node With node The edge weight between the edges, Representation node The order of influence, Representation node There is a fault probability of failure; Indicates An exponential function with base .

5. The method for detecting safe operation of electric power distribution equipment according to claim 1, characterized in that: The migration impact is calculated as: ;in, Representation node For Node The impact of migration, Representation node With node The edge weight between the edges, Representation node The order of influence, Representation node There is a fault probability of failure.

6. The method for detecting safe operation of electric power distribution equipment according to claim 1, characterized in that: Nodes are filtered according to the output fault probability distribution to obtain predicted fault nodes, including setting a fault probability threshold for each fault type. When the fault probability corresponding to any fault type in the fault probability distribution is greater than the fault probability threshold, the node is determined to be a predicted fault node.

7. The method for detecting safe operation of electric power distribution equipment according to claim 1, characterized in that: Nodes are filtered according to the output fault probability distribution to obtain predicted fault nodes, and abnormal nodes are identified as predicted fault nodes using an anomaly detection algorithm.

8. The method for detecting safe operation of electric power distribution equipment according to claim 1, characterized in that: The failure probability of each node is adjusted according to the migration impact of the node to obtain the predicted failure probability, and the product of the migration impact and the failure probability is used as the predicted failure probability.

9. The method for detecting safe operation of electric power distribution equipment according to claim 1, characterized in that: The operating data of the equipment includes current, voltage or power.

10. A power distribution equipment safe operation detection system, characterized in that: include: A processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the method for detecting safe operation of electric power distribution equipment according to any one of claims 1 to 9 is implemented.

Images