Gateway port setting analysis report automatic generation method and system and storage medium

Abstract

This invention relates to the field of power grid management technology, and in particular to a method, system, and storage medium for automatically generating power grid gateway setting analysis reports. The method includes: constructing a topological representation of the power network and identifying ownership boundaries under multi-stakeholder participation; determining the distribution characteristics and quantification results of potential circulating current paths; when the circulating current distribution characteristics exceed a preset threshold, forming an optimized set of locations; constructing and updating a loss allocation model to obtain a quantitative index of loss deviation; calculating the sensitivity score of gateway locations to the balance of interests among multiple stakeholders to obtain the recommended configuration for the final gateway settings; iteratively updating the power grid topology to form a dynamic adaptation scheme for gateway locations and a corresponding gateway setting analysis report. This invention solves the problems of difficulty in dynamically evaluating gateway settings and the reliance on manual generation of analysis reports in multi-source, multi-stakeholder power grids, and achieves intelligent optimization of gateway setting schemes and automatic generation of analysis reports.

Description

Technical Field

[0001] This invention relates to the field of power grid management technology, and in particular to a method, system, and storage medium for automatically generating power grid gateway setting analysis reports. Background Technology

[0002] In the operation and management of power systems, the gate locations serve as key nodes in the power grid for property rights allocation, loss measurement, and benefit settlement. Their rationale directly impacts the stability of the power network, the fairness of loss distribution, and the balance of interests among multiple stakeholders. With the large-scale integration of distributed power sources, renewable energy, and diverse investment entities, the power grid is gradually exhibiting complex characteristics of multiple power sources, multiple stakeholders, strong coupling, and dynamic changes. Traditional gate location setting and analysis methods based on single power sources or static operating conditions are no longer sufficient to meet practical needs.

[0003] In multi-source operation scenarios, fluctuations in power output, load changes, and network topology adjustments can easily lead to circulating currents in the network, resulting in increased losses, local equipment overload, and distorted loss allocation. Improperly positioned gates may cause some entities to bear an unreasonable share of losses for extended periods, while others reap undue benefits, leading to disputes and impacting the safe operation and coordinated development of the power grid. However, current gate location analyses rely heavily on manual experience or single calculations, lacking systematic modeling of multiple operating conditions, uncertainties, and multi-entity interactions. This makes it difficult to accurately assess the comprehensive impact of gate locations on circulating current risks, loss allocation, and the balance of interests. Furthermore, with the expansion of power grids and the rapid growth of operational data, traditional manually compiled gate location analysis reports are inefficient, subjective, and fail to reflect dynamic changes in grid operation status in a timely manner, nor can they provide effective support for continuous optimization of gate locations. Therefore, there is an urgent need for a method that can systematically analyze gate location schemes by combining power grid topology, multi-source operating characteristics, and actual operational data, and automatically generate analysis reports.

[0004] Based on this, a method for automatically generating power grid gateway setting analysis reports is proposed. Through comprehensive analysis of power grid topology, operating conditions and feedback data, the method achieves quantitative assessment and dynamic optimization of the impact of gateway location, providing objective, continuous and traceable technical basis for power grid operation decisions. Summary of the Invention

[0005] This invention provides a method, system, and storage medium for automatically generating power gateway configuration analysis reports. By collaboratively processing power network structure information, operating condition data, and actual feedback data, it enables the dynamic evaluation and automatic generation of dynamic evaluation and analysis reports for gateway configuration schemes in complex multi-source scenarios.

[0006] In a first aspect, the present invention provides a method for automatically generating an analysis report of a power gateway port setting, the method comprising:

[0007] Step S1: Generate a topology representation of the power network by collecting wiring diagram data and multi-source configuration information in the power network; based on the topology representation, obtain the property rights division boundary under the participation of multiple entities, and determine the distribution characteristics and corresponding quantitative results of potential circulation paths by analyzing power flow paths under different operating conditions.

[0008] Step S2: If the quantification result of the distribution feature exceeds the preset threshold, the priority sequence of the gate candidate positions is adjusted to obtain the optimized position set; through the position set, the input parameters of the loss allocation model are obtained and the model is updated to simulate the impact of mutual supply phenomenon and determine the quantification index of loss deviation.

[0009] Step S3: Based on the quantitative indicators, generate dynamic scene samples and determine the sensitivity score of the checkpoint location to the balance of interests; if the sensitivity score is lower than the preset score, select alternative solutions from the optimized location set to obtain the recommended configuration of the final checkpoint setting;

[0010] Step S4: Obtain feedback data of the power network under actual operating conditions through the recommended configuration, determine whether it is necessary to iteratively adjust the topology representation based on the feedback data, and generate a dynamic adaptation scheme and gateway setting analysis report.

[0011] As a preferred embodiment of the present invention, step S1, generating the topological structure representation, includes:

[0012] Obtain wiring diagram data from the power network and combine it with multi-power source configuration information to construct an initial network topology model; use a graph neural network to automatically parse the wiring diagram data and extract the connection relationships between nodes and edges; based on the connection relationships, generate a topology representation of candidate gateway locations, the topology representation including power transmission paths between nodes and power source distribution characteristics.

[0013] As a preferred embodiment of the present invention, step S1, determining the distribution characteristics of potential circulation paths, includes:

[0014] Based on the topology representation, the property rights demarcation boundary under multi-entity participation is extracted to determine the power interaction area between each entity; Monte Carlo simulation is used to simulate and sample power flow paths under different operating conditions to generate multiple sets of flow path data; by analyzing the flow path data, the distribution characteristics of potential circulation paths are identified, including the frequency of occurrence and the range of influence of circulation paths; based on the distribution characteristics, the potential impact of circulation paths on power network stability is quantified and used as the quantification result of the distribution characteristics.

[0015] As a preferred embodiment of the present invention, step S2, obtaining the optimized set of positions, includes:

[0016] The quantization result of the distribution feature is obtained and compared with a preset threshold. If the quantization result exceeds the preset threshold, the influence of the candidate gate positions on the circulation path is analyzed for multi-power source scenarios. Based on the influence, the priority sequence of the candidate gate positions is adjusted to prioritize the selection of positions with less influence on the circulation path. The candidate gate positions that meet the conditions are selected through the priority sequence to generate an optimized set of positions.

[0017] As a preferred embodiment of the present invention, step S2, which involves obtaining the input parameters of the loss allocation model and updating the model to simulate the impact of mutual supply phenomenon, includes:

[0018] Relevant parameters of the checkpoint locations are extracted from the optimized location set and used as input parameters for the loss allocation model. The loss allocation model is updated using a graph neural network to simulate the impact of mutual power supply in a multi-power supply scenario on loss allocation. The loss distribution of each checkpoint location under different operating conditions is calculated using the updated loss allocation model. Based on the loss distribution, a quantitative index of loss deviation is determined, which reflects the fairness and accuracy of loss allocation.

[0019] As a preferred embodiment of the present invention, step S3, generating dynamic scene samples and determining the sensitivity score of the checkpoint location to the balance of interests, includes:

[0020] Based on the quantitative index of loss deviation and the accuracy requirements of measurement in complex networks, a dynamic scenario generation framework is constructed. Multiple sets of dynamic scenario samples are generated using Monte Carlo simulation, including power flow and loss data under different operating conditions. By analyzing the dynamic scenario samples, the degree of influence of each checkpoint location on the balance of interests is calculated. Based on the degree of influence, the sensitivity score of the checkpoint location to the balance of interests is determined.

[0021] As a preferred embodiment of the present invention, step S3, obtaining the recommended configuration for the final checkpoint setting, includes:

[0022] The sensitivity score is obtained and compared with a preset score. If the sensitivity score is lower than the preset score, a suitable alternative is extracted from the optimized location set. The performance of each alternative in terms of benefit balance and loss allocation is analyzed, and the optimal alternative is selected as the recommended configuration for the final gate setting based on the analysis results. The recommended configuration is output as the final setting scheme for the power network gate location.

[0023] As a preferred embodiment of the present invention, step S4, generating a dynamic adaptation scheme and a gate setting analysis report, includes:

[0024] Based on the recommended configuration of the final checkpoint setting, deploy checkpoint locations and collect feedback data on actual operating conditions; by analyzing the feedback data, evaluate the loss distribution and benefit balance effect of the recommended configuration in actual operation; based on the evaluation results, determine whether iterative adjustment of the topology representation is required; if adjustment is required, update the topology representation and regenerate the topology characteristics of the checkpoint candidate locations; through iterative adjustment, generate a dynamic adaptation scheme for the checkpoint locations and a corresponding checkpoint setting analysis report, so that the scheme adapts to changes in actual operating conditions.

[0025] Secondly, the present invention also provides an automatic generation system for power gateway port setting analysis reports, used to implement the above-mentioned method, the system comprising:

[0026] The topology modeling unit is used to generate a topology representation of the power network by collecting wiring diagram data and multi-source configuration information in the power network.

[0027] The circulation analysis unit is used to obtain the property rights division boundary under the participation of multiple entities based on the topology representation, and to determine the distribution characteristics and corresponding quantitative results of potential circulation paths through the analysis of power flow paths under different operating conditions.

[0028] The location optimization unit is used to adjust the priority sequence of the candidate locations of the checkpoints when the quantization result of the distribution feature exceeds a preset threshold, so as to obtain an optimized set of locations.

[0029] The loss assessment unit is used to obtain the input parameters of the loss allocation model through the location set and update the model to simulate the impact of mutual supply phenomenon, and to determine the quantitative index of loss deviation.

[0030] The recommended configuration unit is used to generate dynamic scenario samples and determine the sensitivity score of the checkpoint location to the balance of interests based on the quantitative indicators; if the sensitivity score is lower than the preset score, then alternative solutions are selected from the optimized location set to obtain the recommended configuration of the final checkpoint setting.

[0031] The report generation unit is used to obtain feedback data of the power network under actual operating conditions through the recommended configuration, determine whether the topology representation needs to be iteratively adjusted based on the feedback data, and generate a dynamic adaptation scheme and gateway setting analysis report.

[0032] Thirdly, the present invention also provides a computer-readable storage medium storing instructions that, when executed by a processor, implement the above-described method.

[0033] The beneficial effects of this invention are as follows:

[0034] This invention constructs a topological representation of the power network by collecting wiring diagram data and multi-source configuration information, enabling a unified description of the physical connections and multi-entity participation characteristics of the power grid. Based on this, it identifies property rights boundaries and power interaction areas. Combined with power flow path analysis under different operating conditions, it can systematically reveal the distribution characteristics and quantitative results of potential circulating current paths. When the circulating current distribution characteristics exceed a preset threshold, an optimized set of locations is formed by adjusting the priority sequence of candidate gate locations. A loss allocation model is introduced to simulate the loss distribution under multi-source interconnection conditions. Through model updates and quantitative calculations, a loss deviation index reflecting the rationality of gate location settings can be obtained, allowing for a direct depiction of the impact of gate location on the fairness of loss allocation. Subsequently, dynamic scenario samples are generated based on the loss deviation quantitative index, and different gate locations are analyzed under multiple operating conditions. A comprehensive assessment of the impact on balancing the interests of multiple stakeholders is conducted to obtain a sensitivity score. By comparing the sensitivity score with the preset score, the stability and robustness of the gate setting scheme are judged, and the recommended configuration of the final gate setting is selected accordingly, so that the selection result takes into account both the rationality of loss allocation and the stability of interest balance. Combined with feedback data collected under actual operating conditions, the gate setting scheme is verified and iteratively adjusted. The feedback data drives the dynamic update of the topology and the re-evaluation of gate candidate positions, forming a gate setting scheme that can adaptively adjust with changes in operating conditions, and simultaneously generating a structured gate setting analysis report. Through the cooperation of the above technical solutions, not only can the accuracy and efficiency of gate setting analysis be improved, but the automatic generation and continuous updating of analysis reports can also be achieved, which is applicable to complex power grid scenarios with multiple power sources and multiple stakeholders. Attached Figure Description

[0035] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0036] Figure 1 This is a flowchart of a method for automatically generating analysis reports for setting up an electrical gateway port, as described in this embodiment.

[0037] Figure 2 This is a heat map showing the distribution of the circulation path in the embodiment;

[0038] Figure 3 This is a schematic diagram of the circulating frequency analysis in the embodiment;

[0039] Figure 4The embodiment provides a system structure diagram for automatically generating analysis reports on a power gateway port setting. Detailed Implementation

[0040] This invention provides a method, system, and storage medium for automatically generating an analysis report of a power gateway interface settings. The terms "first," "second," "third," "fourth," etc. (if present) in the specification, claims, and accompanying drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in a sequence other than that illustrated or described herein. Furthermore, the terms "comprising" or "having" and any variations thereof are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or device that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or devices.

[0041] For ease of understanding, the specific process of the embodiments of the present invention will be described below, such as... Figure 1 As shown in the figure, an embodiment of the present invention provides a method for automatically generating an analysis report of a power gateway port setting, comprising:

[0042] Step S1: Generate a topology representation of the power network by collecting wiring diagram data and multi-source configuration information in the power network; based on the topology representation, obtain the property rights division boundary under the participation of multiple entities, and determine the distribution characteristics and corresponding quantitative results of potential circulation paths by analyzing power flow paths under different operating conditions.

[0043] In step S1, generating the topological structure representation includes:

[0044] Obtain wiring diagram data from the power network and combine it with multi-power source configuration information to construct an initial network topology model; use a graph neural network to automatically parse the wiring diagram data and extract the connection relationships between nodes and edges; based on the connection relationships, generate a topology representation of candidate gateway locations, the topology representation including power transmission paths between nodes and power source distribution characteristics.

[0045] Specifically, in this embodiment, in order to realize the automatic generation of the power gateway setting analysis report, it is necessary to uniformly model and structurally express the structural state of the power network, the characteristics of multiple power source access, and the power transmission relationship between nodes, so as to provide a reliable data foundation for subsequent property boundary identification, circulation path analysis, and gateway location optimization. Therefore, this embodiment generates a topology representation that reflects the actual operating state of the power network by collecting wiring diagram data in the power network and combining it with multiple power source configuration information.

[0046] Specifically, the wiring diagram data of the power network is first obtained through data acquisition equipment installed in the power system. The wiring diagram data includes at least the connection relationships between substations, generators, transmission and distribution lines, and load nodes, as well as the corresponding physical topology information. On this basis, multi-power source configuration information is obtained simultaneously. The multi-power source configuration information includes, but is not limited to, the access location, installed capacity, and grid connection method of power sources such as distributed photovoltaic power sources and wind power generation units. By integrating the wiring diagram data and multi-power source configuration information, an initial network topology model is constructed, in which nodes represent power equipment or power source access points, and edges represent power connection lines between nodes, thereby forming an initial graph model that can describe the overall structure of the power network.

[0047] After constructing the initial network topology model, to avoid the error problem of traditional manual wiring diagram analysis in complex multi-power source scenarios, a graph neural network is introduced to automatically parse the wiring diagram data. Specifically, the initial network topology model is used as the input graph of the graph neural network, where each node is assigned an initial feature vector. The initial feature vector includes at least the node's spatial coordinates, equipment type, and whether it is a power source node. Subsequently, graph convolution is performed on the input graph, and the neighborhood information of the nodes is aggregated using the message passing mechanism of the graph neural network to calculate the node's update vector. By performing multi-level iterative calculations on the update vector, the direct connection relationships between nodes and the indirect path relationships formed through multi-level lines are gradually extracted, enabling the graph neural network to simultaneously learn the local connectivity characteristics and overall topology characteristics of the power network.

[0048] The aforementioned gateway candidate locations refer to a class of node locations identified from the power network that have the potential to serve as gateway locations. These locations are typically situated on the main power transmission path between the power source and the load, or at the power interaction boundary between different power entities. Under conditions of multiple power source access and multiple operating conditions, they have a significant impact on power flow, power distribution, and the formation of potential circulating currents. In the topology, they are characterized by high connectivity, large transmission weight, or key nodes crossing property boundaries. They are used as the analysis objects and optimization candidate sets for subsequent circulating current risk assessment, loss allocation analysis, benefit balance calculation, and automatic generation of gateway location setting analysis reports. After obtaining the connection relationships between nodes and edges, a topological representation of the gateway candidate locations is further generated based on these connections. Specifically, path analysis is performed on the extracted connection relationships to calculate the shortest path matrix between any two nodes in the network. This shortest path matrix is ​​used to record the minimum number of connecting edges between nodes, thereby identifying the main path for power transmission from the power source node to the load node. Simultaneously, combined with multi-power source configuration information, statistical analysis is performed on the influence range of each power source node. For example, by calculating the subgraph density centered on power supply nodes, the coverage of different power sources in the network can be determined. Subsequently, the power transmission path information and power distribution characteristics are fused to construct a topology representation. This topology representation can be expressed in the form of an embedded matrix, where the rows of the matrix correspond to nodes in the power network, and the columns correspond to parameters characterizing power transmission and supply characteristics, such as the path length from the node to the power source and the power source's influence weight. Based on this topology representation, key nodes in the power network are identified and labeled. Specifically, by scanning and analyzing the embedded matrix, nodes with a degree higher than the network average are identified as key nodes. The node degree is the number of edges directly connected to the node. Combined with its connection relationship in the topology, its current connection status is labeled, such as whether the line is in an open or closed state. At the same time, combined with the multi-power source operation characteristics, the topology characteristics of key nodes under the influence of multiple power sources are labeled, such as the node's sensitivity to voltage fluctuations or the power transmission weight it bears under different power supply paths, thereby providing refined data for subsequent circulation path analysis and gateway location assessment.

[0049] After generating and labeling the aforementioned topology representation, the topology representation is stored as basic data. Specifically, the topology representation is converted into a structured database format for storage. Through matching and verification with actual operating data, it is ensured that the topology representation can truly reflect the actual layout and operating status of the power network. It is used to obtain the property rights demarcation boundary under multi-stakeholder participation, support Monte Carlo simulation to determine the distribution characteristics of potential circulating current paths, and ultimately serve the optimization analysis of power gateway locations and the automatic generation of gateway setting analysis reports.

[0050] Further, in step S1, the distribution characteristics of potential circulation paths are determined, including:

[0051] Based on the topology representation, the property rights demarcation boundary under multi-entity participation is extracted to determine the power interaction area between each entity; Monte Carlo simulation is used to simulate and sample power flow paths under different operating conditions to generate multiple sets of flow path data; by analyzing the flow path data, the distribution characteristics of potential circulation paths are identified, including the frequency of occurrence and the range of influence of circulation paths; based on the distribution characteristics, the potential impact of circulation paths on power network stability is quantified and used as the quantification result of the distribution characteristics.

[0052] Specifically, in this embodiment, by parsing the topology representation, the property rights division boundaries under the participation of multiple entities are automatically identified, thereby clarifying the physical and functional boundaries between different power entities. The aforementioned multiple entities include, but are not limited to, power generation companies, power transmission and distribution companies, and park-level power users. Through comprehensive analysis of node attributes, branch ownership information, and connection relationships in the topology, boundary lines reflecting property rights boundaries are extracted, and areas where power interaction exists between entities are delineated accordingly. These power interaction areas limit the possible range of cross-entity power flow at the topology level, enabling subsequent analysis of circulation paths to focus on areas where property rights interaction actually exists, avoiding interference from irrelevant paths in the analysis results.

[0053] After determining the property rights boundaries and power interaction areas, based on the aforementioned topology representation and property rights division results, a Monte Carlo simulation mechanism is introduced to sample power flow paths under different operating conditions. Specifically, based on the topology representation and property rights boundaries, parameter ranges for different operating conditions are defined. These operating condition parameter ranges include at least the load level variation range, power output fluctuation characteristics, and multi-power source parallel operation conditions. Each parameter can be set as a probability distribution based on historical statistical data or planning scenarios to reflect uncertainties in power network operation. For example, power output can be modeled as a random distribution with rated power as the mean and a preset standard deviation. Subsequently, the operating conditions are simulated using the Monte Carlo simulation method, and multiple random samples are taken. In each sample, power flow calculation is called to obtain the specific path of power flowing from the power source node through the grid topology to the load node, thereby generating corresponding power flow path data. By repeatedly sampling different operating conditions, multiple sets of flow path data covering peak and valley load variations and multi-power source combinations are obtained.

[0054] After obtaining multiple sets of power flow path data corresponding to multiple sets of operating condition parameters, statistical analysis is performed on the path data to identify the distribution characteristics of potential circulation paths. Specifically, the sampled flow paths are first merged and standardized, and paths with the same or highly similar node sequences and branch structures are identified as the same potential circulation path. Subsequently, the frequency of occurrence of each potential circulation path is calculated by statistically analyzing the number of times it appears in all samplings. This frequency is used to characterize the probability of the circulation path occurring under multiple operating conditions. At the same time, the circulation path is quantified by tracking the number of nodes, the number of branches, and the power interaction area it crosses. The impact range is determined to reflect the degree of impact on the power network structure and operating area once it occurs. By weighting and summing the frequency of occurrence and the impact range, and using the weighted result as the quantitative result of the above distribution characteristics, a complete description of the distribution characteristics of potential circulating current paths is formed. The above technical solution, through the above continuous data processing and analysis process, realizes a complete closed loop from topology analysis, random sampling of operating conditions, statistical identification of circulating current paths to stability quantitative assessment. This enables potential circulating current risks to serve the automatic generation of power gateway setting analysis reports in a structured and quantifiable manner, thereby providing a clear, reliable and directly usable technical foundation for subsequent gateway location optimization.

[0055] Step S2: If the quantification result of the distribution feature exceeds the preset threshold, the priority sequence of the gate candidate positions is adjusted to obtain the optimized position set; through the position set, the input parameters of the loss allocation model are obtained and the model is updated to simulate the impact of mutual supply phenomenon and determine the quantification index of loss deviation.

[0056] In step S2, the optimized set of locations is obtained, including:

[0057] The quantization result of the distribution feature is obtained and compared with a preset threshold. If the quantization result exceeds the preset threshold, the influence of the candidate gate positions on the circulation path is analyzed for multi-power source scenarios. Based on the influence, the priority sequence of the candidate gate positions is adjusted to prioritize the selection of positions with less influence on the circulation path. The candidate gate positions that meet the conditions are selected through the priority sequence to generate an optimized set of positions.

[0058] Specifically, in this embodiment, after the identification and quantification of the distribution characteristics of potential circulation paths are completed, the quantification results corresponding to the above distribution characteristics are first obtained, and the quantification results are compared with a preset threshold to determine whether the circulation risk under the current power grid operation state is within an acceptable range. When the quantification result value is higher than the preset threshold, it is determined that the current set of candidate checkpoint locations has a potential high circulation risk under this operating scenario, thus entering the checkpoint candidate location impact analysis stage. Figure 2 As shown, the above comparison process enables a preliminary assessment of the risk of power grid circulating currents.

[0059] After confirming that the distribution characteristics exceed a preset threshold, and considering the multi-source configuration characteristics, a quantitative analysis is performed on the influence of each candidate location on the formation of the circulation path. Based on the acquired multi-source configuration parameters and topology representation, Monte Carlo simulation is used to simulate various operating conditions and repeatedly sample randomly. Power flow calculations are performed in each sampling to generate corresponding power flow path data. For each candidate location, such as... Figure 3 As shown, the frequency with which a candidate location triggers or participates in the formation of circulation paths in all sampling paths is statistically analyzed, and this frequency is defined as one of the indicators of influence, used to measure the direct influence level of a candidate location on the formation of circulation paths.

[0060] Meanwhile, to avoid the limited analytical dimension caused by relying solely on statistical frequency, and to further characterize the structural impact of candidate gateway locations on the overall circulation distribution, a graph neural network is introduced to perform deep processing on the power grid topology representation. By mapping candidate gateway locations to graph nodes and power lines to graph edges, and using voltage level, connectivity, and multi-source access information as feature inputs for nodes and edges, the message passing and node embedding mechanisms of the graph neural network are utilized to learn the structural role of different candidate gateway locations in the overall topology. This quantifies the contribution of each candidate location to the circulation path distribution. This contribution, together with the aforementioned circulation occurrence frequency, constitutes a comprehensive evaluation index of the influence of candidate gateway locations, enabling the analysis results to reflect not only the probability of occurrence at the statistical level but also the intensity of influence at the topological level.

[0061] After obtaining the impact index corresponding to each candidate checkpoint location, the priority sequence of the candidate checkpoint locations is adjusted according to the aforementioned impact index. Specifically, all candidate checkpoint locations are sorted in ascending order of their impact on the circulation path, and the ranking weight of candidate locations with less impact on the circulation is increased in the generated priority sequence, thus forming an optimized priority sequence of candidate checkpoint locations. This priority sequence directly reflects the relative rationality of different candidate checkpoint locations under the current power grid operation scenario. Subsequently, based on the above priority sequence, candidate checkpoint locations are screened according to preset screening conditions, such as selecting candidate locations whose priority meets specific risk threshold requirements. The optimized set of gate locations is formed and serves as an important component of the power grid gate setting analysis results. On the one hand, it guides actual gate setting decisions; on the other hand, it serves as the input basis for subsequent loss allocation models and mutual power supply simulation models to verify the performance of the gate location set in terms of loss allocation fairness and rationality. Through this verification process, not only is a technical closed-loop verification of the gate location optimization results achieved, but also an analytical basis with logical consistency and data traceability is provided for the automatic generation of the power grid gate setting analysis report. This ensures that the final analysis report can truly reflect the comprehensive impact of gate setting adjustments on the stability and economy of power grid operation.

[0062] Further, in step S2, the input parameters of the loss allocation model are obtained and the model is updated to simulate the impact of mutual supply phenomenon, including:

[0063] Relevant parameters of the gate locations are extracted from the optimized location set and used as input parameters for the loss allocation model. The loss allocation model is updated using a graph neural network to simulate the impact of mutual power supply in a multi-power supply scenario on loss allocation. The loss distribution of each gate location under different operating conditions is calculated using the updated model. Based on the loss distribution, a quantitative index of loss deviation is determined, which reflects the fairness and accuracy of loss allocation.

[0064] Specifically, in this embodiment, in order to realize the automatic generation of the power gateway port setting analysis report and ensure that the analysis results can truly reflect the loss distribution characteristics under the multi-power supply operation conditions, after completing the optimization of the candidate port positions and obtaining the optimized position set, the system further constructs and dynamically updates the parameters of the loss distribution model based on the above position set, so as to simulate the impact of the mutual supply phenomenon under the multi-power supply scenario on the loss distribution results.

[0065] In the implementation process, relevant parameters of the checkpoint locations are first extracted from the optimized location set and used as input parameters for the loss allocation model. Each candidate checkpoint location in the location set corresponds to a node in the power network topology. Therefore, during parameter extraction, the spatial coordinate information of each candidate checkpoint location in the power network topology and its directly connected electrical connection parameters are obtained by traversing the location set. The coordinate information is used to characterize the specific node location of the checkpoint in the wiring diagram, while the connection parameters are used to describe the connection strength, line weight, or power transmission characteristics between the checkpoint and adjacent power supply nodes and load nodes. Subsequently, the extracted multidimensional parameters are uniformly encoded and normalized to convert them into vectorized data, thereby forming a parameter set that matches the input interface of the loss allocation model, providing structured input for subsequent model calculations.

[0066] After the input parameters are constructed, the loss allocation model is updated through a graph neural network to simulate the dynamic impact of mutual power supply on loss allocation in a multi-power source scenario. In this process, the input graph structure of the graph neural network is first constructed based on the above parameters. The nodes in the graph represent the gateway nodes and power source nodes in the power network, the edges in the graph represent the power flow path or electrical connection relationship, and the node features are composed of the coordinate parameters, connection relationship parameters and power source characteristic parameters extracted from the location set. Through the construction of this graph structure, the physical topology and operating characteristics of the power network are uniformly expressed in the form of graph data. Subsequently, the graph neural network performs multi-layer convolution operations on the input graph. Through the message passing mechanism between nodes, it continuously updates the embedding representation of each node. In the specific implementation, each convolutional layer updates the feature vector of the target node by aggregating the feature information of neighboring nodes. For example, in the initial convolutional layer, each gateway node aggregates the power supply features and connection weight information of its neighboring power supply nodes to obtain a node representation that can reflect the local mutual supply relationship. As the number of convolutional layers increases, the node embedding gradually integrates a wider range of topological information, enabling the model to capture the mutual supply phenomenon formed by the cross flow of power between different paths under a multi-power supply configuration.

[0067] After obtaining the updated node embedding representation, the output layer of the graph neural network explicitly models the cross-supply phenomenon. By jointly calculating the node embedding and edge weights, it simulates the loss path changes caused by the cross-supply of power from one power source to another power source or load node under multi-power source operation conditions. Based on this, the adjusted loss weight distribution is calculated. The loss weights reflect the contribution of different paths to the overall loss under cross-supply. Subsequently, the simulation results are fed back into the loss allocation model, and the model parameters are updated through the backpropagation mechanism, enabling the loss allocation model to adaptively reflect the loss change law under multi-power source cross-supply conditions. In a further implementation, for complex power networks containing multiple power sources such as distributed photovoltaics and wind power, an attention mechanism can be introduced into the graph neural network structure. By assigning dynamic weights to different power flow edges, the model prioritizes critical paths that contribute significantly to losses when calculating the impact of cross-supply. For example, by calculating attention scores based on power source capacity differences or output fluctuations, cross-supply paths between high-capacity power sources receive higher weights in the model, thereby improving the accuracy of loss simulation, reducing loss deviations caused by cross-supply, and providing a more reliable data foundation for subsequent analysis.

[0068] After updating the loss allocation model, the updated model is used to calculate the loss distribution at each checkpoint location under different operating conditions. Specifically, operating condition data reflecting different load levels, power output fluctuations, and changes in operating modes are input into the updated loss allocation model. The model inferences to obtain the loss values ​​corresponding to each checkpoint location and outputs them in a distribution form, thus forming a loss distribution result that reflects the trend of loss changes under different operating conditions. Based on the above loss distribution results, a quantitative analysis of loss deviation is performed to determine the fairness and accuracy of the current checkpoint location setting in loss allocation. The training data for the above loss allocation model includes historical topology data, power output data and load change data obtained from multi-power supply configuration schemes and operating condition sampling, and historical actual line loss data. By calculating the statistical characteristics of the loss distribution, statistical indicators including mean and variance are obtained, and based on the above statistical indicators... A quantitative index for loss deviation is constructed. For example, by calculating the ratio of the variance to the mean of the loss distribution, a deviation ratio reflecting the dispersion of loss allocation is obtained. The lower the deviation ratio, the more balanced and accurate the loss allocation. The quantitative index is also compared with a preset threshold to determine whether the current loss allocation result meets the requirements of accuracy and fairness, thus providing a quantitative basis for the rationality of the gateway setting. Finally, the quantitative index of loss deviation is stored as structured data in a database, serving as important reference data for subsequent benefit balance analysis and automatic generation of power gateway gateway setting analysis reports. Through the above technical solution, not only is the data connection between the gateway location optimization results and the loss allocation analysis results realized, but the analysis report also ensures that it has a clear data source, complete calculation logic, and traceable analysis basis during the generation process, thereby effectively supporting the automatic generation and dynamic updating of power gateway gateway setting analysis reports in complex multi-power source operation scenarios.

[0069] Step S3: Based on the quantitative indicators, generate dynamic scene samples and determine the sensitivity score of the checkpoint location to the balance of interests; if the sensitivity score is lower than the preset threshold, select alternative solutions from the optimized location set to obtain the recommended configuration for the final checkpoint setting;

[0070] In step S3, dynamic scene samples are generated and the sensitivity score of the checkpoint location to the balance of interests is determined, including:

[0071] Based on the quantitative index of loss deviation and the accuracy requirements of measurement in complex networks, a dynamic scenario generation framework is constructed. Multiple sets of dynamic scenario samples are generated using Monte Carlo simulation, including power flow and loss data under different operating conditions. By analyzing the dynamic scenario samples, the degree of influence of each checkpoint location on the balance of interests is calculated. Based on the degree of influence, the sensitivity score of the checkpoint location to the balance of interests is determined.

[0072] Specifically, in one implementation, in order to enable the power gateway port setting analysis report to fully reflect the impact of the port location on the balance of interests of various stakeholders under multi-power supply operation conditions, after obtaining the quantitative index of loss deviation, the system further constructs a dynamic scenario generation and analysis mechanism based on the quantitative index to achieve a quantitative assessment of the port location sensitivity.

[0073] Specifically, the system first constructs a dynamic scenario generation framework for subsequent simulation analysis based on the loss deviation quantification index output by the aforementioned loss allocation model, combined with the metering accuracy requirements in complex power networks. The loss deviation quantification index characterizes the deviation of the actual loss allocation result from the ideal fair allocation state under the current threshold setting scheme, while the metering accuracy requirement limits the allowable measurement error range during the analysis process, for example, limiting the loss metering error to no more than a preset proportion. By unifying the quantification index and the metering accuracy requirement, the system forms a structured dynamic scenario generation framework. This framework clearly defines the input parameter types, value ranges, and data structures of the output samples required for simulation, thus providing a consistent data foundation for subsequent multi-scenario simulations.

[0074] After constructing the dynamic scenario generation framework, the system uses Monte Carlo simulation to generate multiple sets of dynamic scenario samples to characterize the randomness and uncertainty of the power network under different operating conditions. In this process, the system first obtains simulation input parameters from the dynamic scenario generation framework, including loss deviation quantification indicators, metering accuracy constraints, and variables reflecting the grid's operating state such as load levels and power source output configurations. Then, the Monte Carlo simulation process is initialized. By setting the number of simulation iterations and randomly selecting operating condition variables in each iteration, the operating state of the power network is sampled multiple times. In each sampling process, the system calculates the power flow path based on the aforementioned topology representation and estimates the corresponding line loss value by combining the power balance relationship, thus obtaining a single simulation result containing power flow data and loss data. By summarizing all iteration results, multiple sets of dynamic scenario samples are formed. Each set of samples corresponds to a possible operating condition and fully describes the power flow state and loss distribution characteristics under that condition. With the help of this Monte Carlo simulation process, the system can statistically cover the possible operating states under multi-source scenarios, thereby improving the robustness and reliability of subsequent analysis results.

[0075] After obtaining the dynamic scenario samples, the system further analyzes and calculates the impact of each checkpoint location on the balance of interests among multiple stakeholders. Specifically, the system extracts relevant power flow and loss data from the dynamic scenario samples and uses this as the basis for the benefit balance assessment. For each candidate checkpoint location, the system compares the loss allocation results corresponding to that location under different scenarios and calculates the rate of change in losses caused by the change in checkpoint location, thereby quantifying the impact of that location on the loss distribution. Based on this, the system further combines the property rights boundaries of each stakeholder in the power grid to weight the rate of change in losses, reflecting the relative weight of different stakeholders in loss allocation, and thus obtaining an indicator characterizing the impact of the checkpoint location on the degree of disturbance to the balance of interests among multiple stakeholders. By performing the above calculations on all candidate checkpoint locations, the system forms a set of distribution characteristic data reflecting the degree of influence of each checkpoint location.

[0076] Subsequently, the system determines the sensitivity score of the checkpoint location to the balance of interests based on the impact data. In this process, the system first normalizes the impact indicators corresponding to each checkpoint location, mapping them to a unified numerical range to eliminate the influence of different units and value ranges on subsequent comparisons. Then, it calculates the sensitivity score for each checkpoint location by weighted summation, which characterizes the sensitivity of that checkpoint location to the balance of interests among multiple stakeholders under different operational scenarios.

[0077] After calculating the sensitivity score, the system compares the score with a preset threshold to provide a basis for decision-making in subsequent gate location selection and analysis report generation. When the sensitivity score of a gate location is lower than the preset threshold, the system classifies it as a low-sensitivity location with minimal impact on the balance of interests and prioritizes it in the subsequent gate setting scheme selection process; conversely, it is classified as a high-sensitivity location, and its potential risks need to be highlighted in the report. Through this comparison mechanism, the system can automatically incorporate the analysis conclusions of the benefit balance dimension when generating power grid gate setting analysis reports. This ensures that the reports not only reflect the impact of gate locations on power grid operation and loss distribution but also take into account fairness and stability among multiple stakeholders, thereby achieving automatic generation and intelligent decision support for analysis reports in complex multi-source power grid scenarios.

[0078] Further, in step S3, the recommended configuration for the final checkpoint setting is obtained, including:

[0079] The sensitivity score is obtained and compared with a preset score. If the sensitivity score is lower than the preset score, a suitable alternative is extracted from the optimized location set. The performance of each alternative in terms of benefit balance and loss allocation is analyzed, and the optimal alternative is selected as the recommended configuration for the final gate setting based on the analysis results. The recommended configuration is output as the final setting scheme for the power network gate location.

[0080] Specifically, in one implementation, in order to automatically generate and reasonably recommend the final gateway configuration scheme in the power gateway gateway setting analysis report, after completing the sensitivity score calculation of the gateway location to the balance of interests, the candidate gateway locations are further screened and comprehensively evaluated based on the above sensitivity score, so as to determine the final recommended gateway setting configuration.

[0081] During implementation, the sensitivity scores of the checkpoint locations obtained through the Monte Carlo simulation and dynamic scenario analysis are first acquired. The corresponding score values ​​are then read from the storage module and compared with the preset scores to determine whether the current set of candidate checkpoint locations meets the stability requirements of the interest balance. When the sensitivity score is lower than the preset score, it indicates that the corresponding checkpoint location has a smaller disturbance to the interest balance of multiple entities in a multi-power source operation scenario. It is necessary to further select alternative solutions from the optimized set of locations to form the final recommended configuration. Specifically, the optimized set of locations is traversed, and based on the topology representation and circulation path analysis results, checkpoint candidate locations that meet the preset conditions are selected as alternative solutions. In the specific selection process, the distribution characteristics of the potential circulation paths corresponding to each checkpoint candidate location in the topology representation are examined. Only those candidate locations whose circulation path distribution intensity does not exceed the preset threshold are retained. This ensures that the checkpoint location has a smaller impact on the interest balance while further reducing circulation risk and laying a reliable foundation for subsequent analysis.

[0082] After obtaining the set of alternative solutions, an in-depth analysis is conducted on the performance of each alternative in terms of benefit balance and loss allocation. Specifically, the topological structure representation of the corresponding key locations is obtained from each alternative solution. This topological structure representation is generated automatically by parsing the power network wiring diagram using a graph neural network. Through node embedding and message passing mechanisms, the connection relationships and operating characteristics between substations, lines, and multiple power source access nodes are accurately depicted. Based on this, combined with the property rights demarcation boundary under multi-stakeholder participation, the Monte Carlo simulation method is used to sample power flow paths under different operating conditions. Through statistical analysis of multiple random sampling results, the distribution characteristics and occurrence probability of potential circulating current paths under each alternative solution are estimated, and the circulating current intensity is further quantified. The analysis results show that the circulation path distribution characteristics of a certain alternative scheme exceed a preset threshold. For multi-source operation scenarios, the priority of the candidate locations in the alternative scheme is dynamically adjusted. This is achieved by reducing the ranking weight of high-circulation-risk locations and reordering the location set to reduce the adverse impact of mutual power supply on grid operation. Simultaneously, the adjusted location set is used to obtain the input parameters of the loss allocation model, and a graph neural network is used to update the model. Iterative training simulates the loss distribution changes under multi-source mutual power supply conditions, thereby calculating the quantitative index of loss deviation for each alternative scheme. This quantitative index reflects the performance of different schemes in terms of loss allocation fairness, providing a direct basis for benefit balance analysis.

[0083] Furthermore, to address the accuracy requirements of metering in complex power networks, Monte Carlo simulation is used to generate multiple sets of dynamic scenario samples. Based on these samples, the benefit balance sensitivity scores of each alternative scheme are recalculated. By averaging or weighted summing the results of multiple sets of samples, a sensitivity score reflecting the comprehensive performance of each scheme under various operating conditions is obtained, thus avoiding biases caused by single-condition analysis. After completing the above analysis, a comprehensive evaluation of the benefit balance performance and loss allocation effect of each alternative scheme is conducted. This involves standardizing the aforementioned loss deviation quantification index and the sensitivity score of the threshold position to the benefit balance, converting evaluation parameters of different dimensions to the same numerical range. Based on pre-set weighting coefficients, the standardized loss deviation quantification index and the sensitivity score are weighted and fused to obtain a comprehensive score characterizing the overall performance of each alternative scheme. The weights are... The coefficients are used to reflect the relative importance of loss allocation fairness and benefit balance stability in the overall evaluation, and the scheme with the lowest comprehensive score is selected as the recommended configuration for the final gate setting. Finally, the above recommended configuration is output as the final setting scheme for the power network gate location, and the corresponding gate location parameters can be directly transmitted to the network control system or report generation module to support the automatic generation of power network gate setting analysis reports and the dynamic adaptive adjustment of gate setting schemes. Through the above technical solution, the technical effect of automatically screening and optimizing gate setting schemes based on sensitivity scores and multi-dimensional analysis results is achieved in complex multi-source operation scenarios. This makes the generated power network gate setting analysis report not only have clear data basis and complete calculation logic, but also provide decision support in terms of benefit balance, loss allocation and operational stability.

[0084] Step S4: Obtain feedback data of the power network under actual operating conditions through the recommended configuration; determine whether iterative adjustment of the topology representation is needed based on the feedback data; and generate a dynamic adaptation scheme and gateway setting analysis report; specifically including:

[0085] Based on the recommended configuration of the final checkpoint setting, deploy checkpoint locations and collect feedback data on actual operating conditions; by analyzing the feedback data, evaluate the loss distribution and benefit balance effect of the recommended configuration in actual operation; based on the evaluation results, determine whether iterative adjustment of the topology representation is necessary; if adjustment is needed, update the topology representation and regenerate the topology characteristics of the checkpoint candidate locations; through iterative adjustment, generate a dynamic adaptation scheme for the checkpoint locations and a corresponding checkpoint setting analysis report, so that the scheme adapts to changes in actual operating conditions.

[0086] Specifically, after finalizing the recommended configuration for the gate settings, the gate setting scheme is verified and dynamically optimized in a closed loop by deploying the recommended configuration and combining it with actual operational feedback. This generates a dynamic adaptation scheme that reflects the actual operating status of the power network and a corresponding gate setting analysis report.

[0087] During implementation, based on the recommended configuration of the final gate settings, the actual deployment of gate locations in the power network is completed. This deployment includes setting metering points or gate nodes at corresponding locations on the wiring diagram of the multi-source configuration, enabling real-time metering of power flow through these gate locations. After deployment, feedback data on the power network under actual operating conditions is continuously collected from these gate locations. This feedback data includes at least real-time operating indicators such as power flow path information, loss values ​​corresponding to each gate location, and the distribution of potential circulating current paths, thus forming a raw data set reflecting the current operating status. After acquiring the feedback data, further analysis is performed to evaluate the loss distribution and benefit balance effects of the recommended configuration in actual operation, specifically through Monte Carlo methods. The Carlow simulation method performs multiple random samplings of the feedback data, constructs various possible operating condition samples, and conducts statistical analysis on different power flow paths to generate input parameters for the loss allocation model calculation. During this process, it estimates the average loss deviation level under actual operating conditions and analyzes the power exchange phenomenon among multiple entities to determine whether there is a significant imbalance in loss allocation. To improve the evaluation accuracy, a graph neural network is used to update the loss allocation model. Node embedding and message passing mechanisms characterize the topological relationships and mutual supply effects between key points, thereby obtaining more accurate quantitative indicators of loss allocation. Based on the changing trend of the aforementioned loss deviation, a sensitivity score for the balance of interests is calculated to characterize the degree of impact of changes in key point locations on the overall stability of the balance of interests.

[0088] After evaluating the effects of loss allocation and benefit balancing, the evaluation results determine whether iterative adjustments to the power network topology representation are necessary. Specifically, if the evaluation results show that the quantification index of loss deviation or the sensitivity score exceeds the corresponding set threshold, it is determined that the current gate setting scheme can no longer meet the requirements of loss allocation fairness or benefit balancing stability under actual operating conditions, thus triggering the topology representation update process. Otherwise, the current gate configuration remains unchanged. When it is determined that adjustments are needed, a graph neural network is used to update the original topology representation based on the feedback data. By aggregating neighbor node information and relearning node embedding vectors, new topology features and property rights boundaries generated by changes in multi-source access or operating conditions are captured. Based on the updated topology representation, the topology characteristics of gate candidate locations are regenerated, and Monte Carlo simulation is further combined to sample power flow paths under different operating conditions to obtain the latest distribution characteristics and occurrence probabilities of potential circulation paths. Based on this, the priority sequence of gate candidate locations is dynamically adjusted, and the location set is optimized to generate new topology characteristics that can better adapt to the current operating state.

[0089] Through the aforementioned topology update and candidate location reconstruction process, the adjustment results are continuously accumulated and adjusted in multiple iterations, gradually forming a dynamic adaptation scheme for the gate location. This dynamic adaptation scheme can continuously optimize the gate setting according to changes in load, power output fluctuations, and seasonal operating characteristics, ensuring that the gate location maintains low loss deviation and high benefit balance stability during long-term operation. The dynamic adaptation scheme can also be verified by simulating new operating conditions, such as recalculating the sensitivity score corresponding to the scheme, to confirm its adaptability to future operating conditions. Based on this, the deployment process, feedback data analysis results, topology iterative adjustment process, and the final dynamic adaptation scheme are integrated. The system performs structured data processing and automatically generates a gateway configuration analysis report. This report includes at least: a description of the current gateway configuration scheme and its deployment status; loss allocation assessment results under actual operating conditions; conclusions of the benefit balance sensitivity analysis; the criteria for determining whether to trigger topology iterative adjustments; and the final dynamically adaptive gateway configuration scheme and its adaptability verification results. Through this technical solution, a complete closed loop of gateway configuration scheme recommendation, verification, and dynamic optimization is achieved. This ensures that the generated power grid gateway configuration analysis report not only reflects static calculation results but also truly reflects the dynamic changes of the power network in actual operation, thereby providing continuous and reliable technical support for power grid operation decisions.

[0090] This invention also provides an automatic generation system for power gateway port setting analysis reports, used to implement the above-mentioned method, such as... Figure 4 As shown, the system includes:

[0091] The topology modeling unit is used to generate a topology representation of the power network by collecting wiring diagram data and multi-source configuration information in the power network.

[0092] The circulation analysis unit is used to obtain the property rights division boundary under the participation of multiple entities based on the topology representation, and to determine the distribution characteristics and corresponding quantitative results of potential circulation paths through the analysis of power flow paths under different operating conditions.

[0093] The location optimization unit is used to adjust the priority sequence of the candidate locations of the checkpoints when the quantization result of the distribution feature exceeds a preset threshold, so as to obtain an optimized set of locations.

[0094] The loss assessment unit is used to obtain the input parameters of the loss allocation model through the location set and update the model to simulate the impact of mutual supply phenomenon, and to determine the quantitative index of loss deviation.

[0095] The recommended configuration unit is used to generate dynamic scenario samples and determine the sensitivity score of the checkpoint location to the balance of interests based on the quantitative indicators; if the sensitivity score is lower than the preset score, then alternative solutions are selected from the optimized location set to obtain the recommended configuration of the final checkpoint setting.

[0096] The report generation unit is used to obtain feedback data of the power network under actual operating conditions through the recommended configuration, determine whether the topology representation needs to be iteratively adjusted based on the feedback data, and generate a dynamic adaptation scheme and gateway setting analysis report.

[0097] The present invention also provides a computer-readable storage medium storing instructions that, when executed by a processor, implement the above-described method.

[0098] In summary, this invention constructs a topological representation of the power network by collecting wiring diagram data and multi-source configuration information, enabling a unified description of the physical connections and multi-entity participation characteristics of the power grid. Based on this, it identifies property rights boundaries and power interaction areas. Combined with power flow path analysis under different operating conditions, it can systematically reveal the distribution characteristics and quantitative results of potential circulating current paths. When the circulating current distribution characteristics exceed a preset threshold, an optimized set of locations is formed by adjusting the priority sequence of candidate locations. A loss allocation model is introduced to simulate the loss distribution under multi-source interconnection conditions. Through model updates and quantitative calculations, a loss deviation index reflecting the rationality of the location can be obtained, allowing for a direct depiction of the impact of location on the fairness of loss allocation. Subsequently, dynamic scenario samples are generated based on the loss deviation quantitative index, and the effects of different location locations on multi-source interconnection are analyzed. A comprehensive assessment of the impact on the balance of interests among multiple stakeholders under operating conditions is conducted to obtain a sensitivity score. By comparing the sensitivity score with the preset score, the stability and robustness of the gate setting scheme are judged, and the recommended configuration of the final gate setting is selected accordingly, so that the selection result takes into account both the rationality of loss allocation and the stability of interest balance. Combined with feedback data collected under actual operating conditions, the gate setting scheme is verified and iteratively adjusted. The feedback data drives the dynamic update of the topology and the re-evaluation of gate candidate positions, forming a gate setting scheme that can adaptively adjust with changes in operating conditions, and simultaneously generating a structured gate setting analysis report. Through the cooperation of the above technical solutions, not only can the accuracy and efficiency of gate setting analysis be improved, but the automatic generation and continuous updating of analysis reports can also be achieved, which is applicable to complex power grid scenarios with multiple power sources and multiple stakeholders.

[0099] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0100] If the integrated unit is implemented as 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 technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0101] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for automatically generating an analysis report on gateway port settings, characterized in that, The method includes: Step S1: Generate a topology representation of the power network by collecting wiring diagram data and multi-source configuration information in the power network; based on the topology representation, obtain the property rights division boundary under the participation of multiple entities, and determine the distribution characteristics and corresponding quantitative results of potential circulation paths by analyzing power flow paths under different operating conditions. Step S2: If the quantification result of the distribution feature exceeds the preset threshold, the priority sequence of the gate candidate positions is adjusted to obtain the optimized position set; through the position set, the input parameters of the loss allocation model are obtained and the model is updated to simulate the impact of mutual supply phenomenon and determine the quantification index of loss deviation. Step S3: Based on the quantitative indicators, generate dynamic scene samples and determine the sensitivity score of the checkpoint location to the balance of interests; if the sensitivity score is lower than the preset score, select alternative solutions from the optimized location set to obtain the recommended configuration of the final checkpoint setting; Step S4: Obtain feedback data of the power network under actual operating conditions through the recommended configuration, determine whether it is necessary to iteratively adjust the topology representation based on the feedback data, and generate a dynamic adaptation scheme and gateway setting analysis report; In step S2, obtaining the input parameters of the loss allocation model and updating the model to simulate the impact of mutual power supply includes: extracting relevant parameters of the gate positions from the optimized location set as input parameters of the loss allocation model; updating the loss allocation model using a graph neural network to simulate the impact of mutual power supply in a multi-power supply scenario on loss allocation; calculating the loss distribution of each gate position under different operating conditions using the updated loss allocation model; and determining a quantitative index of loss deviation based on the loss distribution, wherein the quantitative index reflects the fairness and accuracy of loss allocation. In step S3, generating dynamic scenario samples and determining the sensitivity score of the checkpoint location to the balance of interests includes: constructing a dynamic scenario generation framework based on the quantitative index of loss deviation and the measurement accuracy requirements in complex networks; generating multiple sets of dynamic scenario samples using Monte Carlo simulation, wherein the dynamic scenario samples include power flow and loss data under different operating conditions; calculating the degree of influence of each checkpoint location on the balance of interests by analyzing the dynamic scenario samples; and determining the sensitivity score of the checkpoint location to the balance of interests based on the degree of influence.

2. The method as described in claim 1, characterized in that, In step S1, the topological representation is generated, including: Obtain wiring diagram data from the power network and combine it with multi-power source configuration information to construct an initial network topology model; use a graph neural network to automatically parse the wiring diagram data and extract the connection relationships between nodes and edges; based on the connection relationships, generate a topology representation of candidate gateway locations, the topology representation including power transmission paths between nodes and power source distribution characteristics.

3. The method as described in claim 2, characterized in that, In step S1, the distribution characteristics of potential circulation paths are determined, including: Based on the topology representation, the property rights demarcation boundary under multi-entity participation is extracted to determine the power interaction area between each entity; Monte Carlo simulation is used to simulate and sample power flow paths under different operating conditions to generate multiple sets of flow path data; by analyzing the flow path data, the distribution characteristics of potential circulation paths are identified, including the frequency of occurrence and the range of influence of circulation paths; based on the distribution characteristics, the potential impact of circulation paths on power network stability is quantified and used as the quantification result of the distribution characteristics.

4. The method as described in claim 1, characterized in that, In step S2, the optimized set of locations is obtained, including: The quantization result of the distribution feature is obtained and compared with a preset threshold. If the quantization result exceeds the preset threshold, the influence of the candidate gate positions on the circulation path is analyzed for multi-power source scenarios. Based on the influence, the priority sequence of the candidate gate positions is adjusted to prioritize the selection of positions with less influence on the circulation path. The candidate gate positions that meet the conditions are selected through the priority sequence to generate an optimized set of positions.

5. The method as described in claim 1, characterized in that, In step S3, the recommended configuration for the final checkpoint setting is obtained, including: The sensitivity score is obtained and compared with a preset score. If the sensitivity score is lower than the preset score, a suitable alternative is extracted from the optimized location set. The performance of each alternative in terms of benefit balance and loss allocation is analyzed, and the optimal alternative is selected as the recommended configuration for the final gate setting based on the analysis results. The recommended configuration is output as the final setting scheme for the power network gate location.

6. The method as described in claim 1, characterized in that, In step S4, a dynamic adaptation scheme and checkpoint setting analysis report is generated, including: Based on the recommended configuration of the final checkpoint setting, deploy checkpoint locations and collect feedback data on actual operating conditions; by analyzing the feedback data, evaluate the loss distribution and benefit balance effect of the recommended configuration in actual operation; based on the evaluation results, determine whether iterative adjustment of the topology representation is required; if adjustment is required, update the topology representation and regenerate the topology characteristics of the checkpoint candidate locations; through iterative adjustment, generate a dynamic adaptation scheme for the checkpoint locations and a corresponding checkpoint setting analysis report, so that the scheme adapts to changes in actual operating conditions.

7. A system for automatically generating analysis reports on power gateway port settings, used to implement the method as described in any one of claims 1-6, characterized in that, The system includes: The topology modeling unit is used to generate a topology representation of the power network by collecting wiring diagram data and multi-source configuration information in the power network. The circulation analysis unit is used to obtain the property rights division boundary under the participation of multiple entities based on the topology representation, and to determine the distribution characteristics and corresponding quantitative results of potential circulation paths through the analysis of power flow paths under different operating conditions. The location optimization unit is used to adjust the priority sequence of the candidate locations of the checkpoints when the quantization result of the distribution feature exceeds a preset threshold, so as to obtain an optimized set of locations. The loss assessment unit is used to obtain the input parameters of the loss allocation model through the location set and update the model to simulate the impact of mutual supply phenomenon, and to determine the quantitative index of loss deviation. The recommended configuration unit is used to generate dynamic scenario samples and determine the sensitivity score of the checkpoint location to the balance of interests based on the quantitative indicators; if the sensitivity score is lower than the preset score, then alternative solutions are selected from the optimized location set to obtain the recommended configuration of the final checkpoint setting. The report generation unit is used to obtain feedback data of the power network under actual operating conditions through the recommended configuration, determine whether the topology representation needs to be iteratively adjusted based on the feedback data, and generate a dynamic adaptation scheme and gateway setting analysis report.

8. A computer-readable storage medium storing instructions thereon, characterized in that, When the instructions are executed by the processor, they implement the method as described in any one of claims 1-6.

Images