Regional polymorphic resource virtual aggregation method for power grid real-time scheduling demand

Abstract

The invention discloses a regional polymorphic resource virtual aggregation method for a power grid real-time scheduling demand, and belongs to the technical field of resource scheduling, and the method comprises the following steps: constructing a power grid model, clustering regional polymorphic resources based on the operation data of the regional polymorphic resources in the power grid model, and obtaining a final clustering result based on a target function; dividing the power grid model based on the topological characteristics of the power grid model to obtain a graph model, and aggregating the regional polymorphic resources based on the graph model to obtain a virtual unit; and performing power grid dispatching by using the virtual unit and the final clustering result according to the actual dispatching requirement of the power grid. Characteristic differences of different polymorphic resources in the actual scheduling process are considered, and the collaborative scheduling efficiency of the regional polymorphic resources is remarkably improved.

Description

Technical Field

[0001] The present invention relates to the technical field of resource scheduling, and in particular to a method for virtual aggregation of regional multi-modal resources oriented to real-time scheduling requirements of power grids. Background Art

[0002] With the development of power systems and energy transformation, the scale and variety of multi-modal resources are constantly increasing. To better utilize distributed resources, existing technologies cluster distributed resources and then classify similar multi-modal resources into one category to better perform resource scheduling and optimization. For example, patent publication number CN119298071A describes a method for aggregated scheduling of adjustable resources in a distribution network. The method includes: obtaining various types of demand-side adjustable resources and classifying them based on preset classification rules; extracting resource characteristics of each category of adjustable resources and establishing an adjustable resource model based on the resource characteristics; aggregating all categories of adjustable resource models using a clustering algorithm to obtain a resource aggregate; constructing a multi-objective optimization model for the resource aggregate and determining the objective function and constraints of the multi-objective optimization model; and solving the multi-objective optimization model using a Pareto solution method to obtain an optimized scheduling solution. However, the multi-objective optimization model constructed using the resource aggregate does not take into account the characteristic differences exhibited by different multi-modal resources in the actual scheduling process, resulting in low efficiency in the coordinated scheduling of regional multi-modal resources. Summary of the Invention

[0003] In order to address the technical problem that the existing technology does not take into account the differences in characteristics presented by different polymorphic resources in the actual scheduling process, resulting in low efficiency in the coordinated scheduling of regional polymorphic resources, the present invention provides a regional polymorphic resource virtual aggregation method for real-time scheduling needs of the power grid. The final clustering result is obtained through the objective function, and then the regional polymorphic resources are aggregated to obtain virtual units through a graph model. When scheduling is required, the virtual units are directly located, and then the scheduling tasks are allocated to the regional polymorphic resources in the virtual units according to the characteristics of the regional polymorphic resources in the final clustering results. This takes into account the differences in characteristics presented by different polymorphic resources in the actual scheduling process, and significantly improves the efficiency of the coordinated scheduling of regional polymorphic resources.

[0004] To solve the above technical problems, the present invention provides a method for virtual aggregation of regional multi-modal resources oriented to the real-time dispatching needs of power grids, comprising the following steps: Build a power grid model, cluster regional multi-modal resources based on their operating data, and obtain the final clustering results based on the objective function; Based on the topological characteristics of the power grid model, the power grid model is divided to obtain a graph model, and based on the graph model, the regional multi-modal resources are aggregated to obtain virtual units; According to the actual dispatching requirements of the power grid, virtual units and the final clustering results are used to dispatch the power grid.

[0005] After adopting the above technical solution, the present invention has the following advantages: By clustering regional polymorphic resources with the same characteristics and obtaining the final clustering result based on an objective function that reflects the rationality of the final clustering result, the rationality of the final clustering result is improved. Regional polymorphic resources are aggregated to obtain virtual units through a graph model. When scheduling is required, they are directly located in the virtual units. Then, scheduling tasks are assigned to the regional polymorphic resources in the virtual units based on the characteristics of the regional polymorphic resources in the final clustering result. This takes into account the differences in characteristics of different polymorphic resources in the actual scheduling process, significantly improving the efficiency of collaborative scheduling of regional polymorphic resources. The grid model is divided into graph models based on its topological characteristics, and regional polymorphic resources are aggregated through the graph model to obtain virtual units, thus avoiding the aggregation of distant regional polymorphic resources and nodes, thereby improving the response speed of scheduling.

[0006] Preferably, clustering the regional polymorphic resources based on the operating data of the regional polymorphic resources in the power grid model and obtaining the final clustering result based on the objective function includes: Based on the operating data, clustering the regional polymorphic resources using spectral clustering to obtain initial clustering results; Obtain the center point of each cluster in the initial clustering result, input the center point into the objective function to match the initial clustering result with the requirements. If the requirements match, the initial clustering result is the final clustering result. If the requirements match fails, the initial clustering result is modified based on the constraint conditions to obtain the final clustering result.

[0007] Preferably, clustering the regional polymorphic resources using spectral clustering based on the operating data to obtain an initial clustering result includes: Calculate the similarity of the running data at different time sections in the running data, and obtain the graph Laplacian matrix based on the similarity; The eigenvalue inflection points of the graph Laplacian matrix are obtained, the total number of clusters is obtained based on the eigenvalue inflection points, and the regional polymorphic resources are clustered according to the total number of clusters to obtain the initial clustering results.

[0008] Preferably, the objective function is: Where N y represents the number of all scenarios corresponding to the operating data of different time sections, Y represents the average energy loss value corresponding to all scenarios, [L j] represents the cluster to which the multi-modal resources in each time period of the j-th scene belong in all scenes, num represents the number of time periods in the j-th scene, A represents the total number of clusters, P loss，α Indicates the energy loss value corresponding to the center point of the αth cluster in the initial clustering result.

[0009] Preferably, the constraints include node active power balance constraints, node reactive power balance constraints, distributed power supply installation quantity constraints and regional multi-modal resource group number constraints.

[0010] Preferably, the step of dividing the power grid model based on the topological features of the power grid model to obtain a graph model includes: The power grid model is divided into an ultra-high voltage main network, a high voltage main network, and a high voltage distribution network according to the topological characteristics, and the busbar node in the high voltage main network is used as the first vertex and the branch in the high voltage main network is used as the first edge to construct a first graph model; The busbar node in the high-voltage distribution network is taken as the second vertex, and the branch in the high-voltage distribution network is taken as the second edge to construct the second graph model; the ultra-high voltage main network is taken as the third vertex, and the third vertex, the first graph model and the second graph model are associated through the association equipment between the ultra-high voltage main network, the high voltage main network and the high-voltage distribution network to obtain the graph model.

[0011] Preferably, the step of aggregating regional polymorphic resources based on a graph model to obtain virtual units includes: The node where the regional polymorphic resource is located is used as the starting node, and the graph model is traversed in a breadth-first manner. The aggregation node is obtained according to the preset rules, and the regional polymorphic resource is aggregated at the aggregation node to obtain a virtual unit.

[0012] Preferably, the actual grid dispatching demand includes peak load regulation demand and grid flexible regulation demand.

[0013] Preferably, the method of using virtual units and final clustering results to perform grid dispatch according to actual grid dispatch requirements includes: When the actual dispatching demand of the power grid is a peak-shaving demand, the total amount of backup energy storage of the virtual unit is obtained, and the virtual unit is compared with the final clustering result to obtain the adjustment characteristics of the virtual unit, and the power grid is dispatched based on the total amount of backup energy storage and the adjustment characteristics; when the actual dispatching demand of the power grid is a flexible regulation demand of the power grid, the power grid is dispatched based on the adjustment characteristics.

[0014] Beneficial effects of this program: By clustering regional polymorphic resources with the same characteristics and obtaining the final clustering result based on an objective function that reflects the rationality of the final clustering result, the rationality of the final clustering result is improved. Regional polymorphic resources are aggregated to obtain virtual units through a graph model. When scheduling is required, they are directly located in the virtual units. Then, scheduling tasks are assigned to the regional polymorphic resources in the virtual units based on the characteristics of the regional polymorphic resources in the final clustering result. This takes into account the differences in characteristics of different polymorphic resources in the actual scheduling process, significantly improving the efficiency of collaborative scheduling of regional polymorphic resources. The grid model is divided into graph models based on its topological characteristics, and regional polymorphic resources are aggregated through the graph model to obtain virtual units, thus avoiding the aggregation of distant regional polymorphic resources and nodes, thereby improving the response speed of scheduling.

[0015] The present invention also provides a computer device comprising: a processor, a memory and a bus, wherein the memory stores machine-readable instructions executable by the processor. When the computer device is running, the processor and the memory communicate via the bus, and the processor executes the machine-readable instructions to perform the steps of the method for virtual aggregation of regional polymorphic resources for real-time scheduling needs of the power grid. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Other features, objects, and advantages of the present invention will become more apparent upon reading the detailed description of non-limiting embodiments made with reference to the following drawings. The drawings are provided for the purpose of illustrating preferred embodiments only and are not to be construed as limiting the present invention. Like reference characters are used throughout the drawings to designate like parts.

[0017] Figure 1 The figure is a flow chart of the method for virtual aggregation of regional multi-modal resources for real-time power grid dispatching requirements according to the present invention. DETAILED DESCRIPTION

[0018] In order to make the objectives, technical solutions and advantages of the present invention more clear, the present invention is further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific implementation method described herein is only an optimal embodiment of the present invention, which is only used to explain the present invention and does not limit the scope of protection of the present invention. All other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0019] Before discussing the exemplary embodiments in more detail, it should be mentioned that some exemplary embodiments are described as processes or methods depicted as flow charts. Although the flow charts describe the operations (or steps) as sequential processes, many of the operations (or steps) therein can be performed in parallel, concurrently, or simultaneously. In addition, the order of the operations can be rearranged. The process can be terminated when its operations are completed, but can also have additional steps not included in the figures; the process can correspond to a method, function, procedure, subroutine, subprogram, etc.

[0020] Example 1: like Figure 1 As shown, a virtual aggregation method of regional polymorphic resources for real-time dispatching needs of power grids includes the following steps: constructing a power grid model, clustering regional polymorphic resources based on the operating data of regional polymorphic resources in the power grid model, and obtaining the final clustering result based on the objective function.

[0021] In this embodiment, the power grid model includes an original power grid model and a regional multi-morphic resource model, wherein the construction of the original power grid model realizes the acquisition and storage of the physical model of the power grid. The regional power grid model above 10kV is obtained through the CIM / E model file or model service provided by the external system, including the attributes and connection relationships of the primary equipment of the power grid. Since the data is generally relational list data based on the device type, it is first converted into hierarchical data describing the physical connection relationship of the power grid and the original power grid model is generated. The multi-morphic resource model is obtained through a data file or model service in a standardized description format, including the device attributes and connection relationships of various types of distributed sources, loads, and storage resources in various forms. The association relationship between the original power grid model and the regional multi-morphic resource model is established through the Chinese name or keyword in the device attribute, and each resource in the regional multi-morphic resource model is connected to the nearest bus node above 10kV in the original power grid model to obtain the power grid model.

[0022] As a preferred embodiment, clustering regional polymorphic resources based on the operating data of regional polymorphic resources in the power grid model and obtaining a final clustering result based on an objective function includes: Based on the operating data, clustering the regional polymorphic resources using spectral clustering to obtain initial clustering results; Obtain the center point of each cluster in the initial clustering result, input the center point into the objective function to match the initial clustering result with the requirements. If the requirements match, the initial clustering result is the final clustering result. If the requirements match fails, the initial clustering result is modified based on the constraint conditions to obtain the final clustering result.

[0023] In this embodiment, operational data is obtained from 96-point data files or data services for each region's polymorphic resources. Data is stored as 15-minute intervals, organized by date. This generates 96 data files daily, resulting in over 35,000 data files annually. In this embodiment, the objective function represents the user's actual needs. If the user's need is to minimize loss, the initial clustering result that matches this need is the final clustering result that minimizes loss. In this case, if the objective function value is less than a preset value, the need is matched. The preset value is flexibly set based on user needs. If the objective function value is greater than or equal to the preset value, the need is matched. In this case, the initial clustering result is modified by adjusting the parameters in spectral clustering using constraints to obtain the final clustering result. This improves the rationality of the final clustering result while ensuring its alignment with actual needs.

[0024] The clustering of regional polymorphic resources using spectral clustering based on the operating data to obtain initial clustering results includes: Calculate the similarity of the running data at different time sections in the running data, and obtain the graph Laplacian matrix based on the similarity; The eigenvalue inflection points of the graph Laplacian matrix are obtained, the total number of clusters is obtained based on the eigenvalue inflection points, and the regional polymorphic resources are clustered according to the total number of clusters to obtain the initial clustering results.

[0025] Taking operational data from different time sections as a sample, we used a Gaussian kernel function to calculate the similarity between different samples and construct a similarity matrix S. The degree matrix D of this similarity matrix was then calculated. The graph Laplacian matrix L was obtained using the formula L = DS. Eigendecomposition of L was performed to obtain eigenvalues ​​a, b, and c. The differences between these eigenvalues ​​were calculated. When cb significantly changed from ba, b was considered an inflection point, and the total number of clusters was 2. Based on the total number of clusters, K-means clustering was used to cluster the regional polymorphic resources to obtain initial clustering results. Determining the total number of clusters based on the eigenvalue inflection point avoids bias caused by artificially presetting the total number of clusters, improving the objectivity and rationality of the total number of clusters.

[0026] Specifically, the objective function is: Where N y represents the number of all scenarios corresponding to the operating data of different time sections, Y represents the average energy loss value corresponding to all scenarios, [L j ] represents the cluster to which the multi-modal resources in each time period of the j-th scene belong in all scenes, num represents the number of time periods in the j-th scene, A represents the total number of clusters, P loss，α Indicates the energy loss value corresponding to the center point of the αth cluster in the initial clustering result.

[0027] In this embodiment, [P loss，α ] A×1 =[P loss1，α ,...,P lossa，α ,...,P lossA，α ] T ;[L j ] contains num row vectors l line , describes the classification of random conditions in 96 time periods every day of the year. Each row vector has A-1 0 elements and 1 1 element. The position of the 1 element corresponds to the class to which the time period belongs. For example, a certain time period of a scene belongs to the first class, that is, the row vector describing the time period is [1,0,0,...], and the corresponding loss value is P loss1 .

[0028] Specifically, the constraints include node active power balance constraints, node reactive power balance constraints, distributed power supply installation quantity constraints, and regional multi-modal resource group quantity constraints.

[0029] In this embodiment, the node active power balance constraint, node reactive power balance constraint, and distributed power supply installation quantity constraint are shown in the following formulas: in, and are respectively the active and reactive outputs of the conventional power supply at node i, and are the active and reactive loads at node i, respectively, w i is an integer variable indicating the number of distributed wind turbines installed at node i. is the active power output of the distributed wind turbine, r i is an integer variable for the distributed photovoltaic installed at node i, is the distributed volt-active power output, n is the number of grid nodes, and are the voltage phase angles of nodes i and q, respectively, and Y iq is the admittance modulus between nodes i and q, θ iq is the admittance angle between nodes i and q, and The voltage phase angles at nodes i and q, respectively, are i 、n i are 0 and 1 variables for the locations of distributed wind power and photovoltaic installations, respectively, w max and w min are the maximum and minimum number of distributed wind turbines installed at node i, r max and r min are the maximum and minimum number of distributed photovoltaics installed at node i, respectively. The group number constraint for regional polymorphic resources is as follows: Among them, U min is the lower limit of the node voltage; U max is the upper limit of the node voltage; P rw and P rs are the rated power of each unit of new energy; M B and M N The maximum limits on the total number of renewable energy deployment points are as follows. The maximum limit on the access capacity of distributed resources is given by the following formula: The total access capacity should not exceed the sum of γ1 times the output of conventional power sources and γ2 times the load. In practical applications, this limit can be adjusted based on local policy restrictions.

[0030] Based on the topological characteristics of the power grid model, the power grid model is divided to obtain a graph model, and based on the graph model, the regional multi-form resources are aggregated to obtain virtual units.

[0031] As a preferred embodiment, the method of dividing the power grid model based on the topological characteristics of the power grid model to obtain a graph model includes: The power grid model is divided into an ultra-high voltage main network, a high voltage main network, and a high voltage distribution network according to the topological characteristics, and the busbar node in the high voltage main network is used as the first vertex and the branch in the high voltage main network is used as the first edge to construct a first graph model; The busbar node in the high-voltage distribution network is taken as the second vertex, and the branch in the high-voltage distribution network is taken as the second edge to construct the second graph model; the ultra-high voltage main network is taken as the third vertex, and the third vertex, the first graph model and the second graph model are associated through the association equipment between the ultra-high voltage main network, the high voltage main network and the high-voltage distribution network to obtain the graph model.

[0032] In this embodiment, the ultra-high voltage main grid is the 500kV and above portion of the power grid, the high voltage main grid is the 220kV portion of the power grid, and the high voltage distribution network is the portion of the power grid below 220kV. The associated equipment between the ultra-high voltage main grid and the high voltage main grid is the 500kV main transformer, and the associated equipment between the high voltage main grid and the high voltage distribution network is the 220kV main transformer. By dividing the power grid model into the ultra-high voltage main grid, the high voltage main grid, and the high voltage distribution network, and constructing graph models for each, the graph model can support the collaborative optimization of hierarchical partitioning. At the same time, the graph model aggregates regional multi-modal resources to obtain virtual units, avoids aggregating remote regional multi-modal resources and nodes, and thus improves the response speed of scheduling.

[0033] As a preferred embodiment, the step of aggregating regional polymorphic resources based on a graph model to obtain virtual units includes: The node where the regional polymorphic resource is located is used as the starting node, and the graph model is traversed in a breadth-first manner. The aggregation node is obtained according to the preset rules, and the regional polymorphic resource is aggregated at the aggregation node to obtain a virtual unit.

[0034] In this embodiment, the preset rules are specifically as follows: a breadth-first traversal search is performed on the graph model, and the first 220kV bus node found is the aggregation node; a breadth-first traversal search is performed on the graph model, and the bus nodes connected together are the same partition, and the nodes belong to the corresponding partition according to the 220kV bus node to which they are attached. In this embodiment, the virtual unit is a partition or node that includes regional polymorphic resources. The regional polymorphic resources are aggregated through the graph model to obtain a virtual unit. When scheduling is required, the virtual unit can be directly located so that the regional polymorphic resources in the virtual unit are allocated scheduling tasks according to the characteristics of the regional polymorphic resources in the final clustering result. This takes into account the differences in characteristics of different polymorphic resources in the actual scheduling process, and significantly improves the efficiency of collaborative scheduling of regional polymorphic resources.

[0035] According to the actual dispatching requirements of the power grid, virtual units and the final clustering results are used to dispatch the power grid.

[0036] Specifically, the actual grid dispatching demand includes peak load regulation demand and grid flexible regulation demand.

[0037] As a preferred embodiment, the method of using virtual units and final clustering results to perform grid scheduling according to actual grid scheduling requirements includes: When the actual dispatching demand of the power grid is a peak-shaving demand, the total amount of backup energy storage of the virtual unit is obtained, and the virtual unit is compared with the final clustering result to obtain the adjustment characteristics of the virtual unit, and the power grid is dispatched based on the total amount of backup energy storage and the adjustment characteristics; when the actual dispatching demand of the power grid is a flexible regulation demand of the power grid, the power grid is dispatched based on the adjustment characteristics.

[0038] In this embodiment, peak-shaving requirements include both peak load regulation and peak-load shifting, while grid flexible regulation requirements include overload mitigation and renewable energy integration. Peak load regulation refers to the process by which generators adjust their output to meet power demand based on changes in grid load. The primary purpose of peak load regulation is to maintain power balance and maintain system frequency stability. Regulation characteristics are regulation rate and regulation capacity. Based on peak load regulation requirements, multi-modal resources within a zone are screened and allocated, effectively supporting grid peak load regulation. Peak load shifting aims to reduce peak loads, fill troughs, and minimize the difference between peak and valley loads, thereby balancing power generation and consumption. By analyzing the regulation requirements of each zone over different time periods and combining them with the power consumption patterns of multi-modal resources, multi-modal resources within the zone are screened and allocated, enabling rational and planned scheduling of regional multi-modal resource usage. Overload mitigation involves real-time or advanced power flow analysis of key sections of the regional power grid. When overload or overload risks occur, the power flow in the section is quickly reduced to eliminate the overload risk. A sensitivity analysis is performed on the busbar nodes to screen out busbar nodes that are highly sensitive to over-limit nodes. Combined with the regulation rate and regulation capacity of multi-form resources, the multi-form resources attached to the screening busbar nodes are screened and the regulation amount is apportioned, thereby effectively reducing the section current. New energy consumption refers to ensuring the full consumption of new energy output when new energy is generated and electricity load is low. It usually occurs at noon and holidays in sunny weather, but the strong randomness, intermittency and volatility of new energy may also appear in other time periods. A sensitivity analysis is performed on the busbar nodes to screen out busbar nodes that are highly sensitive to over-limit nodes. Combined with the regulation capacity of multi-form resources, the multi-form resources attached to the screening busbar nodes are screened and the regulation amount is apportioned, thereby ensuring the consumption of new energy. The present invention achieves the functions of peak-shaving and frequency regulation, peak shaving and valley filling, and over-limit elimination by effectively guiding various types of regional multi-form resources to participate in the scheduling and balancing of the new power system, thereby improving the safety and reliability of power grid operation.

[0039] Example 2: This embodiment also provides a computer device, including: a processor, a memory and a bus, wherein the memory stores machine-readable instructions executable by the processor. When the computer device is running, the processor and the memory communicate through the bus, and the processor executes the machine-readable instructions to perform the steps of the regional multi-modal resource virtual aggregation method for real-time scheduling needs of the power grid.

[0040] The specific implementation method described above is a preferred implementation method of the regional multi-form resource virtual aggregation method for the real-time scheduling needs of the power grid of the present invention. It does not limit the specific implementation scope of the present invention. The scope of the present invention includes but is not limited to this specific implementation method. Any equivalent changes made in accordance with the shape and structure of the present invention are within the protection scope of the present invention.

Claims

1. A regional multi-modal resource virtual aggregation method for real-time grid dispatching needs, characterized by: The following steps are involved: Build a power grid model, cluster regional multi-modal resources based on their operating data, and obtain the final clustering results based on the objective function; Based on the topological characteristics of the power grid model, the power grid model is divided to obtain a graph model, and based on the graph model, the regional multi-modal resources are aggregated to obtain virtual units; According to the actual dispatching requirements of the power grid, virtual units and the final clustering results are used to dispatch the power grid.

2. The method for virtual aggregation of regional multi-modal resources oriented to real-time grid dispatching requirements according to claim 1 is characterized in that: Clustering the regional multi-form resources based on the operation data of the regional multi-form resources in the power grid model and obtaining the final clustering result based on the objective function includes: Based on the operating data, clustering the regional polymorphic resources using spectral clustering to obtain initial clustering results; Obtain the center point of each cluster in the initial clustering result, input the center point into the objective function to match the initial clustering result with the requirements. If the requirements match, the initial clustering result is the final clustering result. If the requirements match fails, the initial clustering result is modified based on the constraint conditions to obtain the final clustering result.

3. The method for virtual aggregation of regional multi-modal resources oriented to real-time grid dispatching requirements according to claim 2 is characterized in that: The clustering of regional polymorphic resources using spectral clustering based on the operating data to obtain initial clustering results includes: Calculate the similarity of the running data at different time sections in the running data, and obtain the graph Laplacian matrix based on the similarity; The eigenvalue inflection points of the graph Laplacian matrix are obtained, the total number of clusters is obtained based on the eigenvalue inflection points, and the regional polymorphic resources are clustered according to the total number of clusters to obtain the initial clustering results.

4. The method for virtual aggregation of regional multi-modal resources oriented to real-time grid dispatching requirements according to claim 3 is characterized in that: The objective function is: Where N y represents the number of all scenarios corresponding to the operating data of different time sections, Y represents the average energy loss value corresponding to all scenarios, [L j ] represents the cluster to which the multi-modal resources in each time period of the j-th scene belong in all scenes, num represents the number of time periods in the j-th scene, A represents the total number of clusters, P loss，α Indicates the energy loss value corresponding to the center point of the αth cluster in the initial clustering result.

5. The method for virtual aggregation of regional multi-modal resources oriented to real-time grid dispatching requirements according to claim 2 is characterized in that: The constraints include node active power balance constraints, node reactive power balance constraints, distributed power supply installation quantity constraints and regional multi-modal resource group number constraints.

6. The method for virtual aggregation of regional multi-modal resources oriented to real-time grid dispatching requirements according to claim 1 is characterized in that: The method of dividing the power grid model based on the topological features of the power grid model to obtain a graph model includes: The power grid model is divided into an ultra-high voltage main network, a high voltage main network, and a high voltage distribution network according to the topological characteristics, and the busbar node in the high voltage main network is used as the first vertex and the branch in the high voltage main network is used as the first edge to construct a first graph model; A second graph model is constructed by taking the busbar node in the high-voltage distribution network as the second vertex and the branch in the high-voltage distribution network as the second edge; The ultra-high voltage main network is taken as the third vertex, and the third vertex, the first graph model and the second graph model are associated through the association devices between the ultra-high voltage main network, the high voltage main network and the high voltage distribution network to obtain the graph model.

7. The method for virtual aggregation of regional multi-modal resources oriented to real-time grid dispatching requirements according to claim 1 is characterized in that: The method of aggregating regional multi-modal resources based on a graph model to obtain virtual units includes: The node where the regional polymorphic resource is located is used as the starting node, and the graph model is traversed in a breadth-first manner. The aggregation node is obtained according to the preset rules, and the regional polymorphic resource is aggregated at the aggregation node to obtain a virtual unit.

8. The method for virtual aggregation of regional multi-modal resources oriented to real-time dispatching requirements of power grids according to claim 1 is characterized in that: The actual dispatching demand of the power grid includes peak load regulation demand and flexible control demand of the power grid.

9. The method for virtual aggregation of regional multi-modal resources oriented to real-time grid dispatching requirements according to claim 8, characterized in that: The method of using virtual units and the final clustering results to perform grid dispatch according to the actual grid dispatch requirements includes: When the actual dispatching demand of the power grid is peak-shaving demand, the total amount of backup energy storage of the virtual unit is obtained, and the virtual unit is compared with the final clustering result to obtain the regulation characteristics of the virtual unit. The power grid is dispatched based on the total amount of backup energy storage and the regulation characteristics. When the actual dispatching demand of the power grid is the flexible regulation demand of the power grid, the power grid is dispatched based on the regulation characteristics.

10. A computer device comprising: A processor, a memory and a bus, wherein the memory stores machine-readable instructions executable by the processor. When the computer device is running, the processor and the memory communicate via the bus, and the processor executes the machine-readable instructions to perform the steps of the method for virtual aggregation of regional polymorphic resources for real-time scheduling needs of power grids as described in any one of claims 1 to 9.

Images