Knowledge graph-based power grid multi-energy dynamic aggregation equivalent control method and system

Abstract

The present invention discloses a method and system for controlling the dynamic aggregation equivalent of multiple energy sources in a power grid based on a knowledge graph, which relates to the technical field of power grid energy aggregation equivalent modeling, and includes the following steps: obtaining multi-source data of the power grid energy structure, and constructing a knowledge graph library for the aggregation equivalent of multiple types of energy in the power grid; based on the electrical similarity between the knowledge graph library and the residual voltage of the node, quickly aggregating and dynamically regulating multiple types of energy in the power grid, and constructing an aggregation equivalent model of multiple energy sources in the power grid; simulating the flow of the aggregation equivalent model, obtaining evaluation data of the aggregation equivalent model, and controlling the structure and parameters of the aggregation equivalent model according to the evaluation data; the present application performs flexible equivalence of multiple types of energy in the power grid by considering the rapid and real-time aggregation equivalent method of multiple types of energy in the power grid based on the knowledge graph, so as to solve the problem that centralized, distributed new energy and load-shedding energy structures cannot be automatically and flexibly aggregated to any voltage level power grid in real time, and is not suitable for rapid and real-time aggregation equivalent of multiple types of energy.

Description

Technical Field

[0001] The present invention relates to the technical field of power grid energy aggregation equivalent modeling technology, and more specifically, to a knowledge graph-based power grid multi-energy dynamic aggregation equivalent control method and system. Background Art

[0002] Currently, with the continuous improvement of the national economic level and the continuous advancement of energy internet technology, the power grid energy structure has gradually shown characteristics of multiple energy types, complex structure, wide distribution, and strong coupling. Compared with traditional thermal power generation, the existing energy structure has significant differences in characteristics, operating mode, and response speed. At the same time, the complex and changing power grid topology poses a huge challenge to the stable operation, real-time optimization scheduling, and analytical calculation of the power grid. Traditional power grid equivalence methods often focus on establishing equivalence models for the main grid of large power grids. However, they are unable to effectively integrate information from various distributed energy sources in real time and accurately reflect their comprehensive role in the power grid. They are no longer suitable for the rapid and real-time aggregation of equivalence of multiple energy types, resulting in limited accuracy and efficiency of power grid operation control.

[0003] For example, the invention patent with announcement number: CN119644722A discloses a method for modeling the aggregation equivalent of a new energy base containing multiple control strategies, including: constructing a relational equivalent model that satisfies the terminal voltage of a new energy station containing multiple control strategies and the bus voltage of the new energy base; constructing a positive-sequence network equivalent model and a negative-sequence network equivalent model of the new energy station; performing aggregation equivalent of the new energy base according to the relational equivalent model and the positive-sequence network equivalent model to obtain a positive-sequence network aggregation equivalent model; performing series-parallel aggregation equivalent calculation according to the topological relationship of the new energy base and the negative-sequence network equivalent model to obtain a negative-sequence network aggregation equivalent model; constructing a new energy base aggregation equivalent model according to the positive-sequence network aggregation equivalent model and the negative-sequence network aggregation equivalent model, wherein the new energy base aggregation equivalent model is used to perform short-circuit calculation on the power system containing the new energy base.

[0004] For example, the invention patent with announcement number CN114513004A discloses a new energy station equivalent method and application based on an improved k-means algorithm, including deriving an analytical expression for the short-circuit current of multiple machines of a new energy power source and performing phase conversion; establishing an analytical relationship between an equivalent error and a station electrical quantity; based on the voltage-controlled current source characteristics of the new energy power source, using an optimization algorithm to calculate the equivalent voltage and impedance of the units classified into the same category, using the obtained equivalent error as the particle spacing, using the equivalent voltage as the cluster center, using the algorithm to perform group equalization on the new energy units, determining the final number of groups according to the silhouette coefficient, and completing the group group equalization. The present invention uses the equivalent error as the particle spacing of the clustering algorithm, and uses the improved k-means algorithm to perform group equalization, which simplifies the model complexity of the new energy power source, ensures a certain accuracy, reduces the amount of calculation and simulation time, reduces the number of nodes in the current calculation process of the network containing the new energy power source, and improves the calculation speed and iterative calculation efficiency.

[0005] The above disclosed technical solutions have at least the following technical problems:

[0006] Existing grid equivalence methods struggle to effectively integrate distributed energy information in real time. They are unable to quickly and flexibly aggregate and equivalence energy structures, including centralized and distributed renewable energy, wind power, thermal power, and shelvable loads, into any voltage grid. This approach is no longer suitable for rapidly and real-time aggregation of multiple energy sources. To address this issue, the present invention proposes a solution. Summary of the Invention

[0007] In order to overcome the above-mentioned defects of the prior art, an embodiment of the present invention provides a method and system for controlling the dynamic aggregation of multiple energy sources in a power grid based on a knowledge graph. By considering a method for rapid and real-time aggregation of multiple energy types in a power grid based on a knowledge graph, flexible equivalence of multiple energy types in a power grid is performed to solve the problem that energy structures such as centralized and distributed new energy, wind power, thermal power and cuttable loads cannot be automatically and flexibly aggregated to any voltage level power grid in real time, and are no longer suitable for rapid and real-time aggregation of multiple energy types.

[0008] To achieve the above object, the present invention provides the following technical solutions:

[0009] A knowledge graph-based dynamic aggregation equivalent control method for multiple energy sources in a power grid includes the following steps: obtaining multi-source data on the power grid energy structure and constructing a knowledge graph library for the aggregation equivalent of multiple types of energy in the power grid; based on the electrical similarity between the knowledge graph library and the node residual voltage, quickly aggregating and dynamically regulating multiple types of energy in the power grid to construct an aggregation equivalent model for multiple energy sources in the power grid; simulating the flow of the aggregation equivalent model to obtain evaluation data of the aggregation equivalent model, and controlling the structure and parameters of the aggregation equivalent model based on the evaluation data, wherein the evaluation data includes the error of the aggregation equivalent model.

[0010] In a preferred embodiment, the method of obtaining multi-source data on the energy structure of the power grid and constructing a knowledge graph library of the aggregation of multiple types of energy in the power grid is as follows: preprocessing the multi-source structure data of the measured area to establish a structured data set; extracting the features of the processed multi-source structure data, and treating various types of energy equipment as entity nodes in the knowledge graph; obtaining the entity relationship data of the power grid according to the triple pattern, and setting three types of attribute information for each entity node to construct a triple knowledge graph model; mapping the processed multi-source structure data to the entities and attributes in the triple knowledge graph, and automatically generating edges between entities according to the power grid topology structure to obtain the knowledge graph of the aggregation of multiple types of energy in the power grid.

[0011] In a preferred embodiment, the electrical similarity between the knowledge graph library and the node residual voltage is used to quickly aggregate and dynamically regulate multiple types of energy in the power grid, and to construct an aggregated equivalent model of multiple energy sources in the power grid, specifically as follows: identify aggregation targets based on graph attribute information, retrieve adjustable energy devices under all voltage levels through the knowledge graph, screen out node devices that meet aggregation conditions, and identify groups of node devices of the same type and access conditions; obtain the electrical similarity of different nodes based on node residual voltage and device information, and divide node devices with similar characteristics into coherent groups; equate the aggregated coherent group to a single equivalent device, and output the operating data of the aggregated equivalent device through the weighted average method; construct an aggregated equivalent model based on the node residual voltage attribute in the equivalent information class in the knowledge graph library as a physical quantity to measure electrical distance.

[0012] In a preferred embodiment, the node residual voltage attribute in the equivalent information class in the knowledge graph library is used as a physical quantity to measure the electrical distance to construct an aggregated equivalent model, specifically as follows: the node residual voltage value of each node is extracted according to the power grid operation data and the knowledge graph information, and the external system and the system to be equivalent are divided according to the node residual voltage value; through the knowledge graph structure, the shortest path from the source power generation equipment node to the target bus is retrieved; based on the node residual voltage value and its changing trend, the electrical coupling degree is analyzed, and the equivalent nodes are selected according to the electrical coupling degree and the shortest path to construct an aggregated equivalent model under the target voltage level.

[0013] In a preferred embodiment, the electrical coupling degree is obtained based on the node residual voltage value and its changing trend, and equivalent nodes are selected according to the electrical coupling degree and the shortest path to construct an aggregated equivalent model under the target voltage level, specifically as follows: the operating data of the device to be equivalent is obtained, and the node residual voltage value between the node of the device to be equivalent and the node of the target voltage level is extracted; based on the power grid topology and impedance model, the equivalent electrical distance from the source node to the target node is output; the electrical coupling degree is output according to the node residual voltage value and the electrical distance, and the injection node of the aggregated equivalent model is screened out; according to the set residual voltage threshold and distance threshold, the equivalent devices in the equivalent nodes participating in the equivalent modeling are screened; the operating parameters of the device to be equivalent are weighted and aggregated to form an equivalent device model; the equivalent device model is connected to the injection node in parallel in the power grid flow model, and the original device node is deleted in the network topology to obtain an aggregated equivalent model.

[0014] In a preferred embodiment, the flow simulation of the aggregate equivalent model is performed to obtain evaluation data of the aggregate equivalent model, and the structure and parameters of the aggregate equivalent model are controlled according to the evaluation data, specifically as follows: real-time operation data of the equipment is obtained, and input into the aggregate equivalent model to perform steady-state flow and short-circuit simulation to generate flow simulation prediction results; the flow simulation prediction results are compared with the actual operation data to obtain the error of the aggregate equivalent model; the structure and parameters of the aggregate equivalent model are controlled according to the error results; the operation status of multiple energy sources of the power grid is simulated by the corrected aggregate equivalent model, and the knowledge graph library is dynamically updated in combination with the topology structure and energy data.

[0015] In a preferred embodiment, the structure and parameters of the aggregate equivalent model are controlled according to the error results, specifically as follows: based on the error heat map of the flow simulation prediction results and the actual operation data, the error distribution characteristics and concentrated areas are identified, and the error position and error intensity of the node are extracted; based on the error position and error intensity, the operating parameters of the aggregate equivalent model are corrected, and the model structure is reconstructed, and the model structure reconstruction includes the judgment of the mismatch state of the injection node.

[0016] In a preferred embodiment, the model structure is reconstructed as follows: based on the node residual voltage and electrical distance, the ownership boundary of the aggregated device is re-divided, and the subordinate relationship between the device entity and the target node in the knowledge graph is updated; based on the error heat map, the error intensity of the adjacent nodes of the injection node is obtained, and if the average difference in error intensity between adjacent nodes exceeds a threshold, the current injection node is determined to be mismatched; based on real-time electrical coupling and path length analysis, the injection node is reselected, the aggregation structure is adjusted, and the equivalent connection relationship is updated.

[0017] A system for a dynamic aggregation equivalent control method of multiple energy sources in a power grid based on a knowledge graph is characterized in that it includes a knowledge graph library module, a module for constructing an aggregation equivalent model, and a module for adjusting an aggregation equivalent model, and there are connections between the modules; the knowledge graph library module is used to obtain multi-source data on the power grid energy structure and construct a knowledge graph library for the aggregation equivalent of multiple types of energy in the power grid; the aggregation equivalent model construction module is used to quickly aggregate and dynamically regulate multiple types of energy in the power grid based on the electrical similarity between the knowledge graph library and the node residual voltage, and construct an aggregation equivalent model of multiple energy sources in the power grid; the aggregation equivalent model adjustment module is used to simulate the flow of the aggregation equivalent model, obtain evaluation data of the aggregation equivalent model, and control the structure and parameters of the aggregation equivalent model according to the evaluation data, wherein the evaluation data includes the error of the aggregation equivalent model.

[0018] The technical effects and advantages of the knowledge graph-based power grid multi-energy dynamic aggregation equivalent control method and system of the present invention are as follows:

[0019] 1. The present invention constructs a multi-source data-driven power grid energy structure model through knowledge graph technology, realizing efficient aggregation of distributed energy equipment in the power grid. Based on the entity node and attribute mapping of the knowledge graph, it can accurately identify groups of equipment of the same type and with the same access conditions, and form synchronization groups through electrical similarity analysis, significantly reducing the complexity of the model. For example, the aggregated synchronization generator group is equivalent to a single device, which greatly reduces the amount of calculation. At the same time, the operating data of the equivalent equipment is output through the weighted average method, ensuring the accuracy and real-time performance of the model. The knowledge graph library module realizes the deep integration and intelligent analysis of multi-source data of the power grid through unified structured data sets and triple models. This module not only supports the rapid retrieval and relationship reasoning of equipment entities, but also provides high-precision data support for subsequent aggregation and equivalence through attribute mapping and topology generation of entity nodes. For example, when processing distributed energy access, the knowledge graph can automatically identify the device type and access conditions, significantly improving the modeling efficiency.

[0020] 2. The present invention constructs an aggregated equivalent model based on the knowledge graph through dynamic analysis of node residual voltage and electrical distance. This method can extract node residual voltage values and divide system boundaries in real time based on grid operation data and knowledge graph information, and screen equivalent nodes through the shortest path and electrical coupling to form a dynamic equivalent model under the target voltage level. For example, when the grid topology changes or equipment fails, the system can quickly adjust the equivalent nodes to ensure that the model is always consistent with the actual grid status. The electrical similarity analysis method based on node residual voltage and equipment information breaks through the limitations of traditional aggregation methods. This method can accurately identify node devices with similar characteristics through the joint analysis of residual voltage change trends and electrical coupling, ensuring that the aggregated equivalent devices are highly consistent with the original system in electrical characteristics. For example, when constructing an equivalent model, the system can screen equivalent nodes based on the residual voltage threshold and distance threshold to avoid model errors caused by parameter mismatch. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 This is a flow chart of the knowledge graph-based method for dynamic aggregation and equivalent control of multiple energy sources in a power grid according to the present invention.

[0022] Figure 2 This is a schematic diagram of the system structure of the knowledge graph-based power grid multi-energy dynamic aggregation equivalent control method of the present invention.

[0023] Figure 3 It is the triple knowledge graph model diagram.

[0024] Figure 4 Partition diagrams for external systems and systems requiring equivalence. DETAILED DESCRIPTION

[0025] The following will provide a clear and complete description of the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0026] Example 1, Figure 1 The present invention provides a method for controlling the dynamic aggregation of multiple energy sources in a power grid based on a knowledge graph, which includes the following steps:

[0027] S1, obtain multi-source data on the power grid energy structure and build a knowledge graph library of the aggregation equivalent of multiple types of energy in the power grid.

[0028] In this embodiment, multi-source data on the power grid energy structure is obtained to construct a knowledge graph library of aggregated equivalent values of multiple types of energy in the power grid, as follows:

[0029] Acquire multi-source structural data of the area to be measured, establish a structured data set, and use a rule engine to identify and eliminate abnormal data in the multi-source structural data. Standardize parameters with large dimensional differences (such as voltage level and capacity) to ensure data comparability. The multi-source structural data includes main network data and distribution network data.

[0030] Main grid data is obtained through the power grid full-time and space system, the power grid "one map", PMS, and D5000 platform;

[0031] Distribution network data obtains distributed new energy related data through the distribution network diagram system and marketing system;

[0032] Extract the features of the processed multi-source structured data. Based on the characteristics of the multi-source structured data, various types of energy equipment (such as wind farms, photovoltaic stations, load nodes, small hydropower stations, etc.) are used as entity nodes in the knowledge graph. Each type of entity corresponds to a specific equipment unit.

[0033] Using structured data sets, we automatically extract power grid entity relationship data based on the standard "subject-predicate-object" triplet model. We then assign three types of attribute information to each entity node to construct a triplet knowledge graph model. This model is used to describe the structural characteristics, operating status, and equivalent modeling parameters of devices. These three types of attribute information include device information, equivalent information, and power flow information. The entity relationship data refers to the various types of devices in the power grid and the relationships between them. Specifically, it refers to data represented in the power grid knowledge graph using a triplet model (subject-predicate-object). This data includes the definitions of entities and the relationships between them.

[0034] The equipment information category has four attributes: voltage level, grid-connected power station, equipment model, and energy type; the flow information category has five attributes: voltage, current, active power, reactive power, and power factor; and the equivalent information category has five attributes: node residual voltage, rated capacity, equivalent load, equivalent generator voltage, and equivalent branch current.

[0035] The processed multi-source structured data is mapped to entities and attributes in the triple knowledge graph (such as Figure 3 As shown in Figure 2), the edges between entities are automatically generated according to the topological structure of the power grid, thereby automatically building a knowledge graph library of multi-category energy aggregation equivalents in the power grid.

[0036] S2, based on the electrical similarity between the knowledge graph library and the node residual voltage, quickly aggregates and dynamically regulates multiple types of energy in the power grid, and constructs an aggregation equivalent model of multiple energy sources in the power grid.

[0037] In this embodiment, based on the electrical similarity between the knowledge graph library and the node residual voltage, multiple types of energy in the power grid are quickly aggregated and dynamically regulated to construct an aggregation equivalent model of multiple energy sources in the power grid, as follows:

[0038] Identify aggregation targets based on graph attribute information, conduct parallel automated searches for adjustable energy devices at all voltage levels through the knowledge graph, quickly screen out node devices that meet aggregation conditions, and identify groups of node devices of the same type and with the same access conditions based on the entity's voltage level, grid-connected power station, device model, energy type, and other attributes;

[0039] Based on node residual voltage and device information, the electrical similarity of different nodes is obtained, and node devices with similar characteristics are divided into synchronization groups (for example, multiple wind turbines are merged into a wind power group);

[0040] The aggregated synchronous generator group is treated as an equivalent device to reduce the model complexity, and the operating data of the aggregated equivalent device is output through the weighted average method;

[0041] Based on the node residual voltage attribute in the equivalent information class in the knowledge graph library, an aggregated equivalent model is constructed as a physical quantity to measure the electrical distance.

[0042] It's important to note that node electrical similarity is a key metric for determining device ownership and injection node selection in aggregate modeling. By integrating voltage deviation, electrical distance, and power output characteristics, a unified scoring function is constructed, forming a matrix analysis system. This approach dynamically identifies electrically similar regions, improving equivalent model matching accuracy and the reliability of power flow simulations.

[0043] In this embodiment, an aggregated equivalent model is constructed based on the node residual voltage attribute in the equivalent information class in the knowledge graph library as a physical quantity to measure the electrical distance, as follows:

[0044] Based on the grid operation data and knowledge graph information, the residual voltage value of each node is extracted (i.e., the steady-state deviation of the node voltage recovery after a fault or disturbance). Nodes with residual voltage values higher than the threshold are classified as external systems, and nodes with residual voltage values lower than the threshold are classified as systems requiring equalization (e.g., Figure 4 shown);

[0045] Through the knowledge graph structure, the shortest path from the source power generation equipment node to the target bus is retrieved;

[0046] Based on the node residual voltage and the residual voltage change trend, the electrical coupling degree is obtained. Equivalent nodes are selected according to the electrical coupling degree and the shortest path, and an aggregated equivalent model is established under the target voltage level.

[0047] In this embodiment, the electrical coupling degree is obtained based on the node residual voltage and the residual voltage variation trend. Equivalent nodes are selected based on the electrical coupling degree and the shortest path, and an aggregated equivalent model is established at the target voltage level. The details are as follows:

[0048] Obtain the operating data of the device to be equivalent, and extract the node residual voltage value between the node of the device to be equivalent and the node of the target voltage level;

[0049] Based on the grid topology and impedance model, the equivalent electrical distance from the source node to the target node is output;

[0050] Output electrical coupling according to node residual voltage value and electrical distance, and select injection nodes of aggregate equivalent model according to electrical coupling;

[0051] Based on the set residual voltage threshold and distance threshold, the equivalent devices in the equivalent nodes participating in the equivalent modeling are selected;

[0052] Perform weighted aggregation on the operating parameters of the equipment to be equivalent to form an equivalent equipment model, wherein the operating parameters include rated capacity and dynamic data (inertia, damping);

[0053] In the power flow model, equivalent devices are connected to the injection nodes in parallel, and the original device nodes are deleted in the network topology to obtain an aggregated equivalent model.

[0054] It should be noted that after selecting the injection node, it is necessary to evaluate the equivalent impedance path between the target node and the virtual connection of the equivalent device. If the reactance of the paths connecting the original multiple device nodes to the injection node is significantly different, it is necessary to construct an equivalent connection impedance to reflect the equivalent coupling relationship between the aggregate device and the power grid, so as to avoid electrical model distortion caused by direct connection of the aggregate model to the busbar and improve simulation accuracy.

[0055] The aggregation calculation formula for dynamic data is as follows:

[0056]

[0057] Where: is the total dynamic data of the equivalent unit after aggregation, is the dynamic data of the i-th power generation unit, is the rated capacity of the i-th device, is the total number of devices being aggregated.

[0058] The calculation formula for electrical coupling is as follows:

[0059]

[0060] Where: is the electrical coupling, is the node residual voltage difference, , is the node residual pressure of node i, is the node residual pressure of node j, is the electrical distance from node i to node j, and is the weight coefficient, satisfying , used to balance the effects of residual voltage and electrical distance on coupling degree.

[0061] It should be noted that the node residual voltage difference indicates the similarity of voltage behavior, especially whether the response after the system disturbance is consistent. The electrical distance reflects the topology and impedance coupling, and reflects the path cost of energy transmission or disturbance propagation. The larger the node residual voltage difference and electrical distance, the more dissimilar or farther they are. Taking the reciprocal form, the greater the coupling degree, the closer and more similar the two nodes are, and the tighter the coupling.

[0062] S3, simulating the aggregate equivalent model power flow, obtaining evaluation data of the aggregate equivalent model, and controlling the structure and parameters of the aggregate equivalent model according to the evaluation data, wherein the evaluation data includes an error of the aggregate equivalent model.

[0063] In this embodiment, the aggregate equivalent model power flow is simulated to obtain evaluation data of the aggregate equivalent model, and the structure and parameters of the aggregate equivalent model are controlled according to the evaluation data, as follows:

[0064] Acquire real-time operating data of the equipment and input the real-time operating data into an aggregate equivalent model to perform steady-state power flow and short-circuit simulations to generate power flow simulation prediction results, including predicted node voltages and branch power distributions;

[0065] Compare the power flow simulation prediction results with the actual operation data, evaluate the error of the aggregate equivalent model, and generate an error heat map;

[0066] Control the structure and parameters of the aggregate equivalent model based on the error results;

[0067] The modified aggregate equivalent model is used to simulate the operating status of multiple energy sources in the power grid, and the knowledge graph library is dynamically updated by combining topology and energy data.

[0068] The corrected model is re-applied to the power flow simulation to compare whether the system responses before and after the correction are consistent. If the error is reduced and the stability is improved, the correction is confirmed to be effective. Otherwise, the model rollback mechanism is activated to restore to the last valid aggregation state.

[0069] In this embodiment, the structure and parameters of the aggregated equivalent model are controlled according to the error results, as follows:

[0070] Based on the error heat map of the power flow simulation prediction results and the actual operation data, the error distribution characteristics and concentrated areas are identified, and the error location and error intensity of the node are extracted;

[0071] If the error is concentrated near a certain injection node, it may indicate that the current injection node selection is unreasonable;

[0072] If the error shows a systematic deviation, it may be due to distortion of device parameter aggregation or deviation in dynamic data setting;

[0073] If the error occurs after a topology change, it is necessary to consider the mismatch of the aggregate structure or the topological path error;

[0074] According to the error location and intensity, the operating parameters in the aggregated equivalent model are corrected in a targeted manner, wherein the targeted correction method includes replacing the corresponding data in the aggregated device with the latest real-time data and re-aggregating the dynamic data;

[0075] When the injection node is mismatched, the aggregate equivalent model structure is controlled.

[0076] In this embodiment, the aggregate equivalent model structure is controlled as follows:

[0077] Based on the node residual voltage and electrical distance, the ownership boundary of the aggregated device is re-divided, and the subordinate relationship between the device entity and the target node in the knowledge graph is updated;

[0078] The error intensity of the adjacent nodes of the injected node is obtained based on the error heat map. If the average difference in the error intensity between adjacent nodes exceeds the threshold, the current injected node is determined to be mismatched.

[0079] Based on the latest electrical coupling and path length analysis, the injection nodes are reselected, the aggregation structure is adjusted, and the equivalent connection relationship is updated;

[0080] Based on the real-time residual voltage and flow sensitivity, the node with the highest electrical coupling is selected as the injection point of the equivalent model. If the system supports multi-node injection, a distributed parallel injection method is used.

[0081] In this embodiment, the modified aggregate equivalent model is used to simulate the operating status of multiple energy sources in the power grid, and the knowledge graph library is dynamically updated by combining the topology structure and energy data, as follows:

[0082] After completing the structure and parameter correction, perform the power flow simulation again to verify the correction effect;

[0083] Compare the simulation error values and thermal maps before and after correction, and write the corrected parameter values, aggregation boundaries, and injection node position updates into the knowledge graph library to achieve dynamic accumulation and updating of model knowledge.

[0084] Example 2, Figure 2 The present invention provides a system for a method for controlling the dynamic aggregation equivalent value of multiple energy sources in a power grid based on a knowledge graph, comprising a knowledge graph library module, an aggregation equivalent value model building module, and an aggregation equivalent value model adjustment module, wherein the modules are connected;

[0085] The knowledge graph library module is used to obtain multi-source data on the power grid energy structure and build a knowledge graph library with aggregated equivalent values for multiple types of energy in the power grid;

[0086] Construct an aggregation equivalent model module to quickly aggregate and dynamically regulate multiple types of energy in the power grid based on the electrical similarity between the knowledge graph library and the node residual voltage, and build an aggregation equivalent model for multiple energy sources in the power grid;

[0087] The aggregation equivalent model adjustment module is used to simulate the aggregation equivalent model flow, obtain the evaluation data of the aggregation equivalent model, and control the structure and parameters of the aggregation equivalent model according to the evaluation data.

[0088] The above formulas are all dimensionless and numerical calculations. The formulas are obtained by collecting a large amount of data and performing software simulation to obtain the most recent real situation. The preset parameters in the formulas are set by technicians in this field according to actual conditions.

[0089] The above embodiments may be implemented in whole or in part through software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments may be implemented in whole or in part in the form of a computer program product.

[0090] Those skilled in the art will appreciate that the modules and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0091] In addition, each functional module in each embodiment of the present application may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module.

[0092] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

[0093] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A knowledge graph-based multi-energy dynamic aggregation equivalent control method for power grids, characterized by: The steps include: Obtain multi-source data on the power grid energy structure and build a knowledge graph library of aggregated equivalent values of multiple types of energy in the power grid; Based on the electrical similarity between the knowledge graph library and the node residual voltage, multiple types of energy in the power grid are quickly aggregated and dynamically regulated to build an aggregation equivalent model for multiple energy sources in the power grid; Simulating the aggregate equivalent model power flow, obtaining evaluation data of the aggregate equivalent model, and controlling the structure and parameters of the aggregate equivalent model according to the evaluation data, wherein the evaluation data includes an error of the aggregate equivalent model; Based on the electrical similarity between the knowledge graph library and the node residual voltage, the multiple energy types of the power grid are quickly aggregated and dynamically regulated to construct an aggregation equivalent model of the multiple energy sources of the power grid, as follows: Identify aggregation targets based on graph attribute information, retrieve adjustable energy devices at all voltage levels through the knowledge graph, screen out node devices that meet aggregation conditions, and identify groups of node devices of the same type and with the same access conditions; Based on node residual voltage and device information, the electrical similarity of different nodes is obtained, and node devices with similar characteristics are divided into synchronization groups; The aggregated coherent group is converted into a single equivalent device, and the operating data of the aggregated equivalent device is output through the weighted average method; An aggregated equivalent model is constructed based on the node residual voltage attribute in the equivalent information class in the knowledge graph library as a physical quantity to measure the electrical distance. The node residual voltage attribute is the steady-state deviation of the node voltage recovery after the disturbance.

2. The method for controlling the dynamic aggregation of multiple energy sources in a power grid based on knowledge graph according to claim 1 is characterized in that: The acquisition of multi-source data on the grid energy structure and the construction of a knowledge graph library of aggregated equivalent values of multiple types of energy in the grid are as follows: Preprocess the multi-source structural data of the area to be measured and establish a structured data set; Extract the features of the processed multi-source structured data and treat various energy devices as entity nodes in the knowledge graph; Obtain power grid entity relationship data based on the triple pattern, set three types of attribute information for each entity node, and build a triple knowledge graph model; The processed multi-source structured data is mapped to the entities and attributes in the triple knowledge graph, and the edges between the entities are automatically generated according to the power grid topology structure to obtain the equivalent knowledge graph of the aggregation of multiple types of energy in the power grid.

3. The method for controlling the dynamic aggregation of multiple energy sources in a power grid based on knowledge graph according to claim 2 is characterized in that: According to the node residual voltage attribute in the equivalent information class in the knowledge graph library, as a physical quantity to measure the electrical distance, an aggregate equivalent model is constructed, as follows: Extract the node residual voltage value of each node based on the grid operation data and knowledge graph information, and divide the external system and the equivalent value system according to the node residual voltage value; Through the knowledge graph structure, the shortest path from the source power generation equipment node to the target bus is retrieved; Based on the node residual voltage value and its changing trend, the electrical coupling degree is analyzed and obtained. Equivalent nodes are selected according to the electrical coupling degree and the shortest path, and an aggregated equivalent model is constructed under the target voltage level.

4. The method for controlling the dynamic aggregation of multiple energy sources in a power grid based on knowledge graph according to claim 3 is characterized in that: Based on the node residual voltage value and its change trend, the electrical coupling degree is analyzed and obtained, and equivalent nodes are selected according to the electrical coupling degree and the shortest path to construct an aggregate equivalent model at the target voltage level, as follows: Obtain the operating data of the device to be equivalent, and extract the node residual voltage value between the node of the device to be equivalent and the node of the target voltage level; Based on the grid topology and impedance model, the equivalent electrical distance from the source node to the target node is output; Output electrical coupling degree according to node residual voltage value and electrical distance, and select injection nodes of aggregate equivalent model; Based on the set residual voltage threshold and distance threshold, the equivalent devices in the equivalent nodes participating in the equivalent modeling are selected; The operating parameters of the equipment to be equivalent are weighted and aggregated to form an equivalent equipment model; In the power grid flow model, the equivalent device model is connected to the injection node in parallel, and the original device node is deleted in the network topology to obtain the aggregated equivalent model.

5. The method for controlling the dynamic aggregation of multiple energy sources in a power grid based on knowledge graph according to claim 4 is characterized in that: The aggregation equivalent model power flow simulation is performed to obtain evaluation data of the aggregation equivalent model, and the structure and parameters of the aggregation equivalent model are controlled according to the evaluation data, as follows: Obtain real-time operating data of the equipment and input it into the aggregate equivalent model to perform steady-state power flow and short-circuit simulation to generate power flow simulation prediction results; Compare the power flow simulation prediction results with the actual operation data to obtain the error of the aggregate equivalent model; Control the structure and parameters of the aggregate equivalent model based on the error results; The modified aggregation equivalent model is used to simulate the operating status of multiple energy sources in the power grid, and the knowledge graph library is dynamically updated by combining the topology structure and energy data.

6. The method for controlling the dynamic aggregation of multiple energy sources in a power grid based on knowledge graph according to claim 5 is characterized in that: The structure and parameters of the aggregation equivalent model are controlled according to the error results, as follows: Based on the error heat map of the power flow simulation prediction results and the actual operation data, the error distribution characteristics and concentrated areas are identified, and the error location and error intensity of the node are extracted; According to the error position and error intensity, the operating parameters of the aggregate equivalent model are corrected, and the model structure is reconstructed. The model structure reconstruction includes determining the mismatch state of the injection node.

7. The method for controlling the dynamic aggregation of multiple energy sources in a power grid based on knowledge graph according to claim 6 is characterized in that: The model structure reconstruction is as follows: Based on the node residual voltage and electrical distance, the ownership boundary of the aggregated device is re-divided, and the subordinate relationship between the device entity and the target node in the knowledge graph is updated; The error intensity of the adjacent nodes of the injected node is obtained based on the error heat map. If the average difference in the error intensity between adjacent nodes exceeds the threshold, the current injected node is determined to be mismatched. Based on real-time electrical coupling and path length analysis, injection nodes are reselected, the aggregation structure is adjusted, and the equivalent connection relationship is updated.

8. A system using the knowledge graph-based multi-energy dynamic aggregation equivalent control method for a power grid as described in any one of claims 1 to 7, characterized in that: It includes a knowledge graph library module, an aggregation equivalence model building module, and an aggregation equivalence model adjustment module, and there are connections between the modules; The knowledge graph library module is used to obtain multi-source data on the power grid energy structure and build a knowledge graph library with aggregated equivalent values for multiple types of energy in the power grid; Construct an aggregation equivalent model module to quickly aggregate and dynamically regulate multiple types of energy in the power grid based on the electrical similarity between the knowledge graph library and the node residual voltage, and build an aggregation equivalent model for multiple energy sources in the power grid; The aggregation equivalent model adjustment module is used to simulate the aggregation equivalent model power flow, obtain evaluation data of the aggregation equivalent model, and control the structure and parameters of the aggregation equivalent model according to the evaluation data, wherein the evaluation data includes the error of the aggregation equivalent model.

Images