Capacity expansion method and apparatus for distribution transformers, device, and storage medium

Abstract

Provided are a capacity expansion method and apparatus for distribution transformers, a device, and a storage medium. The method comprises: generating a topological relationship diagram for a plurality of distribution transformers (101); in response to scheduling power between a first distribution transformer and a second distribution transformer, respectively collecting a power scheduling value output by the first distribution transformer and a power scheduling value output by the second distribution transformer (102); converting the topological relationship diagram into a power scheduling relationship diagram on the basis of the power scheduling value output by the first distribution transformer and the power scheduling value output by the second distribution transformer (103); respectively computing power scheduling levels of the plurality of distribution transformers on the basis of the power scheduling relationship diagram (104); during operation of the plurality of distribution transformers, respectively collecting operation data of the plurality of distribution transformers (105); converting the power scheduling relationship diagram into a graph neural network on the basis of the operation data (106); respectively identifying the operation states of the plurality of distribution transformers on the basis of the graph neural network (107); and selecting some distribution transformers from among the plurality of distribution transformers on the basis of the operation states and the power scheduling levels, and performing capacity expansion on the selected distribution transformers (108).

Description

Method, device and equipment for expanding capacity of distribution transformer and storage medium

[0001] The present application claims priority to the Chinese patent application No. 202410424930.3, filed on April 10, 2024, to the Chinese Patent Office, the whole content of the above application being incorporated herein by reference. TECHNICAL FIELD

[0002] The present application relates to the technical field of power grid, for example, to a method, device and equipment for expanding capacity of distribution transformer and storage medium. BACKGROUND

[0003] In the power grid, a station area is divided, that is, the power supply range or area of a distribution transformer is defined. With the development of economy, the number of electricity users in the station area is increasing, and the power equipment used by the electricity users is increasing, so that the station area occasionally experiences heavy overload, at which time, the station area experiences problems such as low voltage, increased line loss, and reduced power supply reliability.

[0004] In the related art, power is dispatched between two adjacent distribution transformers to dispatch the power of the distribution transformer with lower load to the distribution transformer with higher load, so as to realize power sharing.

[0005] However, power dispatching is a supplementary measure for the distribution transformer itself. With the increasing frequency of power dispatching, the volatility of voltage, power and other parameters between the station areas begins to increase, and the operation risk continues to increase.

[0006] SUMMARY

[0007] The present application provides a method, device and equipment for expanding capacity of distribution transformer and storage medium to solve the problem of how to increase the safety of station area operation.

[0008] According to an aspect of the present application, a method for expanding capacity of distribution transformer is provided, comprising:

[0009] generating a topology relationship graph for a plurality of distribution transformers;

[0010] in response to dispatching power between a first distribution transformer and a second distribution transformer, respectively collecting a power dispatching value output by the first distribution transformer and a power dispatching value output by the second distribution transformer;

[0011] converting the topology relationship graph into a power dispatching relationship graph according to the power dispatching value output by the first distribution transformer and the power dispatching value output by the second distribution transformer;

[0012] respectively calculating power dispatching levels of the plurality of distribution transformers according to the power dispatching relationship graph;

[0013] collecting operation data of the plurality of distribution transformers respectively when the plurality of distribution transformers are operating;

[0014] converting the power scheduling relationship graph into a graph neural network according to the operation data;

[0015] identifying operation states of the plurality of distribution transformers respectively according to the graph neural network;

[0016] screening part of the plurality of distribution transformers according to the operation states and the power scheduling levels, and expanding capacity of the screened part of the plurality of distribution transformers.

[0017] According to another aspect of the present application, there is provided an expansion device for a distribution transformer, comprising:

[0018] a topological relationship graph generation module configured to generate a topological relationship graph for a plurality of distribution transformers;

[0019] a power scheduling value collection module configured to collect a power scheduling value output by a first distribution transformer and a power scheduling value output by a second distribution transformer respectively in response to scheduling power between the first distribution transformer and the second distribution transformer;

[0020] a power scheduling relationship graph conversion module configured to convert the topological relationship graph into a power scheduling relationship graph according to the power scheduling value output by the first distribution transformer and the power scheduling value output by the second distribution transformer;

[0021] a power scheduling level calculation module configured to calculate power scheduling levels of the plurality of distribution transformers respectively according to the power scheduling relationship graph;

[0022] an operation data collection module configured to collect operation data of the plurality of distribution transformers respectively when the plurality of distribution transformers are operating;

[0023] a graph neural network conversion module configured to convert the power scheduling relationship graph into a graph neural network according to the operation data;

[0024] an operation state identification module configured to identify operation states of the plurality of distribution transformers respectively according to the graph neural network;

[0025] a comprehensive expansion module configured to screen part of the plurality of distribution transformers according to the operation states and the power scheduling levels, and expand capacity of the screened part of the plurality of distribution transformers.

[0026] According to another aspect of the present application, there is provided an electronic device, comprising:

[0027] at least one processor; and

[0028] a memory connected with the at least one processor in communication; wherein,

[0029] The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor to enable the at least one processor to perform the capacity expansion method of the power distribution transformer according to any one of the embodiments of the present application.

[0030] According to another aspect of the present application, a computer readable storage medium is provided, which stores a computer program for enabling a processor to implement the capacity expansion method of the power distribution transformer according to any one of the embodiments of the present application when executed. BRIEF DESCRIPTION OF DRAWINGS

[0031] Fig. 1 is a flow chart of a capacity expansion method of a power distribution transformer according to an embodiment of the present application;

[0032] Fig. 2 is an example diagram of a topology relationship diagram according to an embodiment of the present application;

[0033] Fig. 3 is a schematic diagram of a scheduling device according to an embodiment of the present application;

[0034] Fig. 4 is a schematic diagram of a power scheduling relationship diagram according to an embodiment of the present application;

[0035] Fig. 5 is a structural schematic diagram of a capacity expansion device of a power distribution transformer according to an embodiment of the present application;

[0036] Fig. 6 is a structural schematic diagram of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0037] The terms "first", "second", and the like in the description and claims of the present application and the above drawings are used to distinguish similar objects, and do not necessarily indicate a specific order or sequence. It should be understood that the data thus used can be interchanged, where appropriate, so that the embodiments of the present application described herein can encompass orders other than those illustrated or described herein. In addition, the terms "comprise" and "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, processes, methods, systems, products, or devices that comprise a series of steps or units shown in the embodiments of the present application can also include other processes, methods, systems, products, and devices that are not clearly listed, or other steps or units inherent to these processes, methods, systems, products, or devices.

[0038] FIG. 1 is a flowchart of a capacity expansion method of a distribution transformer according to an embodiment of the present application. The method can be performed by a capacity expansion device of a distribution transformer, which can be implemented in the form of hardware and / or software and can be configured in an electronic device. As shown in FIG. 1, the method includes steps 101-108.

[0039] In step 101, a topological relationship graph is generated for a plurality of distribution transformers.

[0040] In the construction of a plurality of distribution transformers, information of the plurality of distribution transformers and their cables is recorded in a geographic information system (GIS) of a power grid.

[0041] In actual application, as shown in FIG. 2, information of the plurality of distribution transformers and their cables can be read from the GIS system, the topological relationship thereof is analyzed, the plurality of distribution transformers are all converted into first nodes (i.e., A, B, C, D, E, F, G, H and I in FIG. 2), and the cables between adjacent two distribution transformers are converted into first edges, so as to convert the topological relationship into a graph, thereby obtaining the topological relationship graph.

[0042] In step 102, in response to scheduling power between a first distribution transformer and a second distribution transformer, a power scheduling value output by the first distribution transformer and a power scheduling value output by the second distribution transformer are respectively collected.

[0043] For example, adjacent two distribution transformers can be close to each other in position, and a scheduling device can be arranged between the adjacent two distribution transformers, so that the adjacent two distribution transformers can schedule power with each other.

[0044] For example, as shown in FIG. 3, the scheduling device is a rectifier and an inverter connected with each other between adjacent two distribution transformers.

[0045] The rectifier is used to convert three-phase alternating current into direct current, and the inverter is used to convert direct current into three-phase alternating current, and the rectifier and the inverter cooperatively realize scheduling of power.

[0046] According to the conditions of the two distribution transformers, the rectifier is controlled to absorb corresponding power (i.e., active power and reactive power), and the inverter is controlled to output corresponding power (i.e., active power and reactive power), so as to schedule power between the two distribution transformers and realize capacity sharing between the two distribution transformers.

[0047] If the power is scheduled by using the scheduling device between the two power distribution transformers, the power scheduling value output by the first power distribution transformer and the power scheduling value output by the second power distribution transformer can be collected, and the information (such as ID) of the two power distribution transformers and the information (such as time, value number, etc.) of the power scheduling value are recorded in the system of the power grid.

[0048] The power scheduling value refers to the power value output by one of the power distribution transformers to the other power distribution transformer. Considering the efficiency of the rectifier, the efficiency of the inverter and other factors, the power value received by the other power distribution transformer has a certain loss.

[0049] In 103, the topology relationship graph is converted into a power scheduling relationship graph according to the power scheduling value output by the first power distribution transformer and the power scheduling value output by the second power distribution transformer.

[0050] Generally, the scheduling device is installed between the power distribution transformers based on the original line structure, so the scheduling relationship of the power between the power distribution transformers is established on the basis of the topology relationship, and therefore the topology relationship graph can be reused, and the topology relationship graph is modified to a certain extent on the basis of the power scheduling value, so as to obtain the power scheduling relationship graph representing the scheduling relationship, and the calculation amount of generating the power scheduling relationship graph can be reduced.

[0051] In an embodiment of the present application, 103 can include the following steps:

[0052] In 1031, each first node in the topology relationship graph is set as a second node corresponding to the first node in the power scheduling relationship graph, and each first edge in the topology relationship graph is set as a second edge corresponding to the first edge in the power scheduling relationship graph.

[0053] In the embodiment, a blank scheduling graph can be generated, and each first node in the topology relationship graph is directly reused as a second node corresponding to the first node in the power scheduling relationship graph, and each first edge in the topology relationship graph is directly reused as a second edge corresponding to the first edge in the power scheduling relationship graph.

[0054] The first node and the second node corresponding to the first node represent a power distribution transformer, the first edge represents a cable between two power distribution transformers, and the second edge represents a scheduling device corresponding to the first edge between the two power distribution transformers. The scheduling device is a proper subset of the cable, and at this time, the first edge is equal to the second edge, and therefore the second edge needs to be pruned subsequently.

[0055] In 1032, the first power distribution transformer and the second power distribution transformer are respectively set as a first scheduling transformer and a second scheduling transformer.

[0056] In a specific implementation, two power distribution transformers that generate the power scheduling values, i.e., the first power distribution transformer and the second power distribution transformer, can be queried in the system of the power grid, and the first power distribution transformer and the second power distribution transformer are respectively taken as a first scheduling transformer and a second scheduling transformer.

[0057] In 1033, in the power scheduling relationship graph, a second edge connecting the first scheduling transformer and the second scheduling transformer at the same time is retained, and a second edge not connecting the first scheduling transformer and the second scheduling transformer at the same time is deleted.

[0058] All the second edges in the power scheduling relationship graph are traversed, and it is judged whether a corresponding scheduling device exists for each second edge.

[0059] If the two end points of the second edge are respectively the first scheduling transformer and the second scheduling transformer that generate the same power scheduling value, it indicates that a scheduling device exists between the first scheduling transformer and the second scheduling transformer, a corresponding scheduling device exists for the second edge, and the second edge can be retained at this time.

[0060] If the two end points of the second edge are not respectively the first scheduling transformer and the second scheduling transformer that generate the same power scheduling value, it indicates that no scheduling device exists between the first scheduling transformer and the second scheduling transformer, a corresponding scheduling device does not exist for the second edge, and the second edge can be deleted at this time.

[0061] For example, as shown in FIG. 2 and FIG. 4, the topology relationship graph is directly converted into the power scheduling relationship graph, the second edge between the second node A and the second node G, the second edge between the second node C and the second node F, and the second edge between the second node C and the second node G are respectively deleted on the basis of the power scheduling relationship graph, and the pruning of the second edge is completed.

[0062] In 1034, the second edge is configured with an edge weight according to the power scheduling value of the first scheduling transformer and the power scheduling value of the second scheduling transformer.

[0063] Generally, the first edge in the topology relationship graph does not have an edge weight, and when the second edge corresponding to the first edge is retained, the second edge can be configured with an edge weight in combination with the power scheduling value provided by the first scheduling transformer and the power scheduling value provided by the second scheduling transformer.

[0064] In an embodiment of the present application, 1034 can further include the following steps:

[0065] In 10341, a first rated power value of the first scheduling transformer and a second rated power value of the second scheduling transformer are queried.

[0066] The first rated power value of the first dispatch transformer and the second rated power value of the second dispatch transformer can be queried respectively.

[0067] In 10342, the first dispatch value and the second dispatch value are determined according to the power dispatch value of the first dispatch transformer and the power dispatch value of the second dispatch transformer.

[0068] In this embodiment, the power dispatch value can be divided into the first dispatch value and the second dispatch value in the dimension of the output.

[0069] The first dispatch value is the power dispatch value output by the first dispatch transformer to the second dispatch transformer, and the second dispatch value is the power dispatch value output by the second dispatch transformer to the first dispatch transformer.

[0070] In 10343, the ratio between the first dispatch value and the first rated power value is calculated to obtain the first dispatch ratio.

[0071] In this embodiment, the ratio between the first dispatch value and the first rated power value is calculated to obtain the first dispatch ratio, that is, the first dispatch ratio = the first dispatch value / the first rated power value.

[0072] In 10344, the ratio between the second dispatch value and the second rated power value is calculated to obtain the second dispatch ratio.

[0073] In this embodiment, the ratio between the first dispatch value and the second rated power value is calculated to obtain the second dispatch ratio, that is, the second dispatch ratio = the second dispatch value / the second rated power value.

[0074] Generally, a distribution transformer is responsible for power supply to the area where it is located, and does not supply power to adjacent areas. If the distribution transformer supplies power to adjacent areas, it will occupy a part of its own power, and the utilization rate of this part of power will decrease. The more power occupied, the greater the impact on itself.

[0075] Therefore, the first dispatch ratio can represent the fluctuation of the first dispatch transformer to a certain extent when dispatching power, and the second dispatch ratio can represent the fluctuation of the second dispatch transformer to a certain extent when dispatching power.

[0076] In 10345, the first dispatch value is configured with a first dispatch weight in the dimension of the first dispatch transformer, and the second dispatch value is configured with a second dispatch weight in the dimension of the second dispatch transformer.

[0077] For the first dispatch transformer, the first dispatch value can be configured with a first dispatch weight based on the dispatching significance of the second dispatch transformer to the first dispatch transformer.

[0078] For the second dispatch transformer, a second dispatch weight can be configured for its second dispatch value based on the dispatch significance of the first dispatch transformer to the second dispatch transformer.

[0079] In one example, in one aspect, a sum of the power dispatch values output by the first dispatch transformer to all other distribution transformers (including the second dispatch transformer) is calculated to obtain a first total power value, and a ratio between the first dispatch value and the first total power value is calculated to obtain a first dispatch weight, i.e., the first dispatch weight is represented as α = w1 / W 1总 , where α is the first dispatch weight, w1 is the first dispatch value, and W 1总 is the first total power value.

[0080] In another aspect, a sum of the power dispatch values output by the second dispatch transformer to all other distribution transformers (including the first dispatch transformer) is calculated to obtain a second total power value, and a ratio between the second dispatch value and the second total power value is calculated to obtain a second dispatch weight, i.e., the second dispatch weight is represented as β = w2 / W 2总 , where β is the second dispatch weight, w2 is the second dispatch value, and W 2总 is the second total power value.

[0081] In 10346, a product between the first dispatch value and the first dispatch weight, and a product between the second dispatch value and the second dispatch weight are added to obtain an edge weight of the second edge.

[0082] In the present embodiment, a product between the first dispatch value and the first dispatch weight is calculated, and a product between the second dispatch value and the second dispatch weight is calculated, and the two products are added to obtain an edge weight of the second edge, i.e., the edge weight of the second edge is a linear fusion of the first dispatch value and the second dispatch value.

[0083] Then, the edge weight of the second edge is represented as W 边 = w1*α + w2*β, where W 边 is the edge weight of the second edge, α is the first dispatch weight, w1 is the first dispatch value, β is the second dispatch weight, and w2 is the second dispatch value.

[0084] In the present embodiment, the steps of 10341-10346 described above can be executed in sequence, or selected steps can be executed. For example, only 10342, 10345 and 10246 can be executed, and 10341, 10343 and 10344 can not be executed.

[0085] In 104, power dispatch levels of the plurality of distribution transformers are calculated according to the power dispatch relationship graph.

[0086] In the embodiment, the power scheduling levels of the plurality of power distribution transformers are divided in the power scheduling relationship graph. The power scheduling level is the level of each power distribution transformer in the power scheduling.

[0087] In a specific implementation, a plurality of fluctuation ranges can be set in advance, and each fluctuation range is configured with a corresponding power scheduling level.

[0088] For each power distribution transformer, the fluctuation degree of the power distribution transformer in the power scheduling is calculated in the power scheduling relationship graph using PageRank.

[0089] The fluctuation degree is compared with the plurality of fluctuation ranges, the fluctuation range into which the fluctuation degree falls is queried in the plurality of preset fluctuation ranges, a target range is obtained, and the power scheduling level configured for the target range is assigned as the power scheduling level of the power distribution transformer.

[0090] In 105, the operation data of the plurality of power distribution transformers is collected when the plurality of power distribution transformers are running.

[0091] In the recent period of time, if each power distribution transformer is running normally, the operation data of each power distribution transformer can be collected by calling the plurality of sensors configured for each power distribution transformer, such as the temperature, voltage, current, power, and the like of the insulating oil.

[0092] In 106, the power scheduling relationship graph is converted into a graph neural network according to the operation data.

[0093] In the embodiment, a blank base graph of a graph neural network (GNN) can be generated, each second node in the power scheduling relationship graph is directly reused as a third node corresponding to the second node in the base graph of the graph neural network, and each second edge in the power scheduling relationship graph is directly reused as a third edge corresponding to the second edge in the base graph of the graph neural network.

[0094] The second node and the third node corresponding to the second node represent a power distribution transformer, and the third edge represents the association between the power distribution transformers. Subsequently, the third edge does not need to be pruned.

[0095] The operation data of the plurality of power distribution transformers is statistically analyzed, so as to extract the operation features of each power distribution transformer, such as the operation features in the time domain and the operation features in the frequency domain, and the operation features are written into the third node corresponding to the power distribution transformer.

[0096] So, the input of the graph neural network includes a base graph G=(V, E), where V is a set of third nodes (also referred to as vertices) and E is a set of third edges (edges), where any third edge e=(v i ,v j ) represents a third edge e connecting a third node v i and a third node v j , and each third node v corresponds to a multi-dimensional operating feature Feat.

[0097] The graph neural network is a generalized neural network based on a graph structure and is also a connection model that captures the dependency of the base graph through message passing between third nodes of the base graph.

[0098] The graph neural network can be divided into five categories, namely, a graph convolution network (GCN), a graph attention network (GAT), a graph auto encoder (GAE), a graph generative network, and a graph spatial-temporal network (ST-GNN).

[0099] In 107, the operating state of each power distribution transformer is identified according to the graph neural network.

[0100] In this embodiment, the graph neural network can be called to classify the plurality of power distribution transformers and identify the operating state of each power distribution transformer.

[0101] The graph neural network is a graph at the bottom layer as a computational graph, and learns neural network primitives by passing, converting, and aggregating the features of the nodes on the entire graph to generate an embedding vector (Embedding) of a single third node. The generated embedding vector can be used as the input of a differentiable prediction layer to classify the third nodes.

[0102] The graph neural network has an advantage in classifying power distribution transformers with strong correlation (i.e., power scheduling), can include power scheduling in the evaluation range, and improve the classification accuracy of the power distribution transformers, i.e., improve the accuracy of identifying the operating state of the power distribution transformers.

[0103] In 108, some power distribution transformers are selected from the plurality of power distribution transformers according to the operating state and the power scheduling level, and the selected power distribution transformers are expanded.

[0104] In actual application, the overall state of each distribution transformer can be identified in combination with the operation state and the power scheduling level of all distribution transformers, and some distribution transformers are selected from the multiple distribution transformers for capacity expansion (also referred to as capacity increase).

[0105] Generally, the most fundamental solution to the problem of area overload is to expand the capacity of the distribution transformer. However, the cost of capacity expansion is high, the reconstruction is large, and the planning and construction of the distribution network has a certain lag.

[0106] In this embodiment, the priority of capacity expansion of the selected distribution transformers is improved, which can effectively alleviate the volatility of the area.

[0107] In an embodiment of the present application, the operation state comprises a first state, a second state and a third state in sequence from normal to abnormal, that is, the abnormality degree of the second state is higher than that of the first state, and the abnormality degree of the third state is higher than that of the second state; the power scheduling level comprises a first level, a second level and a third level in sequence from small to large, that is, the fluctuation degree of the second level power scheduling is higher than that of the first level power scheduling, and the fluctuation degree of the third level power scheduling is higher than that of the second level power scheduling.

[0108] In this embodiment, 108 can include the following steps:

[0109] In 1081, for each distribution transformer, in response to the operation state of the distribution transformer being the third state and the power scheduling level of the distribution transformer being the second level or the third level, the distribution transformer is determined as a candidate transformer; or in response to the operation state of the distribution transformer being the second state and the power scheduling level of the distribution transformer being the third level, the distribution transformer is determined as a candidate transformer.

[0110] Each distribution transformer has both an operation state and a power scheduling level, and the operation state and the power scheduling level can be combined to determine whether each distribution transformer is a candidate transformer.

[0111] For each distribution transformer, if its operation state is the third state and its power scheduling level is the second level or the third level, the distribution transformer can be determined as a candidate transformer.

[0112] For each distribution transformer, if its operation state is the second state and its power scheduling level is the third level, the distribution transformer can be determined as a candidate transformer.

[0113] In 1082, the distance between intervals of all candidate transformers in the power scheduling relationship diagram is calculated.

[0114] In the embodiment, the logical distance between any two candidate transformers in the power scheduling relationship diagram can be calculated.

[0115] For example, in the power scheduling relationship diagram, the shortest path between any two candidate transformers can be planned.

[0116] If there are other candidate transformers on the shortest path, indicating that the current two candidate transformers basically have no power scheduling possibility, the current two candidate transformers are deleted.

[0117] If there are no other candidate transformers on the shortest path, indicating that the current two candidate transformers may have power scheduling possibility, the number of the second edges in the shortest path is taken as the distance between the current two candidate transformers.

[0118] In 1083, for each distance, in response to the distance being less than or equal to a preset threshold, one of the two candidate transformers corresponding to the distance is deleted.

[0119] In the embodiment, the distance between the two candidate transformers is compared with the preset threshold. If the distance is less than or equal to the preset threshold, indicating that the distance between the two candidate transformers corresponding to the distance is close, which is a social point on capacity expansion, and the performance gain of capacity expansion will be reduced if the two candidate transformers are expanded at the same time, one of the two candidate transformers can be deleted.

[0120] For each candidate transformer, the average value of all distances between the candidate transformer and other candidate transformers is calculated. The average value can represent the overall fluctuation degree of the areas around the candidate transformer to some extent. The greater the average value, the lower the overall fluctuation degree of the areas around the candidate transformer, and vice versa.

[0121] For example, when the distance is less than or equal to the preset threshold, the candidate transformer with the maximum average value among the two candidate transformers corresponding to the distance is deleted, and the candidate transformer with the minimum average value among the two candidate transformers is retained, which can bring higher performance gain to the areas around the candidate transformer and reduce the overall fluctuation degree of the areas around the candidate transformer.

[0122] In 1084, in response to the remaining distances being greater than the preset threshold respectively, it is determined to expand the candidate transformers corresponding to the remaining distances.

[0123] The two candidate transformers corresponding to the distance are deleted one by one, the number of candidate transformers is reduced, the number of distances less than or equal to the preset threshold in the remaining distances between the remaining candidate transformers is less and less, the influence of power scheduling is reduced, and until the remaining distances between each of the remaining candidate transformers and other candidate transformers are greater than the preset threshold. If all the remaining distances between a certain candidate transformer and other remaining candidate transformers are greater than the preset threshold, it indicates that the candidate transformer corresponding to the remaining distance belongs to the isolated point of expansion, and it can be determined that the candidate transformer corresponding to the remaining distance is expanded.

[0124] In the embodiment, a topology relationship graph is generated for a plurality of distribution transformers; in response to scheduling power between a first distribution transformer and a second distribution transformer, a power scheduling value output by the first distribution transformer and a power scheduling value output by the second distribution transformer are respectively collected; the topology relationship graph is converted into a power scheduling relationship graph according to the power scheduling value output by the first distribution transformer and the power scheduling value output by the second distribution transformer; power scheduling levels of the plurality of distribution transformers are respectively calculated according to the power scheduling relationship graph; when the plurality of distribution transformers are running, running data of the plurality of distribution transformers are respectively collected; the power scheduling relationship graph is converted into a graph neural network according to the running data; running states of the plurality of distribution transformers are respectively identified according to the graph neural network; and part of the distribution transformers are screened out from the plurality of distribution transformers according to the running states and the power scheduling levels, and the part of the distribution transformers are expanded. In the embodiment, part of the distribution transformers are screened out from two dimensions of running and power scheduling for expansion, so that the number of expansion is reduced as much as possible, the cost of expansion is controlled, and the part of the distribution transformers after expansion can maximize the gain in the performance of the transformer station, reduce the frequency of power scheduling, thereby stabilizing the running of the transformer station, reducing the volatility of the transformer station, and ensuring the safety of the transformer station.

[0125] FIG. 5 is a structural schematic diagram of an expansion device of a distribution transformer provided by an embodiment of the present application. As shown in FIG. 5, the device comprises:

[0126] A topology relationship graph generation module 501 is configured to generate a topology relationship graph for a plurality of distribution transformers;

[0127] A power scheduling value collection module 502 is configured to, in response to scheduling power between a first distribution transformer and a second distribution transformer, respectively collect a power scheduling value output by the first distribution transformer and a power scheduling value output by the second distribution transformer;

[0128] A power scheduling relationship graph conversion module 503 is configured to convert the topology relationship graph into a power scheduling relationship graph according to the power scheduling value output by the first distribution transformer and the power scheduling value output by the second distribution transformer.

[0129] The power scheduling level calculation module 504 is configured to calculate power scheduling levels of the plurality of distribution transformers respectively according to the power scheduling graph;

[0130] The operation data collection module 505 is configured to collect operation data of the plurality of distribution transformers respectively when the plurality of distribution transformers are operating;

[0131] The graph neural network conversion module 506 is configured to convert the power scheduling graph into a graph neural network according to the operation data;

[0132] The operation state recognition module 507 is configured to recognize operation states of the plurality of distribution transformers respectively according to the graph neural network;

[0133] The comprehensive expansion module 508 is configured to filter out part of the plurality of distribution transformers according to the operation states and the power scheduling levels, and expand the part of the distribution transformers.

[0134] In an embodiment of the present application, the power scheduling graph conversion module 503 comprises:

[0135] The graph multiplexing module is configured to set each first node in the topology graph as a second node corresponding to the first node in the power scheduling graph, and set each first edge in the topology graph as a second edge corresponding to the first edge in the power scheduling graph; the first node and the second node corresponding to the first node represent a distribution transformer;

[0136] The scheduling transformer query module is configured to set the first distribution transformer and the second distribution transformer as a first scheduling transformer and a second scheduling transformer respectively;

[0137] The edge pruning module is configured to retain the second edge connecting the first scheduling transformer and the second scheduling transformer simultaneously and delete the second edge not connecting the first scheduling transformer and the second scheduling transformer simultaneously in the power scheduling graph;

[0138] The edge weight configuration module is configured to configure an edge weight of the second edge according to the power scheduling value of the first scheduling transformer and the power scheduling value of the second scheduling transformer.

[0139] In an embodiment of the present application, the edge weight configuration module comprises:

[0140] The scheduling value determination module is configured to determine a first scheduling value and a second scheduling value according to the power scheduling value of the first scheduling transformer and the power scheduling value of the second scheduling transformer; the first scheduling value is a power scheduling value output by the first scheduling transformer to the second scheduling transformer, and the second scheduling value is a power scheduling value output by the second scheduling transformer to the first scheduling transformer;

[0141] The scheduling weight configuration module is configured to configure a first scheduling weight for the first scheduling value in the dimension of the first scheduling transformer and configure a second scheduling weight for the second scheduling value in the dimension of the second scheduling transformer;

[0142] The weighted sum module is configured to add a product of the first scheduling value and the first scheduling weight and a product of the second scheduling value and the second scheduling weight to obtain an edge weight of the second edge.

[0143] In an embodiment of the present application, the scheduling weight configuration module comprises:

[0144] The first total power value calculation module is configured to calculate a sum of the power scheduling values output by the first scheduling transformer to all other distribution transformers to obtain a first total power value;

[0145] The first scheduling weight calculation module is configured to calculate a ratio between the first scheduling value and the first total power value to obtain the first scheduling weight;

[0146] The second total power value calculation module is configured to calculate a sum of the power scheduling values output by the second scheduling transformer to all other distribution transformers to obtain a second total power value;

[0147] The second scheduling weight calculation module is configured to calculate a ratio between the second scheduling value and the second total power value to obtain the second scheduling weight.

[0148] In an embodiment of the present application, the power scheduling level calculation module 504 comprises:

[0149] The fluctuation degree calculation module is configured to calculate, for each distribution transformer, a fluctuation degree of each distribution transformer in power scheduling in the power scheduling relationship graph using a web page ranking;

[0150] The target range query module is configured to query a fluctuation range in which the fluctuation degree falls from a plurality of preset fluctuation ranges to obtain a target range; the plurality of preset fluctuation ranges are each configured with a power scheduling level;

[0151] The power scheduling level assignment module is configured to assign a power scheduling level configured for the target range to a power scheduling level of the power distribution transformer.

[0152] In an embodiment of the present application, the running states from normal to abnormal include a first state, a second state and a third state in sequence; and the power scheduling levels from small to large include a first level, a second level and a third level in sequence.

[0153] The comprehensive expansion module 508 includes:

[0154] The candidate transformer determination module is configured to, for each power distribution transformer, determine the power distribution transformer as a candidate transformer in response to that the running state of the power distribution transformer is the third state and the power scheduling level of the power distribution transformer is the second level or the third level; or determine the power distribution transformer as a candidate transformer in response to that the running state of the power distribution transformer is the second state and the power scheduling level of the power distribution transformer is the third level.

[0155] The distance calculation module is configured to calculate distances between intervals of all candidate transformers in the power scheduling relationship graph.

[0156] The community point deletion module is configured to, for each distance, delete one of two candidate transformers corresponding to the distance in response to that the distance is less than or equal to a preset threshold.

[0157] The expansion determination module is configured to determine to expand the candidate transformers corresponding to the remaining distances in response to that the remaining distances are greater than the preset threshold respectively.

[0158] In an embodiment of the present application, the distance calculation module includes:

[0159] The path planning module is configured to, in the power scheduling relationship graph, select two candidate transformers from the candidate transformers and plan a shortest path for the two candidate transformers.

[0160] The way point deletion module is configured to delete the two candidate transformers if there are other candidate transformers on the path.

[0161] The edge number statistics module is configured to take a number of second edges in the path as the distance between the two candidate transformers if there are no other candidate transformers on the path.

[0162] In an embodiment of the present application, the community point deletion module includes:

[0163] An average value calculation module is configured to calculate, for each candidate transformer, an average value of all distances between the candidate transformer and other candidate transformers;

[0164] An average value deletion module is configured to delete, from the candidate transformers, the candidate transformer with the maximum average value of distances between the candidate transformer and other candidate transformers, which is less than or equal to a preset threshold.

[0165] The power distribution transformer expansion device provided by the embodiments of the present application can execute the power distribution transformer expansion method provided by any of the embodiments of the present application, and has the corresponding function modules and beneficial effects of executing the power distribution transformer expansion method.

[0166] FIG. 6 shows a structural schematic diagram of an electronic device 10 that can be used to implement the embodiments of the present application. The electronic device is intended to represent various forms of digital computers, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. The components shown here, their connections and relationships, and their functions, are meant to be examples only, and are not intended to limit the implementations of the present application described and / or claimed in this document.

[0167] As shown in FIG. 6, the electronic device 10 includes at least one processor 11, and a memory, such as a Read-Only Memory (ROM) 12, a Random Access Memory (RAM) 13, etc., which is communicatively connected to the at least one processor 11, wherein the memory stores a computer program that can be executed by the at least one processor. The processor 11 can perform various appropriate actions and processes according to the computer program stored in the Read-Only Memory (ROM) 12 or loaded from the storage unit 18 into the Random Access Memory (RAM) 13. In the RAM 13, various programs and data required for the operation of the electronic device 10 can also be stored. The processor 11, the ROM 12, and the RAM 13 are connected to each other through a bus 14. An Input / Output (I / O) interface 15 is also connected to the bus 14.

[0168] A plurality of components in the electronic device 10 are connected to the I / O interface 15, including: an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, a loudspeaker, etc.; a storage unit 18, such as a magnetic disk, an optical disk, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices through a computer network, such as the Internet, and / or various telecommunication networks.

[0169] The processor 11 can be various general and / or special purpose processing components with processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a Central Processing Unit (CPU), a Graphics Processing Unit (GPU), various special-purpose Artificial Intelligence (AI) computing chips, various processors running machine learning model algorithms, a Digital Signal Processor (DSP), and any appropriate processor, controller, microcontroller, etc. The processor 11 performs various methods and processes described above, such as the capacity expansion method of a distribution transformer.

[0170] In some embodiments, the capacity expansion method of a distribution transformer can be implemented as a computer program tangibly embodied in a computer readable storage medium, such as the storage unit 18. In some embodiments, part or all of the computer program can be loaded and / or installed onto the electronic device 10 via the ROM 12 and / or the communication unit 19. When the computer program is loaded onto the RAM 13 and executed by the processor 11, one or more steps of the capacity expansion method of a distribution transformer described above can be performed. Alternatively, in other embodiments, the processor 11 can be configured to perform the capacity expansion method of a distribution transformer by other any appropriate means, such as by means of firmware.

[0171] The various embodiments of the systems and techniques described above can be implemented in digital electronic circuitry, integrated circuitry, a field programmable gate array (FPGA), an application specific integrated circuit (ASIC), a system on chip (SOC), a complex programmable logic device (CPLD), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include implementation in one or more computer programs that are executable and / or interpretable on a programmable system including at least one programmable processor, which can be special or general purpose, coupled to receive data and instructions from, and to transmit data and instructions to, a storage system, at least one input device, and at least one output device.

[0172] Computer programs implementing methods of the present application can be written in any combination of one or more programming languages. These computer programs can be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the computer program, when executed by the processor, implements the functions / acts specified in the flow diagrams and / or block diagrams. The computer program can be executed entirely on a machine, partially on a machine, partially on a machine as part of a standalone software package, partially on a machine and partially on a remote machine or entirely on a remote machine or server.

[0173] In the context of this application, a computer readable storage medium can be a tangible medium that can contain or store a computer program for use by or in connection with an instruction execution system, apparatus, or device. A computer readable storage medium can include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. Alternatively, a computer readable storage medium can be a machine readable signal medium. More specific examples of a machine readable storage medium will include one or more lines of a program of instructions in a transitory signal form, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0174] To provide for interaction with a user, the systems and techniques described here can be implemented on an electronic device having a display device (e.g., a Cathode Ray Tube (CRT) or a Liquid Crystal Display (LCD) monitor) for displaying information to the user and a keyboard and a pointing device (e.g., a mouse or a trackball) by which the user can provide input to the electronic device. Other kinds of devices can be used to provide for interaction with a user as well; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form, including acoustic, speech, or tactile input.

[0175] The systems and techniques described herein can be implemented in a computing system that includes a back end component, e.g., as a data server, or that includes a middleware component, e.g., an application server, or that includes a front end component, e.g., a user computer having a graphical user interface or a Web browser through which a user can interact with an implementation of the systems and techniques described herein, or any combination of such back end, middleware, or front end components. The components of the system can be interconnected by any form or medium of digital data communication, e.g., a communication network. Examples of communication networks include a local area network (LAN), a wide area network (WAN), the Internet, and a blockchain network.

[0176] The computing system can include clients and servers. A client and server are generally remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other. A server can be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system, and solves the defects of large management difficulty and weak business scalability in traditional physical host and virtual private server (VPS) services.

[0177] The embodiments of the present application also provide a computer program product, which comprises a computer program, and the computer program implements the capacity expansion method of the power distribution transformer as provided in any of the embodiments of the present application when executed by a processor.

[0178] The computer program product, in implementation, can be written in one or more programming languages or combinations thereof to implement the computer program codes for performing the operations of the present application, including object-oriented programming languages such as Java, Smalltalk, C++, and conventional procedural programming languages such as "C" language or similar programming languages. The program codes can be executed entirely on a user computer, partially on a user computer, as an independent software package, partially on a user computer and partially on a remote computer, or entirely on a remote computer or server. In the case involving a remote computer, the remote computer can be connected to the user computer through any kind of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computer (e.g., through the Internet by using an Internet service provider).

[0179] It should be understood that the various forms of flow shown above can be used to reorder, add, or delete steps. For example, the various steps described in this application can be performed in parallel, in series, or in a different order, as long as the desired results of the technical solutions of the present application can be achieved, which are not limited herein.

Claims

1. A method for expanding the capacity of a distribution transformer, comprising: Generate topology diagrams for multiple distribution transformers; In response to dispatching power between a first distribution transformer and a second distribution transformer, respectively collecting a power dispatch value output by the first distribution transformer and a power dispatch value output by the second distribution transformer; Converting the topology relationship diagram into a power scheduling relationship diagram according to the power scheduling value output by the first distribution transformer and the power scheduling value output by the second distribution transformer; Calculating the power dispatch levels of the plurality of distribution transformers respectively according to the power dispatch relationship diagram; When the plurality of distribution transformers are in operation, respectively collecting operation data of the plurality of distribution transformers; Converting the power scheduling relationship graph into a graph neural network based on the operating data; identifying the operating status of the plurality of distribution transformers respectively according to the graph neural network; Selecting some distribution transformers from the plurality of distribution transformers according to the operating status and the power dispatch level, and expanding the capacity of the selected distribution transformers.

2. The method according to claim 1, wherein The converting the topology relationship diagram into a power scheduling relationship diagram according to the power scheduling value output by the first distribution transformer and the power scheduling value output by the second distribution transformer includes: Setting each first node in the topology relationship graph to a second node corresponding to the first node in the power scheduling relationship graph, and setting each first edge in the topology relationship graph to a second edge corresponding to the first edge in the power scheduling relationship graph; the first node and the second node corresponding to the first node represent a distribution transformer; The first distribution transformer and the second distribution transformer are used as the first dispatching transformer and the second dispatching transformer respectively; In the power scheduling relationship diagram, retain the second side that simultaneously connects the first scheduling transformer and the second scheduling transformer, and delete the second side that does not simultaneously connect the first scheduling transformer and the second scheduling transformer; An edge weight is configured for the second edge according to the power dispatch value of the first dispatching transformer and the power dispatch value of the second dispatching transformer.

3. The method according to claim 2, wherein: The configuring an edge weight for the second edge according to the power scheduling value of the first scheduling transformer and the power scheduling value of the second scheduling transformer includes: According to the power dispatch value of the first dispatching transformer and the power dispatch value of the second dispatching transformer The dispatch value determines a first dispatch value and a second dispatch value; the first dispatch value is a power dispatch value output by the first dispatch transformer to the second dispatch transformer, and the second dispatch value is a power dispatch value output by the second dispatch transformer to the first dispatch transformer; Configuring a first dispatching weight for the first dispatching value under the dimension of the first dispatching transformer, and configuring a second dispatching weight for the second dispatching value under the dimension of the second dispatching transformer; The product of the first scheduling value and the first scheduling weight is added to the product of the second scheduling value and the second scheduling weight to obtain the edge weight of the second edge.

4. The method according to claim 3, wherein: The configuring a first scheduling weight for the first scheduling value under the dimension of the first scheduling transformer, and configuring a second scheduling weight for the second scheduling value under the dimension of the second scheduling transformer, includes: Calculating the sum of the power dispatching values ​​output by the first dispatching transformer to all other distribution transformers to obtain a first total power value; Calculating a ratio between the first scheduling value and the first total power value to obtain a first scheduling weight; Calculating the sum of the power dispatching values ​​output by the second dispatching transformer to all other distribution transformers to obtain a second total power value; A ratio between the second scheduling value and the second total power value is calculated to obtain a second scheduling weight.

5. The method according to claim 1, wherein Calculating the power dispatch levels of the plurality of distribution transformers respectively according to the power dispatch relationship diagram includes: For each distribution transformer, using page ranking to calculate the fluctuation degree of power scheduling of each distribution transformer in the power scheduling relationship diagram; Querying a fluctuation range into which the fluctuation degree falls in a plurality of preset fluctuation ranges to obtain a target range; wherein the plurality of preset fluctuation ranges are each configured with a power scheduling level; The power dispatch level configured for the target range is assigned as the power dispatch level of the distribution transformer.

6. The method according to any one of claims 1 to 5, wherein The operating states include the first state, the second state and the third state in order from normal to abnormal; the power scheduling levels include the first level, the second level and the third level in order from small to large; The selecting some distribution transformers from the plurality of distribution transformers according to the operating state and the power dispatch level, and expanding the capacity of the selected distribution transformers, includes: For each distribution transformer, in response to the operating state of the distribution transformer being the third state and the power dispatch level of the distribution transformer being the second level or the third level, the distribution transformer is determined to be a candidate transformer; or, in response to the operating state of the distribution transformer being the second state and the distribution transformer being the third level, the distribution transformer is determined to be a candidate transformer. If the power dispatch level of the power transformer is the third level, the distribution transformer is determined to be a candidate transformer; Calculating the distances between all candidate transformers in the power scheduling relationship diagram; For each distance, in response to the distance being less than or equal to a preset threshold, deleting one of the two candidate transformers corresponding to the distance; In response to the remaining distances being respectively greater than preset thresholds, it is determined to expand the capacity of the candidate transformers corresponding to the remaining distances.

7. The method according to claim 6, wherein: Calculating the distances between all candidate transformers in the power scheduling relationship diagram includes: In the power scheduling relationship diagram, two candidate transformers are selected from the candidate transformers, and the shortest paths are planned for the two candidate transformers; If there are other candidate transformers on the path, delete the two candidate transformers; If there are no other candidate transformers on the path, the number of second edges in the path is used as the distance between the two candidate transformers; The deleting one of the two candidate transformers corresponding to the distance includes: For each candidate transformer, calculating the average of all distances between the candidate transformer and other candidate transformers; The candidate transformer with the largest average value among the two candidate transformers corresponding to the distance less than or equal to the preset threshold is deleted.

8. A capacity expansion device for a distribution transformer, comprising: a topology diagram generating module configured to generate a topology diagram for a plurality of distribution transformers; a power dispatch value acquisition module, configured to respectively acquire a power dispatch value output by the first distribution transformer and a power dispatch value output by the second distribution transformer in response to dispatching power between the first distribution transformer and the second distribution transformer; a power scheduling relationship diagram conversion module, configured to convert the topology relationship diagram into a power scheduling relationship diagram according to the power scheduling value output by the first distribution transformer and the power scheduling value output by the second distribution transformer; a power dispatch level calculation module, configured to respectively calculate the power dispatch levels of the plurality of distribution transformers according to the power dispatch relationship diagram; an operation data acquisition module, configured to respectively collect operation data of the plurality of distribution transformers when the plurality of distribution transformers are in operation; A graph neural network conversion module is configured to convert the power scheduling relationship graph into Switch to graph neural network; an operating status identification module, configured to respectively identify the operating status of the plurality of distribution transformers based on the graph neural network; The comprehensive capacity expansion module is configured to select some distribution transformers from the plurality of distribution transformers according to the operating status and the power scheduling level, and expand the capacity of the selected distribution transformers.

9. An electronic device, comprising: at least one processor; as well as a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can perform the method for expanding the capacity of a distribution transformer according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, and the computer program is used to enable a processor to implement the capacity expansion method of the distribution transformer according to any one of claims 1 to 7 when executed.

Images