A method, device, equipment and medium for identifying power grid topology

Abstract

The present invention belongs to the field of new energy and energy-saving technology, and specifically relates to a method, device, equipment and medium for identifying power grid topology, including: analyzing power plant data to obtain power grid hierarchical partitioning; deploying smart terminals on the edge side based on power grid clustering hierarchical partitioning, pre-processing equipment data, and obtaining power grid nodes; utilizing power grid state estimation and positioning, combined with power grid node topology characteristics, to realize the identification of network topology information based on the Bayesian network model. The present invention realizes power grid hierarchical partitioning based on clustering analysis method, selects cloud-edge collaborative architecture for edge data processing and cloud data verification, and solves the problem of large-scale and high-frequency data communication transmission congestion of distributed nodes in virtual power plants; combines network topology characteristics and actual on-site conditions to give node topology prior knowledge, realizes the identification of network topology information based on the Bayesian network model, and solves the problem that distributed resources are connected to the grid to change the physical topology structure of the system and adjust the power grid current transmission path, which makes it difficult to identify topology information.

Description

Technical Field

[0001] The present invention belongs to the field of new energy and energy-saving technology, and specifically relates to a power grid topology identification method, device, equipment and medium. Background Art

[0002] With the large-scale access of renewable energy, the widespread use of interactive energy-consuming devices such as electric vehicles, distributed energy, and energy storage, the diversity of resource characteristics of networked devices and the influence of spatiotemporal factors on the accuracy of load regulation are gradually increasing. At the same time, multi-energy complementary power generation is gradually replacing traditional single energy power generation, and power generation is gradually transitioning from centralized to distributed, further exacerbating the instability of the power grid. In this case, the form and planning methods of traditional power grids are no longer applicable, and a new operation and management model and form that conforms to the current trend of complex development of distribution networks is urgently needed. The current virtual power plant resource access, transmission, and verification have the following problems:

[0003] First, grid zoning and stratification are important means of achieving production management and dispatch automation. Currently, commonly used multi-level management methods generally divide regions based on substation locations and administrative regions, and divide them into hierarchical levels according to high, medium, and low voltage structures. However, with the emergence of virtual power plants and the development of modern communication technologies, intelligent terminals have been deployed in large numbers. Centralized control alone is difficult to ensure control effectiveness and accuracy, while distributed control alone hinders the realization of energy complementarity among multiple regions. Second, due to the wide distribution of resource nodes and the high frequency of data transmission, coupled with the diverse and streaming nature of the massive amount of collected data, node information transmission is subject to congestion and verification issues. Based on the idea of ​​cloud-edge collaboration, this paper adopts a technical approach of deeply mining the value of energy Internet big data, streamlining and refining data to improve data value density, and sharing and verifying data in real time between the cloud and the edge. This solves the problems of node information congestion and information verification difficulties that are common in distributed nodes of virtual power plants. Third, the accuracy of grid state estimation and positioning is of great significance to power system analysis and is the foundation of other advanced applications, directly affecting the rationality of planning and operation and the reliability of control and protection. The current mainstream methods for power grid state estimation and positioning are the breadth-first search algorithm and the adjacency matrix method. However, these two traditional algorithms have disadvantages such as difficult expansion of association tables and large computational complexity, and can no longer meet the topology identification needs of today's complex power grid structures. Summary of the Invention

[0004] In order to overcome the shortcomings of the prior art, the present invention aims to provide a method for solving the problem of topology identification of complex power grid structures.

[0005] In order to achieve the above object, the present invention adopts the following technical solutions:

[0006] In a first aspect, the present invention provides a method for identifying a power grid topology, comprising:

[0007] Analyze the power plant data to obtain the grid stratification and partitioning;

[0008] Based on the grid's hierarchical partitioning, smart terminals are deployed at the edge to pre-process device data and obtain grid nodes.

[0009] By utilizing grid state estimation and positioning, combined with grid node topology characteristics, network topology information is identified based on the Bayesian network model.

[0010] Furthermore, the analysis based on the power plant data to obtain the grid stratification and partitioning includes: using the power plant data to calculate the minimum net load value, the calculation formula is: Among them, N d N is the target number of distribution layer power grid construction; c The number of statistical areas for load forecasting and power balance; P Load,max,j is the maximum load of area j; C DG,avr,j is the average distributed generation output of region j based on the distributed generation processing model and the multi-energy complementary generation model; C ES,max,k is the maximum energy storage output of region j.

[0011] Furthermore, the grid clustering and hierarchical partitioning includes: Among them, P avr is the net load value; T is the statistical duration; P L，t is the load at time t; C DG，t is the output of the power generation equipment at time t; C ES，t is the output of the electrical equipment at time t; L i is the geographical location parameter of the i-node device; L j is the geographical location parameter of the i-node device; U i,t is the voltage of node i at time t; P i，t are the output power of node i at time t; U j，t is the voltage of node j at time t; P j，t are the output power of node j at time t; σ1, σ2, and σ3 are the maximum allowable values ​​of the set targets.

[0012] Furthermore, based on the grid clustering, hierarchical partitioning, smart terminals are deployed on the edge side to pre-process the device data. The grid nodes include:

[0013] Edge-side smart terminals are deployed in the grid's clustered, hierarchical, and partitioned regions. The first step is to pre-process the sensory data of each device in different layers to generate knowledge data. The sensory data includes active power, reactive power, device parameters, historical load, and ambient temperature.

[0014] The knowledge data is uploaded to the cloud for the second step of preprocessing and analysis.

[0015] Furthermore, the first step of preprocessing includes: combining the knowledgeization of the sensory data with edge computing, extracting knowledge from the original data on the edge side, and converting it into knowledge data;

[0016] The second step of preprocessing includes: simplifying the knowledge data to improve the data value density; and uniformly calculating and measuring the knowledge data, extracting erroneous data and data exceeding the preset error range, and returning the remaining data to the cloud.

[0017] Furthermore, the knowledge data is uniformly calculated and measured, including: edge power probability balance and cloud power probability balance;

[0018] The edge power probability balance is: Among them, C TtF is the output dispatched from the transmission grid at the edge end; N c is the number of cloud power grids to which the edge belongs; C FtC，k is the output of the k-th cloud power grid dispatched from the dispatching end; σ is the preset error;

[0019] The cloud power probability balance is: Among them, C FtC The output dispatched by the cloud from the dispatching end; C ES，C Contribute to energy storage in the cloud; N D is the number of edge power grids under the cloud power grid; a l is the balance ratio of the l-th edge power grid; P Load，max，l is the maximum load of the lth edge power grid; D is the reserve coefficient.

[0020] Furthermore, the method of utilizing grid state estimation and positioning, combined with grid node topology characteristics, to identify network topology information based on a Bayesian network model specifically includes:

[0021] By using grid state estimation and positioning, combined with grid node topology characteristics, the prior network topology information is defined, where the network topology information is: Among them, a ij represents the possibility of connecting nodes i and j through lines in the distribution network; a ij =1 indicates that node i and node j are connected by a line; ij =0 indicates that there is no line connection between node i and node j; a ij =0.5 means that there is no definite prior relationship between the connection between node i and node j;

[0022] The feature data is extracted using the prior network topology structure, and the link relationship between network topology nodes is solved based on the Bayesian network model to realize the recognition of network topology information.

[0023] In a second aspect, the present invention provides a power grid topology identification device, comprising:

[0024] Hierarchical partitioning module: used to analyze power plant data and obtain grid hierarchical partitioning;

[0025] Data processing module: used to deploy smart terminals on the edge side based on grid clustering, hierarchical partitioning, pre-process device data, and obtain grid nodes;

[0026] Topology identification module: It is used to utilize grid state estimation and positioning, combined with grid node topology characteristics, to identify network topology information based on the Bayesian network model.

[0027] In a third aspect, the present invention provides an electronic device comprising a processor and a memory, wherein the processor is configured to execute a computer program stored in the memory to implement any one of the power grid topology identification methods.

[0028] In a fourth aspect, the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores at least one instruction, and when the at least one instruction is executed by a processor, any one of the power grid topology identification methods is implemented.

[0029] The present invention has at least the following beneficial effects:

[0030] The present invention provides a power grid topology identification method, device, equipment and medium, including: performing grid layering and partitioning according to power plant data, calculating the minimum net load value, and obtaining grid clustering and layering and partitioning; wherein the power plant equipment data includes: equipment type, equipment voltage, output, net load density, geographical location and power plant area; deploying intelligent terminals on the edge side of the power grid clustering and layering and partitioning, pre-processing the equipment data, and obtaining power grid nodes; utilizing grid state estimation and positioning, combined with the topological characteristics of the power grid nodes, and realizing the identification of network topology information based on the Bayesian network model. The present invention realizes grid layering and partitioning based on the clustering analysis method, selects the cloud-edge collaborative architecture for edge data processing and cloud data verification, and solves the problem of large-scale and high-frequency data communication transmission congestion of distributed nodes in virtual power plants; combines the network topology characteristics and the actual situation on site to give node topology prior knowledge, considers the node voltage and power characteristic quantities, and realizes the identification of network topology information based on the Bayesian network model, solving the problem that the physical topology structure of the system is changed by the grid connection of distributed resources, and the power grid current transmission path is adjusted to make the identification of topology information difficult. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] The accompanying drawings, which constitute part of the present invention, are intended to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are intended to explain the present invention and do not constitute an undue limitation of the present invention. In the accompanying drawings:

[0032] Figure 1 This is a schematic diagram of the IEEE33 node power distribution system;

[0033] Figure 2 A schematic diagram of power grid topology partitioning for a power grid topology identification method provided by the present invention;

[0034] Figure 3 A schematic diagram of node partitioning for a method for identifying power grid topology provided by the present invention;

[0035] Figure 4 This is a structural block diagram of an electronic device of the present invention. DETAILED DESCRIPTION

[0036] The present invention will be described in detail below with reference to the accompanying drawings and in combination with embodiments. It should be noted that, in the absence of conflict, the embodiments and features of the embodiments of the present invention can be combined with each other.

[0037] The following detailed description is an exemplary description and is intended to provide further detailed description of the present invention. Unless otherwise indicated, all technical terms used in the present invention have the same meaning as those generally understood by those skilled in the art to which the present invention belongs. The terms used in the present invention are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present invention.

[0038] Example 1

[0039] See also Figures 1 to 3 As shown, the present invention provides a method for identifying power grid topology, comprising:

[0040] S1: Analyze power plant data to obtain grid stratification and zoning. Power plant equipment data includes: equipment type, equipment voltage, output, net load density, geographic location, and power plant area.

[0041] S2: Deploy smart terminals at the edge based on grid clustering, hierarchical partitioning, and pre-process device data to obtain grid nodes.

[0042] S3: Utilize grid state estimation and positioning, combined with grid node topology characteristics, and realize network topology information identification based on the Bayesian network model.

[0043] The step S1 specifically includes: performing a grid topology analysis on the distribution system according to the power plant data, calculating the minimum net load value, and partitioning the grid topology nodes based on the clustering method using characteristic quantities to obtain grid hierarchical partitions.

[0044] The grid topology analysis includes: this embodiment takes the IEEE33-node distribution system as an example, including 32 branches with a voltage level of 10kV; 5 tie switch branches (I25-29, I33-18, I9-15, I21-8, I22-12); node 1 is a substation node; nodes 2, 3, 4, 5, 19, 20, 21, and 22 are equipped with photovoltaic users; nodes 6, 7, 8, 9, and 10 are large industrial users; nodes 22, 23, and 24 are centralized charging stations; nodes 26, 27, 28, 29, and 30 are commercial loads equipped with flexible loads such as central air conditioning; nodes 11, 12, 13, 14, 31, 32, and 33 are residential loads, and the heating and cooling networks use electric heating and cooling methods; nodes 15, 16, 17, and 18 are smart buildings, including flexible loads such as smart switches and lighting.

[0045] The method of using characteristic quantities to partition the power grid topology nodes based on the clustering method specifically includes: determining the distribution network planning range and the range covered by the edge, clarifying all power plant equipment data in the area, wherein the power plant equipment data includes: equipment type, equipment voltage, output, net load density, geographical location and power plant area, and calculating the minimum net load value applied to each edge area;

[0046] The minimum net load value is: Among them, N d N is the target number of distribution layer power grid construction; c The number of statistical areas for load forecasting and power balance; P Load,max,j is the maximum load of area j; C DG,avr,j is the average distributed generation output of region j based on the distributed generation processing model and the multi-energy complementary generation model; C ES,max,k is the maximum output of energy storage in area j. avr The average is minimized, the equipment type is unified, the geographical location L, voltage U, and output P are close as the goal, and the grid hierarchical partitioning is obtained using the clustering method.

[0047] The grid is divided into layers and zones: Among them, P avr is the net load value; T is the statistical duration; P L，t is the load at time t; C DG，t is the output of the power generation equipment at time t; C ES，t is the output of the electrical equipment at time t; L i is the geographical location parameter of the i-node device; L j is the geographical location parameter of the i-node device; U i，t is the voltage of node i at time t; P i，t are the output power of node i at time t; Uj，t is the voltage of node j at time t; P j，t are the output power of node j at time t; σ1, σ2, and σ3 are the maximum allowable values ​​of the set targets.

[0048] The obtained partition topology is as follows: Zone I: nodes 1, 2, 3, 4, 5, 19, 20, 21, 22, including substations and user photovoltaics; Zone II: nodes 6, 7, 8, 9, 10, 26, 27, 28, 29, 30, including large industrial users and commercial loads; Zone III: nodes 11, 12, 13, 14, 31, 32, 33, including residential loads; Zone IV: nodes 15, 16, 17, 18, including smart buildings; Zone V: nodes 23, 24, 25, including centralized charging stations.

[0049] In the S2 step, the deployment of smart terminals on the edge side is specifically as follows: deploying edge-side smart terminals in areas I, II, III, IV, and V respectively, performing the first step of preprocessing on the perception data of each device body to obtain knowledge data, and uploading the corresponding knowledge data to the cloud for the second step of preprocessing and analysis.

[0050] Among them, the sensed quantity data includes: active power, reactive power, equipment parameters, historical load and ambient temperature;

[0051] The first step of preprocessing involves combining the knowledge-based perception data with edge computing, extracting knowledge from the raw data on the edge, and converting it into knowledge data.

[0052] The second step of preprocessing includes: simplifying the knowledge data to improve the data value density; and uniformly calculating and measuring the knowledge data, extracting erroneous data and data that exceeds the preset error range, and returning the rest of the data to the cloud.

[0053] The maximum transmission capacity index A of the line based on a single power source-load node ij , characterizes the role of a transmission line in transmitting power between each power source-load node pair from the perspective of a single power source-load node pair. The power injected and absorbed by a single power source-load node pair is: A ij∈(g，l) =min h∈H P max，h , where when any transmission line h in the power grid (assuming the set of all lines is H) first reaches its maximum transmission capacity P max,h At this time, the maximum transmission capacity of the line is A ij∈(g，l) ;

[0054] Among them, S g is the power node set; S l is the set of load nodes;

[0055] Node transmission contribution A i for: Among them, A i represents the average contribution of all routes connected to node i to the power transmission capacity of node i; D i is the degree value of node i, reflecting the physical topological relationship between node i and other nodes in the power grid; Ω is the set of all nodes directly connected to node i; ∑ j∈Ω A ij Characterizes the transmission and carrying capacity of node i for power flow in the entire power grid;

[0056] The second step of preprocessing specifically includes: edge power probability balancing and cloud power probability balancing;

[0057] The edge power probability balance is: Among them, C TtF is the output dispatched from the transmission grid at the edge end; N c is the number of cloud power grids to which the edge belongs; C FtC,k is the output of the k-th cloud power grid dispatched from the dispatching end; σ is the preset error;

[0058] The cloud power probability balance is: Among them, C FtC The output dispatched by the cloud from the dispatching end; C ES，C Contribute to energy storage in the cloud; N D is the number of edge power grids under the cloud power grid; a l is the balance ratio of the l-th edge power grid; P Load,max,l is the maximum load of the lth edge power grid; D is the reserve coefficient, which is generally taken as 0.2-0.3 considering hierarchical scheduling and local consumption of renewable energy.

[0059] The S3 step includes:

[0060] S31: Utilizing grid state estimation and positioning, combined with grid node topology characteristics, define prior network topology information, where the network topology information is: Among them, a ij represents the possibility of connecting nodes i and j through lines in the distribution network; a ii =1 indicates that node i and node j are connected by a line; ij =0 indicates that there is no line connection between node i and node j; a ij= 0.5 means that there is no a priori relationship between the connection relationship between node i and node j; the diagonal elements in the PDNSI matrix are always 1, because in the actual distribution network, the nodes at the diagonal position are unified nodes. Based on the preset information, the PDNSI can be obtained to initialize the distribution network topology;

[0061] S32: Considering nodes with close electrical distances, the more similar the change characteristics of their voltage curves are; the farther the electrical distance is, the less similar the change characteristics of their voltage curves are; the closer the node is to the head end of the feeder, the higher the voltage; the closer the node is to the end of the feeder, the lower the voltage. Select characteristic data;

[0062] S33: Utilize the prior network topology structure to extract feature data, solve the link relationship between network topology nodes based on the Bayesian network model, and realize the recognition of network topology information.

[0063] In the step S31, PDNSI can be obtained according to the preset information to initialize the distribution network topology, which specifically includes: initializing the distribution network topology structure according to the preset information, determining the number of characteristic data groups T and the termination number λ, and calculating the initial measure e0.

[0064] In the step S32, the selection of characteristic data specifically includes: detecting the autovariance of each measurement at time n and the covariance with other measurements, setting a threshold value for each measurement, and if one of the autovariance or covariance at time n exceeds the threshold value, then eliminating the data at time n, obtaining a new data set, and then obtaining the characteristic data;

[0065] In the step S33, solving the link relationship between network topology nodes based on the Bayesian network model specifically includes:

[0066] S331: Modify the structure (such as adding a line, deleting a line), and ensure that the new structure does not contain directed loops. All new structures generated are represented by set B. s Represents. From set B s A modified new structure e is randomly selected in the equation, and the difference between the structure and the original structure is expressed as Δe. The probability p = exp(-Δe / T) is obtained.

[0067] S332: If the calculated p value is greater than or equal to 1, then select structure e as the new initial topology; otherwise, adopt structure e as the new initial structure with probability p;

[0068] S333: Repeat steps S331 and S332 a times. If the topology is not modified during this period, the algorithm is stopped. The connection status of this topology can represent the connection status of physical devices in the actual distribution network. Otherwise, add 1 to i. If i is greater than λ, the algorithm is stopped. The device connection status of this topology is closest to the device connection status in the actual power grid. If i is less than λ, the T value is reduced. The reduction formula is T = β × T, where the value of β is between 0 and 1. After obtaining the new T value, return to S331 and continue executing the algorithm;

[0069] S334: Model and analyze topology positioning to achieve network topology identification and positioning.

[0070] Step S334 specifically includes modeling the IEEE 33-node system. During modeling, the topological characteristics of the distribution network and actual field conditions are considered. Given the probability of inter-node connections and incorrect node-branch connections, the proposed method is applied to perform topological analysis of the network structure. The results show that the probability of connectivity between nodes 22 and 35 is 0.15, and the probability of connectivity between nodes 10 and 11 is 0.2. The topological location recognition model can identify and locate the network topology. The following table shows the connectivity probability of each node:

[0071]

[0072] Example 2

[0073] The present invention provides a power grid topology identification device, comprising:

[0074] Hierarchical partitioning module: used to analyze power plant data and obtain grid hierarchical partitioning;

[0075] Data processing module: used to deploy smart terminals on the edge side based on grid clustering, hierarchical partitioning, pre-process device data, and obtain grid nodes;

[0076] Topology identification module: It is used to utilize grid state estimation and positioning, combined with grid node topology characteristics, to identify network topology information based on the Bayesian network model.

[0077] Example 3

[0078] See also Figure 4 As shown, the present invention also provides an electronic device 100 for a power grid topology identification method; the electronic device 100 includes a memory 101, at least one processor 102, a computer program 103 stored in the memory 101 and executable on the at least one processor 102, and at least one communication bus 104.

[0079] The memory 101 can be used to store the computer program 103. The processor 102 implements the steps of the power grid topology identification method described in Example 1 by running or executing the computer program stored in the memory 101 and calling the data stored in the memory 101. The memory 101 can mainly include a program storage area and a data storage area. The program storage area can store an operating system and at least one application required for a function (such as a sound playback function, an image playback function, etc.); the data storage area can store data (such as audio data) created based on the use of the electronic device 100. In addition, the memory 101 can include non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (SmartMedia Card, SMC), a secure digital (Secure Digital, SD) card, a flash card, at least one disk storage device, a flash memory device, or other non-volatile solid-state storage device.

[0080] The at least one processor 102 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The processor 102 may be a microprocessor or any conventional processor, etc. The processor 102 is the control center of the electronic device 100 and connects various parts of the entire electronic device 100 using various interfaces and lines.

[0081] The memory 101 in the electronic device 100 stores a plurality of instructions to implement a power grid topology identification method. The processor 102 may execute the plurality of instructions to implement:

[0082] Analyze the power plant data to obtain the grid stratification and partitioning;

[0083] Based on the grid's hierarchical partitioning, smart terminals are deployed at the edge to pre-process device data and obtain grid nodes.

[0084] By utilizing grid state estimation and positioning, combined with grid node topology characteristics, network topology information is identified based on the Bayesian network model.

[0085] Example 4

[0086] If the module / unit integrated in the electronic device 100 is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present invention implements all or part of the process in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned method embodiments when executed by the processor. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form, etc. The computer-readable medium may include: any entity or device that can carry the computer program code, recording medium, U disk, mobile hard disk, magnetic disk, optical disk, computer memory and read-only memory (ROM, Read-Only Memory).

[0087] It will be understood by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Thus, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0088] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts and / or block diagrams. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0089] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0090] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0091] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered by the scope of protection of the claims of the present invention.

Claims

1. A method for identifying power grid topology, characterized in that: include: Analyze the power plant data to obtain the grid stratification and partitioning; Edge-side smart terminals are deployed in the clustered, hierarchical, and partitioned areas of the power grid. The first step of preprocessing is to obtain knowledge data from the sensory data of each device in different layers. The sensory data includes active power, reactive power, equipment parameters, historical load, and ambient temperature. The knowledge data is uploaded to the cloud for the second step of preprocessing and analysis. The first step of preprocessing includes: combining the knowledgeization of sensory data with edge computing, extracting knowledge from the original data on the edge, and converting it into knowledge data. The second step of preprocessing includes: simplifying the knowledge data to improve the data value density. The knowledge data is then uniformly calculated and measured, and erroneous data and data exceeding the preset error range are extracted, and the rest of the data is returned to the cloud. The knowledge data is uniformly calculated and measured, including: edge-side power probability balance and cloud-side power probability balance. The edge-side power probability balance is: Among them, C TtF is the output dispatched from the transmission grid at the edge; N c is the number of cloud power grids to which the edge belongs; C FtC,k is the output of the k-th cloud power grid dispatched from the dispatching end; σ is the preset error; the cloud power probability balance is: Among them, C FtC The output dispatched by the cloud from the dispatching end; C ES,C Contribute to energy storage in the cloud; N D is the number of edge power grids under the cloud power grid; a l is the balance ratio of the l-th edge power grid; P Load,max,l is the maximum load of the l-th edge power grid; D is the reserve coefficient; By utilizing grid state estimation and positioning, combined with grid node topology characteristics, network topology information is identified based on the Bayesian network model.

2. A power grid topology identification method according to claim 1, characterized in that: The analysis based on the power plant data to obtain the grid stratification and partitioning includes: using the power plant data to calculate the minimum net load value, the calculation formula is: Among them, N d N is the target number of distribution layer power grid construction; c The number of statistical areas for load forecasting and power balance; P Load,max,j is the maximum load of area j; C DG,avr,j is the average distributed generation output of region j based on the distributed generation processing model and the multi-energy complementary generation model; C ES,max,k is the maximum energy storage output of region j.

3. A power grid topology identification method according to claim 1, characterized in that: The grid hierarchical partitioning includes: Among them, P avr is the net load value; T is the statistical duration; P L,t is the load at time t; C DG,t is the output of the power generation equipment at time t; C ES,t is the output of the electrical equipment at time t; L i is the geographical location parameter of the i-node device; L j is the geographical location parameter of the i-node device; U i,t is the voltage of node i at time t; P i,t are the output power of node i at time t; U j,t is the voltage of node j at time t; P j,t are the output power of node j at time t; σ1, σ2, and σ3 are the maximum allowable values ​​of the set targets.

4. A method for identifying power grid topology according to claim 1, characterized in that: The method utilizes grid state estimation and positioning, combines grid node topology characteristics, and implements network topology information identification based on a Bayesian network model, specifically including: By using grid state estimation and positioning, combined with grid node topology characteristics, the prior network topology information is defined, where the network topology information is: Among them, a ij represents the possibility of connecting nodes i and j through lines in the distribution network; a ij =1 indicates that node i and node j are connected by a line; ij =0 indicates that there is no line connection between node i and node j; ij =0.5 means that there is no definite prior relationship between the connection between node i and node j; The feature data is extracted using the prior network topology structure, and the link relationship between network topology nodes is solved based on the Bayesian network model to realize the recognition of network topology information.

5. A power grid topology identification device, used to implement the power grid topology identification method according to claim 1, characterized in that: include: Hierarchical partitioning module: used to analyze power plant data and obtain grid hierarchical partitioning; Data processing module: used to deploy smart terminals on the edge side based on grid clustering, hierarchical partitioning, pre-process device data, and obtain grid nodes; Topology identification module: It is used to utilize grid state estimation and positioning, combined with grid node topology characteristics, to identify network topology information based on the Bayesian network model.

6. An electronic device, characterized in that: The method comprises a processor and a memory, wherein the processor is configured to execute a computer program stored in the memory to implement the power grid topology identification method according to any one of claims 1 to 4.

7. A computer-readable storage medium, characterized in that The computer-readable storage medium stores at least one instruction, and when the at least one instruction is executed by a processor, the power grid topology identification method according to any one of claims 1 to 4 is implemented.

Images