A power distribution network-microgrid topology optimization method and device

Abstract

The application discloses a kind of distribution network-microgrid topological structure optimization method and device, it is related to power grid technical field, comprising: setting composite constraint condition, call target topological structure data and carry out multi-objective collaborative optimization, obtain topological structure optimization parameter;Grid-connected off-grid switching analysis is carried out, and transient stability control data is constructed;According to topological structure optimization parameter, control distribution network-microgrid is combined with transient stability control data, obtains operating stability parameter, carries out power grid simulation test to topological structure optimization parameter is verified, according to the iteration correction of topological structure optimization parameter according to verification result, until meet preset index, generate topological structure optimization scheme and carry out control execution.The application solves the technical problem that existing technology cannot carry out high reliability, multi-objective collaborative optimization to distribution network-microgrid topological structure, achieves the technical effect that the topological structure optimization scheme that can dynamically adapt to operating condition change is obtained and reliable execution control is realized.

Description

Technical Field

[0001] This invention relates to the field of power grid technology, and specifically to a method and apparatus for optimizing the topology of a distribution network-microgrid. Background Technology

[0002] Distribution networks and microgrids typically exhibit characteristics such as multi-source access, significant load fluctuations, and frequent switching between grid-connected and off-grid states during actual operation, placing higher demands on the security and flexibility of their topologies. As the scale of distributed power sources continues to expand, topology analysis under single objectives or static conditions is insufficient to meet actual operational needs, failing to simultaneously consider multiple operational goals such as reliability, power distribution, and steady-state and transient characteristics. Topology optimization is often affected by the coupling of multiple constraints, lacking a systematic collaborative solution and dynamic verification mechanism, making it difficult to obtain the optimal operating topology adaptable to complex operating conditions. Summary of the Invention

[0003] This application provides a method and apparatus for optimizing the topology of a distribution network-microgrid, which addresses the technical problem that existing technologies cannot perform high-reliability, multi-objective collaborative optimization of the topology of a distribution network-microgrid.

[0004] In view of the above problems, this application provides a method and apparatus for optimizing the topology of a distribution network-microgrid.

[0005] The first aspect of this application provides a method for optimizing the topology of a distribution network-microgrid, the method comprising: A composite operation analysis based on the distribution network and microgrid is performed, and composite constraints are set. Target topology data of the distribution network and microgrid is retrieved, and multi-objective collaborative optimization is conducted on the target topology data according to the composite constraints to obtain topology optimization parameters. Grid-connected / off-grid switching analysis is performed based on the distribution network and microgrid to construct transient stability control data. The distribution network and microgrid are controlled according to the topology optimization parameters and the transient stability control data to obtain operational stability parameters. Grid simulation tests are conducted based on the operational stability parameters to verify the topology optimization parameters. The topology optimization parameters are iteratively corrected according to the verification results until preset indicators are met. The topology optimization scheme is then generated and distributed to the distribution network and microgrid for control execution.

[0006] A second aspect of this application provides a topology optimization device for a distribution network-microgrid, the device comprising: The system comprises the following modules: a constraint setting module for performing composite operation analysis based on the distribution network and microgrid, and setting composite constraints; a collaborative optimization module for retrieving target topology data of the distribution network and microgrid, and performing multi-objective collaborative optimization on the target topology data according to the composite constraints to obtain topology optimization parameters; a switching analysis module for performing grid-connected / off-grid switching analysis based on the distribution network and microgrid, and constructing transient stability control data; a control module for controlling the distribution network and microgrid according to the topology optimization parameters and the transient stability control data to obtain operational stability parameters; and a correction module for verifying the topology optimization parameters through grid simulation testing based on the operational stability parameters, iteratively correcting the topology optimization parameters according to the verification results until preset indicators are met, generating the topology optimization scheme, and distributing it to the distribution network and microgrid for control execution.

[0007] One or more technical solutions provided in this application have at least the following technical effects or advantages: This application performs composite operation analysis based on a distribution network and microgrid, setting composite constraints; retrieves target topology data of the distribution network and microgrid, and performs multi-objective collaborative optimization on the target topology data according to the composite constraints to obtain topology optimization parameters; performs grid-connected / off-grid switching analysis based on the distribution network and microgrid to construct transient stability control data; controls the distribution network and microgrid according to the topology optimization parameters and the transient stability control data to obtain operational stability parameters; verifies the topology optimization parameters through grid simulation testing based on the operational stability parameters, iteratively corrects the topology optimization parameters according to the verification results until preset indicators are met, generates the topology optimization scheme, and distributes it to the distribution network and microgrid for control execution. This invention solves the technical problem of existing technologies being unable to perform high-reliability, multi-objective collaborative optimization of distribution network and microgrid topology. Through a comprehensive process of composite operation analysis with constraint setting, multi-objective collaborative optimization, grid-connected / off-grid switching analysis, hierarchical collaborative control, and iterative verification and correction based on digital twins, it achieves the technical effect of obtaining a topology optimization scheme that can dynamically adapt to changes in operating conditions and realizing reliable execution control. Attached Figure Description

[0008] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0009] Figure 1A schematic flowchart of a method for optimizing the topology of a distribution network-microgrid provided in this application embodiment; Figure 2 This is a schematic diagram of a topology optimization device for a distribution network-microgrid provided in an embodiment of this application.

[0010] Explanation of reference numerals in the attached diagram: Constraint setting module 11, Collaborative optimization module 12, Switching analysis module 13, Control module 14, Correction module 15. Detailed Implementation

[0011] This application provides a method and apparatus for optimizing the topology of a distribution network-microgrid. It addresses the technical problem that existing technologies cannot perform high-reliability, multi-objective collaborative optimization of the distribution network-microgrid topology. Through a comprehensive process including constraint setting based on composite operation analysis, multi-objective collaborative optimization, grid-connected / off-grid switching analysis, hierarchical collaborative control, and iterative verification and correction based on digital twins, it achieves the technical effect of obtaining a topology optimization scheme that can dynamically adapt to changes in operating conditions and realizing reliable execution control.

[0012] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0013] It should be noted that any variation of the terms "comprising" and "having" is intended to cover non-exclusive inclusion, for example, a process, method, apparatus, product, or server that includes a series of steps or units is not necessarily limited to those steps or units that are explicitly listed, but may include other steps or modules that are not explicitly listed or that are inherent to such process, method, product, or apparatus.

[0014] Example 1, as Figure 1 As shown, this application provides a method for optimizing the topology of a distribution network-microgrid, the method comprising: Step S100: Perform composite operation analysis based on distribution network-microgrid and set composite constraints.

[0015] Furthermore, the method provided in the application embodiments, which performs composite operation analysis based on the distribution network-microgrid and sets composite constraint conditions, also includes: Multi-timescale operation analysis is performed on the distribution network-microgrid system to obtain multi-scale operation modes. Topology analysis is then performed on the distribution network-microgrid system according to these multi-scale operation modes, setting topology constraints. Operational analysis is also performed on the distribution network-microgrid system according to these multi-scale operation modes, setting operational constraints. Power analysis is then performed on the distribution network-microgrid system according to these multi-scale operation modes, setting power constraints. Based on the topology constraints, operational constraints, and power constraints, stochastic chance constraint programming is performed to construct initial composite constraints. Multi-scenario adaptability analysis is then performed on the initial composite constraints to generate multiple adaptation coefficients. Finally, the initial composite constraints are identified based on these multiple adaptation coefficients to obtain the composite constraints.

[0016] In this embodiment, when performing composite operation analysis based on the distribution network-microgrid, the operation analysis is first conducted at multiple time scales, collecting operational data spanning different time scales, including transient data with a sampling period of 1 second, short-cycle data with a sampling period of 15 minutes, and long-cycle data with a sampling period of 1 hour. In actual implementation, 96 sets of short-cycle data such as voltage, current, active power, and reactive power, 86,400 sets of second-level data such as frequency and transient current, and 24 sets of long-cycle data such as load levels are acquired over a continuous 24-hour period. Subsequently, a multi-scale data clustering method is used to classify the above data, dividing data samples with similar operating trends into multiple operating condition groups. For example, low load condition, high photovoltaic output condition, transient disturbance condition, and typical peak load condition. The classification results of data belonging to different time scales are integrated to form a multi-scale operation mode characterizing the features of different operating cycles.

[0017] Next, when performing topology analysis on the distribution network-microgrid according to multi-scale operation modes, a node-branch topology model is established based on the number of nodes, distributed power source access locations, and branch structure corresponding to each operation mode. During topology analysis, a node-branch association matrix A and an adjacency matrix B are constructed. For example, in a network containing N nodes and M branches, the dimension of A is N×M, and the dimension of B is N×N. By processing the node connectivity, branch on / off status, and branch combinations in the matrices, network connectivity, the number of power supply paths, loop structure, and important nodes are identified. When the load level corresponding to a certain operation mode is high, the analysis can focus on whether the power supply paths of key nodes meet the redundancy requirement of at least two paths. Based on the above topology information, a set of node connection constraints, branch on / off constraints, and connectivity constraints is compiled to form topology constraints.

[0018] Subsequently, during the operational analysis according to multi-scale operating modes, steady-state operating parameters are calculated based on the load level, distributed power output, and topology in different operating modes. In this calculation, node injected power, branch power flow, voltage amplitude, and current values ​​are used as key indicators and compared with operating standards. Taking commonly used technical indicators as examples, node voltage must be between 0.95 and 1.05 per unit, branch current must not exceed 90% of the rated thermal stability current, and transformer load rate must be within 80% of the rated capacity. If voltage deviation or branch current approaches its limit in a certain operating mode, the corresponding node or branch can be included in the operating restriction range. By summarizing the voltage limit conditions, current limit conditions, and equipment load conditions under different operating modes, operating constraints are formed.

[0019] Next, when performing power analysis according to the multi-scale operation mode, a power balance check is performed based on the node injected power and absorbed power, and the adjustable space is analyzed in conjunction with the rated power output range of distributed power sources, inverters, and energy storage devices. For example, the rated active power output capacity of the inverter is set to 300 to 500 kW, the charging and discharging capacity of the energy storage device is set to 100 to 200 kW, and the reactive power adjustment range is taken within a certain range according to the equipment model. Then, based on the maximum power limit that the branch can carry, the power transmission limit range of the line is determined. Combining the node power balance constraints, equipment power output constraints, and line power transmission limits, power constraint conditions are formed.

[0020] Subsequently, when performing stochastic chance-constrained programming based on topology constraints, operational constraints, and power constraints, random variables are first defined for load fluctuations, distributed power generation output fluctuations, and line operating states. Several stochastic scenarios are then generated using a disturbance model. For example, 300 stochastic operating scenarios based on load forecast deviations, photovoltaic output fluctuations, and equipment parameter changes at different time periods can be generated. After substituting the three types of constraints into each scenario, the satisfaction of the constraints in each scenario is statistically analyzed to determine whether the constraint satisfaction rate meets the set probability requirements, such as 90% or 95%. By expressing the proportion of scenarios that meet the requirements as a probabilistic expression, the initial composite constraint conditions containing probabilistic constraints can be constructed.

[0021] Next, a multi-scenario adaptability analysis was performed on the initial composite constraints. In this process, the number of constraints was initially set to 30 and the number of scenarios to 100. For each constraint, two indicators were calculated sequentially across all scenarios: a satisfaction mark (1) and an excess percentage (0). The satisfaction mark represents the relative deviation of a variable from its limit range; for example, when the node voltage exceeds the upper limit of 1.05 per unit (1.08 per unit), the excess percentage is 3%. Then, the satisfaction mark ratio for each constraint was averaged across all scenarios to obtain the satisfaction ratio, and the excess percentage ratio was averaged to obtain the average excess degree. To standardize the scale of different constraints, the average excess degree was divided by a preset maximum excess percentage (e.g., 10%), and the reverse result was taken to form a normalized excess index. When the average excess degree exceeds the maximum allowable excess value, the normalized excess index was set to 0 to indicate low adaptability. Then, a comprehensive evaluation is performed according to the set weights. For example, the satisfaction ratio weight is set to 0.7, and the normalized violation index weight is set to 0.3. The fitness coefficient is obtained through weighted calculation. Taking a voltage constraint as an example, when the voltage constraint is satisfied in 92 out of 100 scenarios, the satisfaction ratio is 0.92; when the corresponding average violation rate is 3%, under the condition of a maximum violation rate of 10%, the normalized violation index is 0.7; at this time, the fitness coefficient is equal to 0.92 multiplied by 0.7 plus 0.7 multiplied by 0.3, which is approximately equal to 0.854. The fitness coefficient is calculated for all constraints according to the above method, generating multiple fitness coefficients.

[0022] Finally, initial composite constraints are identified based on multiple fitness coefficients, and final composite constraints are constructed. In this process, constraints with fitness coefficients greater than or equal to 0.85 are classified as critical constraints and retained; constraints with fitness coefficients between 0.6 and 0.85 are classified as adjustable constraints and corrected by adjusting the limitation range or optimizing the constraint expression; constraints with fitness coefficients less than 0.6 are classified as weak constraints and weakened through simplification, merging, or deletion. By screening, integrating, and adjusting constraints of different levels, composite constraints are constructed.

[0023] Step S200: Retrieve the target topology data of the distribution network-microgrid, and perform multi-objective collaborative optimization on the target topology data according to the composite constraint conditions to obtain the topology optimization parameters.

[0024] Furthermore, in the method provided in the application embodiments, retrieving the target topology data of the distribution network-microgrid, performing multi-objective collaborative optimization on the target topology data according to the composite constraints, and obtaining topology optimization parameters, further includes: A topology database of the distribution network-microgrid is constructed, and the target topology data of the distribution network-microgrid is retrieved by traversing the topology database. Multi-objective analysis is performed based on the composite constraints and the target topology data to determine multiple optimization objectives. Based on the composite constraints and the multiple optimization objectives, the target topology data is collaboratively optimized to obtain a combination of topology optimization parameters. The performance of the topology optimization parameter combination is evaluated in multiple scenarios, and the impact of parameter fluctuations is analyzed based on the evaluation results to determine the topology optimization parameters.

[0025] In this embodiment, when retrieving target topology data of the distribution network-microgrid and performing multi-objective collaborative optimization on the target topology data according to composite constraints, a topology database of the distribution network-microgrid is first constructed, and the target topology data of the distribution network-microgrid is retrieved by traversing the topology database. In this process, firstly, topology analysis of the distribution network-microgrid is performed, and nodes are classified and identified according to the analysis results to determine the set of nodes in the network. Then, the connection mode of each branch is identified based on the connection relationship between nodes. Subsequently, graph theory methods are applied to integrate the node set according to the branch connection relationship to generate a network topology graph reflecting the node hierarchy, adjacency relationship, and path structure. Then, node-branch association analysis is performed based on the network topology graph to construct a node-branch association matrix of the network topology. Then, different topology forms are matched based on the node-branch association matrix, and topology screening conditions are set for screening the structure. Finally, a depth-first search is performed on the topology database based on the screening conditions to determine the target topology data.

[0026] Next, a multi-objective analysis is performed based on composite constraints and target topology data. First, power flow calculations are performed under the target topology to obtain the voltage of each node and the current and active power flow of each branch. The rated voltage is set to 10kV. Voltage deviation is calculated for each node, which is equal to the node voltage minus the absolute value of 10kV, divided by 10kV, and then multiplied by 100%. The maximum and average voltage deviations across all nodes are calculated to reflect voltage quality. Current utilization is calculated for each branch, which is equal to the actual current divided by the branch's rated current and then multiplied by 100%. The maximum and average current utilizations are calculated to reflect line safety margin. Regarding losses, the active power loss of each branch is obtained by multiplying the resistance and the square of the current. The total active power loss of the network is obtained by summing the losses of all branches and comparing it with the losses of the baseline topology to see if the total loss is too high. Regarding renewable energy utilization, the power generation of distributed power sources such as photovoltaics and the power load of users are statistically analyzed within a given operating cycle. The renewable energy utilization rate is equal to the renewable energy power divided by the load power and then multiplied by 100%. For power supply reliability, the amount of unsupplied power is estimated based on the load power of critical loads and the possible outage time. The sum of the unsupplied power for all critical loads is used to measure the reliability level. Subsequently, according to the requirements of composite constraints, the voltage deviation, current utilization rate, total loss, renewable energy utilization rate, and unsupplied power are normalized. For example, the actual value is divided by the allowable upper limit or typical reference value, scaling these quantities to a range of 0 to 1. For quantities where smaller values ​​are better, such as voltage deviation, loss, and unsupplied power, the normalized value can be subtracted from 1 to convert them into a form where larger values ​​are better. By comparing these processed values, problems such as excessively large voltage deviation, excessively high current utilization rate, excessively high total loss, excessively low renewable energy utilization rate, or excessively large unsupplied power can be identified. Based on this, the voltage deviation control target, line current safety target, loss reduction target, renewable energy utilization rate improvement target, and power supply reliability improvement target are established as multiple optimization targets.

[0027] Subsequently, based on composite constraints, the target topology data is collaboratively optimized according to multiple optimization objectives. In this process, firstly, a topology clustering structure is constructed based on multiple optimization objectives, forming multiple topology clusters; then, within the constraints of the composite constraints, each cluster is dimensionality-reduced and representative parameter solutions are extracted; next, entropy-weighted multi-attribute evaluation generates parameter attribute advantage and disadvantage information, and combines this with historical operation logs of the distribution network-microgrid to form operation preference data; under the combined effect of the operation preference data and the bidirectional evaluation information of parameter attributes, multiple selected solutions are determined; finally, environmental sensitivity verification screens out the topology optimization parameter combination that meets the comprehensive performance requirements.

[0028] Finally, multi-scenario performance evaluations were conducted on the topology optimization parameter combinations, and parameter fluctuation impact analysis was performed based on the evaluation results. This process involved first constructing a multi-scenario test set based on historical operating scenarios of the distribution network-microgrid, and then using this test set to simulate the topology optimization parameter combinations to obtain multi-scenario simulation data. Subsequently, multi-scenario performance evaluations were performed based on the simulation data to generate performance evaluation results. Then, based on the performance evaluation results, global sensitivity analysis was performed to generate multiple sensitivity indices and construct a parameter-performance interaction response surface. Finally, the topology optimization parameter combinations were sorted in descending order according to the parameter-performance interaction response surface, identifying key and secondary parameters, and using the key parameters as the final topology optimization parameters output.

[0029] Furthermore, the method provided in the application embodiment, which involves constructing a distribution network-microgrid topology database and traversing the topology database to retrieve target topology data of the distribution network-microgrid, further includes: A topology analysis is performed on the distribution network-microgrid system. Based on the analysis results, the nodes are grouped and identified to determine multiple network nodes. Connection analysis is then performed on these nodes to determine multiple branch connection relationships. These multiple network nodes are then integrated using graph theory to construct a network topology graph. Based on this graph, node-branch association analysis is performed to construct a node-branch association matrix. The distribution network-microgrid system is then matched for operation based on this association matrix, and topology structure selection criteria are set. Finally, a depth-first search is performed on the topology database based on these selection criteria to determine the target topology data for the distribution network-microgrid system.

[0030] In this embodiment, when performing topology analysis by traversing the distribution network-microgrid, the primary wiring structure of the distribution network-microgrid is first traversed. The electrical connections of buses, lines, switches, transformers, and distributed power source access points are identified one by one. During the traversal, based on electrical equipotentiality and direct conductor connections, endpoints at the same electrical equipotential and directly connected by conductors or cables are grouped and labeled. Each group of electrical endpoints is marked as a network node, thus obtaining multiple network nodes. Then, based on the actual conductor connections between network nodes, connection analysis is performed on node combinations. It is determined whether any two network nodes have a direct electrical connection. Node pairs with direct connections are defined as branches, thus determining multiple branch connection relationships.

[0031] Next, when integrating multiple network nodes according to multiple branch connection relationships using graph theory, multiple network nodes are used as nodes in the graph structure, and multiple branch connection relationships are used as edges connecting the corresponding nodes. Based on graph theory, all multiple network nodes and multiple branch connection relationships are structurally integrated. By establishing node sets and edge sets, the original primary wiring information is transformed into an abstract network topology graph. During the integration process, isolated nodes and redundant connections without electrical significance are regularized, so that the network topology graph can accurately reflect the power supply path relationship, feeder structure and interconnection structure between multiple network nodes, thus forming the network topology graph.

[0032] Subsequently, when performing node-branch association analysis based on the network topology graph, multiple network nodes in the network topology graph are used as the row dimension of the node-branch association matrix, and multiple branch connections in the network topology graph are used as the column dimension of the node-branch association matrix. The connection status of each network node and each branch in the network topology graph is determined. When a network node is the connection endpoint of a branch, a value of 1 is assigned to the corresponding position in the node-branch association matrix. When there is no direct connection between a network node and a branch, a value of 0 is assigned to the corresponding position in the node-branch association matrix. By performing the above node-branch association analysis on all multiple network nodes and all multiple branch connections, the node-branch association matrix of the network topology is constructed.

[0033] Subsequently, the distribution network-microgrid is matched for operation according to the node-branch correlation matrix. When setting the topology screening conditions, the node-branch correlation matrix is ​​matched with the load distribution, distributed generation access nodes, power supply range of each feeder, and nodes where critical loads are located in the distribution network-microgrid. The consistency between the load power supply demand at the node level and the electrical connection method at the branch level is compared. At the same time, the requirements for node connection, branch operation mode, and power supply path in the composite constraint conditions are combined to form the structure screening conditions, including the connectivity relationships that nodes should meet, the connection mode requirements of branches, and the power supply path requirements of the nodes where loads are located. This allows the topology screening conditions to be directly applied to the screening of the topology database.

[0034] Finally, when performing a depth-first search based on the topology database using the topology filtering criteria, each set of topology data in the database is read sequentially. Multiple network nodes and branch connections are then reconstructed into a node-branch association matrix according to corresponding rules, and this matrix is ​​compared with the topology filtering criteria. During the depth-first search, the search expands layer by layer along the paths formed by network nodes and branches, performing condition checks at each branch. If a path does not meet the filtering criteria, it is excluded through backtracking using the depth-first search mechanism. When a topology data set meets all the filtering criteria in terms of network node connectivity, branch connection methods, and load power supply paths, it is determined as a selectable structure. All topology data that meet the requirements are then aggregated to ultimately determine the target topology data for the distribution network-microgrid.

[0035] Furthermore, in the method provided in the application embodiments, the method further includes: Based on the composite constraints, collaboratively optimizing the target topology data according to the multiple optimization objectives to obtain a combination of topology optimization parameters; Cluster analysis is performed on the target topology data according to the multiple optimization objectives to determine multiple topology clusters; dimensionality reduction is performed on the multiple topology clusters according to the composite constraints to generate multiple representative parameter solutions; entropy weighting is applied to the multiple representative parameter solutions for multi-attribute evaluation to obtain bidirectional evaluation information of parameter attributes, which includes parameter attribute advantage information and parameter attribute disadvantage information; historical operation logs of the distribution network-microgrid are retrieved for preference analysis to obtain operation preference data; weighted analysis is performed on the multiple representative parameter solutions according to the operation preference data combined with the parameter attribute advantage information and the parameter attribute disadvantage information to determine multiple selected solutions; environmental sensitivity verification is performed on the multiple selected solutions to obtain the topology optimization parameter combination.

[0036] In this embodiment, when performing cluster analysis on the target topology data according to multiple optimization objectives, a cluster analysis method based on Euclidean distance is adopted. The voltage deviation, current utilization rate, active power loss, renewable energy utilization rate, and unsupplied power of each topology sample are combined into a 5-dimensional feature vector. A distance matrix is ​​constructed by calculating the Euclidean distance between any two feature vectors, and the distance represents the similarity between samples across multiple optimization objectives. The samples are then classified according to the distance matrix, grouping samples with smaller distances into the same cluster. The cluster centers are iteratively updated until convergence, resulting in multiple topological clusters.

[0037] Next, when performing dimensionality reduction on multiple topological clusters under composite constraints, a variance contribution analysis method is used. The variance of each performance index within each topological cluster is calculated, and the variance of each index reflects its contribution to the cluster differences. All index variances are sorted from largest to smallest, and several performance indices with larger variance values ​​are selected to form a new low-dimensional eigenvector. This ensures that the eigenvector after dimensionality reduction still retains the main differences between the original clusters, thereby generating multiple representative parameter solutions.

[0038] Subsequently, when applying entropy weights for multi-attribute evaluation across multiple representative parameter solutions, the entropy weight method is used to calculate the weights of each performance index. For each representative parameter solution, voltage deviation, current utilization rate, active power loss, renewable energy utilization rate, and unsupplied electricity are first normalized, linearly mapping all index values ​​to the range of 0 to 1. For each index, the proportion of its normalized value across all representative parameter solutions is calculated, and its entropy value is then calculated based on this proportion. The entropy weight of the index is obtained by subtracting the entropy value from 1, reflecting its importance in distinguishing different representative parameter solutions. After obtaining the entropy weights, the target direction for each type of performance index is pre-determined. For voltage deviation, current utilization rate, active power loss, and unsupplied electricity, smaller values ​​are preferred, while for renewable energy utilization rate, larger values ​​are preferred. Then, the average value of all representative parameter solutions is calculated for each index, and this average value is used as the reference value for that index. For indicators where smaller values ​​are generally better, when the indicator value of a representative parameter solution is lower than the reference value, the advantage magnitude is calculated by subtracting the indicator value from the reference value, and this difference is included in the parameter attribute advantage information. When the indicator value is higher than the reference value, the disadvantage magnitude is calculated by subtracting the reference value from the indicator value, and this difference is included in the parameter attribute disadvantage information. For indicators where larger values ​​are generally better, when the indicator value of a representative parameter solution is higher than the reference value, the difference is included in the parameter attribute advantage information. When the indicator value is lower than the reference value, the difference is included in the parameter attribute disadvantage information. Through this process, a two-way evaluation information of parameter attributes, consisting of parameter attribute advantage information and parameter attribute disadvantage information, is obtained.

[0039] Subsequently, when retrieving historical operation logs of the distribution network-microgrid for preference analysis, statistical analysis methods were used to organize and frequency-count the voltage level, line load level, network loss level, renewable energy output ratio, and key load power supply status recorded in the historical operation logs. Specifically, the percentage of time each node voltage fell within ±5% of the rated voltage, the percentage of time line current utilization was between 60% and 80%, the percentage of time active power loss was within different ranges, the percentage of time renewable energy output as a percentage of load was within different ranges, and the percentage of time when the amount of unsupplied power was close to zero was calculated. Based on these time percentages, the degree of attention given to each performance indicator in actual operation was determined. The time percentage values ​​were normalized by summing the time percentages of all indicators to obtain the operational preference weights for each performance indicator. The set of operational preference weights constitutes operational preference data, which reflects the importance ranking of each performance indicator in long-term operation.

[0040] Next, when performing a weighted analysis of multiple representative parameter solutions based on operational preference data combined with parameter attribute advantage and disadvantage information, a weighted summation method is used to comprehensively evaluate each representative parameter solution. The operational preference weights of each performance indicator in the operational preference data are combined with the corresponding parameter attribute advantage and disadvantage information in the representative parameter solution. Specifically, for each performance indicator, the advantage magnitude of the representative parameter solution on that indicator is multiplied by the product of the operational preference weight and the entropy weight to obtain the advantage contribution value; the disadvantage magnitude of the representative parameter solution on that indicator is multiplied by the product of the operational preference weight and the entropy weight to obtain the disadvantage penalty value. Then, the advantage contribution value is subtracted from the disadvantage penalty value to obtain the comprehensive contribution of that indicator to the representative parameter solution. The comprehensive contributions of all performance indicators are summed to obtain the total evaluation value of the representative parameter solution. The total evaluation values ​​of all representative parameter solutions are then ranked, and several representative parameter solutions with the highest total evaluation values ​​are selected as multiple selected solutions.

[0041] Finally, environmental sensitivity verification was performed on multiple selected solutions. A multi-scenario simulation method was used to construct various operating scenarios consisting of different load levels, renewable energy output levels, and line operating states. The parameter configurations of each selected solution were substituted into each operating scenario, and voltage deviation, current utilization rate, active power loss, renewable energy utilization rate, and unsupplied power were obtained under each scenario through power flow calculation and operational analysis. For each selected solution, the difference between the maximum and minimum values ​​of the above indicators in all scenarios, or the standard deviation, was calculated to measure the performance fluctuation of the selected solution under different operating scenarios. The performance fluctuation of all selected solutions was compared, and several selected solutions with performance indicators that met the composite constraints and had small fluctuation amplitudes under multiple scenarios were selected. The parameter configuration sets of these selected solutions were determined as the topology optimization parameter combinations, thereby obtaining optimization parameter results with a good balance between multi-objective performance and changes in the operating environment.

[0042] Furthermore, in the method provided in the application embodiments, the multi-scenario performance evaluation of the topology optimization parameter combination, the parameter fluctuation impact analysis based on the evaluation results, and the determination of the topology optimization parameters further include: Historical operating scenarios of the distribution network-microgrid are retrieved to construct a multi-scenario test set; the topology optimization parameter combination is simulated according to the multi-scenario test set to obtain multi-scenario simulation data; multi-scenario performance evaluation is performed based on the multi-scenario simulation data to generate performance evaluation results; global sensitivity analysis is performed based on the performance evaluation results to generate multiple sensitivity indices for interactive response, constructing a parameter-performance interactive response surface; the topology optimization parameter combination is sorted in descending order according to the parameter-performance interactive response surface, key parameters and minor parameters are identified, the minor parameters are filtered out, and the key parameters are output as topology optimization parameters.

[0043] In this embodiment, historical operating scenarios of the distribution network-microgrid are first retrieved from the historical database. Representative operating days, operating weeks, or typical operating conditions are used as the source of historical operating scenarios, and data such as node voltage, current load, distributed generation output, and equipment operating status are extracted from them. Subsequently, the historical data is structured and classified according to load level, distributed generation output level, and topology differences to form a multi-scenario test that includes various operating conditions.

[0044] Next, the topology optimization parameter combinations are simulated using a multi-scenario test set. During this process, a steady-state power flow analysis method is employed. For each scenario, the topology optimization parameter combinations are input, and the voltage magnitude of each node, the current value of each branch, and the power transmission value of each branch are calculated. The simulation outputs are recorded according to their assigned numbers, forming a multi-scenario simulation dataset containing the results of all scenarios.

[0045] Subsequently, when conducting multi-scenario performance evaluations based on multi-scenario simulation data, performance indicators such as node voltage deviation, current utilization rate, active power loss, renewable energy utilization rate, and unsupplied power are used as the evaluation basis, and the simulation results for each scenario are quantitatively calculated. For example, node voltage deviation is obtained by dividing the absolute value of the difference between the actual voltage and the rated voltage by the rated voltage and then multiplying by 100%; current utilization rate is obtained by dividing the branch current by the rated current of the branch and then multiplying by 100%; active power loss is obtained by multiplying the resistance by the square of the current; renewable energy utilization rate is obtained by dividing the renewable energy generation by the total load power and then multiplying by 100%; and unsupplied power is obtained by multiplying the critical load power by the duration of power outage. All of the above indicators are organized according to the scenario to generate performance evaluation results.

[0046] Subsequently, when performing global sensitivity analysis and generating multiple sensitivity indices based on the performance evaluation results, a single-factor perturbation method was used to process each topology optimization parameter. By applying fixed-amplitude positive and negative perturbations to a specific parameter while keeping other parameters constant, power flow calculations were re-executed, and the corresponding performance index changes were recorded. Dividing the change in each performance index by the corresponding perturbation amplitude yields the parameter's sensitivity index for that index. For each parameter, its sensitivity index across all performance indices was statistically analyzed and recorded to generate the parameter-performance interaction response surface. Then, when constructing the parameter-performance interaction response surface, each topology optimization parameter was used as the independent variable, and the corresponding multiple sensitivity indices were used as dependent variables. The sensitivity indices were mapped to the performance indices using coordinates, forming a visualized response structure based on the multidimensional sensitivity characteristics of each parameter. This response structure characterizes the impact of parameter changes on performance indices such as node voltage deviation, current utilization, active power loss, renewable energy utilization, and unsupplied power, constituting the parameter-performance interaction response surface in a multidimensional space.

[0047] Finally, the topology optimization parameter combinations are sorted in descending order according to the parameter-performance interaction response surface. When identifying key and secondary parameters, the absolute values ​​of the sensitivity indices of each parameter are first averaged to form a comprehensive sensitivity index. Then, the comprehensive sensitivity indices are sorted from largest to smallest, and a classification threshold is set. For example, a parameter with a comprehensive sensitivity index greater than or equal to 0.50 is classified as a key parameter; a parameter with a comprehensive sensitivity index less than 0.50 is classified as a secondary parameter. After classification, secondary parameters are removed, and only the key parameters are retained and output as the final topology optimization parameters.

[0048] Step S300: Perform grid-connected / off-grid switching analysis based on the distribution network-microgrid, and construct transient stability control data.

[0049] In this embodiment, when performing grid-connected / off-grid switching analysis based on a distribution network-microgrid system, the analysis first focuses on the instantaneous switching between grid-connected and off-grid states. An equivalent electrical structure model consisting of a bus, power source, energy storage device, and inverter is established. Voltage and frequency changes during switching are calculated as core transient quantities. The calculation incorporates the switching switch's action time, the inverter control's response time constant, and the energy storage power regulation rate to obtain frequency offset and voltage jump curves during the switching process. Subsequently, the maximum frequency deviation and frequency recovery time are calculated based on the frequency offset curve, and the maximum voltage deviation and voltage stabilization time are calculated based on the voltage jump curve. These quantities are used as the main indicators reflecting the degree of transient stability. The above calculations are then repeated under different load levels, photovoltaic output, and energy storage state of charge conditions to form a transient stability dataset covering multiple operating scenarios. Finally, based on the transient stability dataset, the distribution range of each transient index is statistically analyzed, and the parameter range required for switching control is extracted accordingly, including the maximum allowable switching current, the inverter initial voltage support reference value, and the energy storage transient output adjustment limit value, thereby constructing the transient stability control data.

[0050] Step S400: Control the distribution network-microgrid according to the topology optimization parameters and the transient stability control data to obtain the operating stability parameters.

[0051] Furthermore, in the method provided in the application embodiments, controlling the distribution network-microgrid according to the topology optimization parameters and the transient stability control data to obtain operating stability parameters further includes: The topology optimization parameters are configured to the distribution network-microgrid to generate a grid configuration result; multiple control parameters are tuned based on the transient stability control data; multi-level collaborative control is performed according to the grid configuration result using the multiple control parameters: S1: The multi-level distribution network-microgrid includes a grid equipment layer, a grid unit layer, and a grid system layer, wherein the grid equipment layer, the grid unit layer, and the grid system layer have an information interaction mechanism; S2: Based on the grid equipment layer, the inverter is activated to perform local fast control on the distribution network-microgrid to obtain a first control result; S3: Based on the grid unit layer, the microgrid central controller is activated to perform unit coordination control to obtain a second control result; S4: Based on the grid system layer, the distribution network dispatch controller is activated to perform optimized control to obtain a third control result; S5: Based on the first control result, the second control result, and the third control result, collaborative control analysis is performed through the information interaction mechanism to obtain a collaborative control result; the distribution network-microgrid is monitored in real time according to the collaborative control result to generate the operating stability parameters.

[0052] In this embodiment, when controlling the distribution network-microgrid according to the topology optimization parameters and transient stability control data, the topology optimization parameters are first configured to the distribution network-microgrid. That is, based on the switch states, tie switch positions, active and reactive power output settings of distributed generation sources, and charging and discharging power settings of energy storage devices given in the topology optimization parameters, each parameter is mapped to the corresponding node and branch according to the node number and branch number of the distribution network-microgrid, forming a parameter allocation table for the node layer and branch layer. Then, a power flow calculation is performed based on the parameter allocation table to check whether the node voltage range, branch current exceeds the limit, and power balance is satisfied. When all node voltages are within the allowable range, branch currents do not exceed the allowable current, and power balance of each node is satisfied, the parameter allocation table is recorded as the grid configuration result.

[0053] Subsequently, when tuning multiple control parameters based on transient stability control data, the maximum allowable voltage deviation, maximum allowable frequency deviation, allowable switching current, and transient recovery time requirements given in the transient stability control data are used as constraints. The droop control coefficient of the inverter, the proportional-integral parameters of the voltage and frequency control loops, the active and reactive power regulation coefficients of the energy storage device, and the voltage and frequency thresholds in the microgrid switching criteria are tuned respectively. Specifically, by applying typical grid-connected / off-grid switching conditions in the simulation environment, the control parameters are adjusted step by step. The voltage recovery time, frequency recovery time, and peak switching current are observed after the switch. When the voltage deviation and frequency deviation do not exceed the limits specified in the transient stability control data, the recovery time is within the given range, and the switching current does not exceed the allowable value, the corresponding set of inverter parameters, energy storage regulation parameters, and switching criterion parameters are solidified into multiple control parameters.

[0054] Finally, when implementing multi-level coordinated control based on multiple control parameters according to the grid configuration results, the control scope of the grid equipment layer, grid unit layer, and grid system layer is first defined based on the grid configuration results, and an information interaction mechanism is established among the three to realize the transmission of status variables and the issuance of control variables. At the grid equipment layer, the local fast control of the inverter is activated according to the local control settings among the multiple control parameters, and the active and reactive power outputs are adjusted according to the voltage and frequency deviations to form the first control result. At the grid unit layer, the microgrid central controller performs coordinated control of the execution unit based on the distributed power output and energy storage power allocation settings among the multiple control parameters to form the second control result. At the grid system layer, the distribution network dispatch controller implements optimized control based on the settings of tie switch control and reactive power compensation control among the multiple control parameters to form the third control result. Subsequently, the first, second, and third control results are coordinated through an information exchange mechanism. Any discrepancies in the control quantities are uniformly corrected according to the priority principle to form the final coordinated control result. Finally, based on the coordinated control result, the operating quantities such as node voltage, system frequency, branch power flow, and energy storage output are monitored in real time. Stability indicators such as voltage deviation, frequency deviation, power flow fluctuation amplitude, and power response time are calculated from the monitoring data, and these indicators are output as operating stability parameters, thus generating operating stability parameters.

[0055] Step S500: Based on the stable operating parameters, perform a power grid simulation test to verify the topology optimization parameters. According to the verification results, iteratively correct the topology optimization parameters until the preset indicators are met. Generate the topology optimization scheme and send it to the distribution network-microgrid for control execution.

[0056] In this embodiment, when verifying the topology optimization parameters based on power grid simulation tests using operational stability parameters, a digital twin test platform reflecting the operating characteristics of the distribution network-microgrid is first constructed based on the operational stability parameters. The topology optimization parameters are then synchronized to this platform for power grid simulation tests to generate simulation test results. Subsequently, the performance of the topology optimization parameters is evaluated based on the simulation test results, including transient stability analysis to obtain a first analysis result and steady-state operation analysis to obtain a second analysis result. Finally, the reliability verification is conducted by combining the first and second analysis results to form parameter analysis results, and the parameter verification of the topology optimization parameters is completed accordingly to generate verification results.

[0057] Subsequently, the topology optimization parameters are iteratively corrected based on the verification results. During this process, the topology optimization parameters are gradually adjusted based on the verification results, and the parameter values ​​are continuously updated through an adaptive iterative correction method until the relevant performance indicators meet the preset requirements. After parameter convergence, a topology optimization scheme is formed, and the scheme is distributed through a security verification mechanism to enable the distribution network-microgrid to achieve control execution according to the optimized topology.

[0058] Furthermore, in the method provided in the application embodiment, the topology optimization parameters are verified by performing power grid simulation tests based on the operational stability parameters, and the topology optimization parameters are iteratively corrected according to the verification results until preset indicators are met. The topology optimization scheme is then generated and distributed to the distribution network-microgrid for control execution. The method further includes: A digital twin test platform is constructed based on the aforementioned stable operating parameters. The topology optimization parameters are synchronized to the digital twin test platform to perform power grid simulation tests and generate simulation test results. The effectiveness of the topology optimization parameters is evaluated based on the simulation test results: transient stability analysis is performed on the topology optimization parameters based on the simulation test results to obtain a first analysis result; steady-state operation analysis is performed on the topology optimization parameters based on the simulation test results to obtain a second analysis result; reliability verification is performed based on the first and second analysis results to obtain parameter analysis results; parameter verification is performed on the topology optimization parameters according to the parameter analysis results to generate verification results; an adaptive iterative correction algorithm is used to optimize and adjust the topology optimization parameters based on the verification results. When the optimized parameters meet preset indicators, a topology optimization scheme is generated, and the topology optimization scheme is distributed to the distribution network-microgrid system for execution through a security verification mechanism.

[0059] In this embodiment, when constructing a digital twin test platform based on operational stability parameters, a virtual model framework for simulating the dynamic behavior of a distribution network-microgrid is first formed based on the operational characteristics recorded in the operational stability parameters, such as node voltage, system frequency, branch power flow, and power response time. The switching states, distributed generation output setpoints, and energy storage device power configurations in the topology optimization parameters are then synchronized to the digital twin test platform to achieve a consistent mapping of the topology configuration in the simulation space. Subsequently, grid simulation tests are performed on the digital twin test platform. Typical disturbances are applied in the simulation environment, such as a 10% load step disturbance at 5 seconds and a 20% decrease in photovoltaic output at 15 seconds. Simultaneously, grid-connected and off-grid switching conditions are tested to obtain simulation test results, including voltage curves, frequency change rate, branch power flow timing, and energy storage power regulation response.

[0060] Next, when evaluating the effectiveness of the topology optimization parameters based on the simulation test results, a transient stability analysis is first performed. The transient response quantization method is used to calculate the maximum frequency shift, frequency recovery time, minimum voltage drop, and voltage recovery time during the transient process. For example, when the system frequency drops from 50 Hz to 49.6 Hz after a disturbance, the maximum shift is 0.4 Hz. If it recovers to above 49.9 Hz within 2.5 seconds, the frequency recovery time is recorded as 2.5 seconds. Simultaneously, when the node voltage drops to a minimum of 0.88 per unit and recovers to above 0.95 per unit within 1.8 seconds, the voltage recovery time is recorded as 1.8 seconds. These indicators constitute the first analysis result. Subsequently, steady-state operation analysis was conducted based on the simulation test results. The steady-state power flow calculation method was used to quantify the steady-state voltage deviation at each node, branch current utilization rate, total active power loss, and renewable energy utilization rate. For example, the steady-state node voltage deviation must be limited to within ±5%, the branch current utilization rate must be less than 90%, the acceptable increase in total active power loss compared to the baseline topology should not exceed 5%, and the renewable energy utilization rate must reach above 85%. The above quantitative assessment formed the second analysis result. Then, reliability verification was conducted based on the first and second analysis results. A threshold comparison method was used to compare all transient and steady-state indicators item by item. For example, the maximum frequency offset must be less than 0.5 Hz, the minimum voltage drop must not be less than 0.85 per unit, the voltage recovery time must not exceed 3 seconds, the steady-state voltage deviation must not exceed the allowable range, and the branch current must not exceed the rated current. The parameter analysis results were then used to determine whether the operational technical standards were met.

[0061] Next, the topology optimization parameters are verified based on the parameter analysis results. In this process, the transient stability index, steady-state operation index, and reliability index in the parameter analysis results are compared with the preset reference indexes. If any one of them fails to meet the requirements, its offset is recorded and marked as a parameter to be corrected; if all requirements are met, the verification result is formed.

[0062] Subsequently, based on the verification results, an adaptive iterative correction algorithm is used to optimize and adjust the topology optimization parameters. During this process, target evaluation is performed based on the parameter verification results to determine the parameter correction target. Then, under the constraint of the correction target, local and global iterative searches are performed on the topology optimization parameters to obtain local and global parameter information to be corrected. These two are then combined to form parameter correction information containing correction step size and correction direction information. Next, based on the correction direction information and the correction step size information, multiple rounds of iterative optimization are performed on the topology optimization parameters to obtain optimized parameters reflecting the correction effect. Finally, the correction effect of the optimized parameters is evaluated in real time, and a topology optimization scheme is generated after it meets the preset comprehensive index.

[0063] Finally, a security verification mechanism is used to conduct a security consistency review of the topology optimization scheme, including verification of the legality of switch operations, power flow over-limit verification, control command conflict verification, and parameter range verification. After all verifications are passed, the topology optimization scheme is sent to the distribution network-microgrid system for execution, thereby ensuring that the scheme can operate safely in the actual power grid.

[0064] Furthermore, in the method provided in the application embodiment, based on the verification results, an adaptive iterative correction algorithm is used to optimize and adjust the topology optimization parameters. When the optimization parameters meet the preset indicators, a topology optimization scheme is generated, which further includes: Based on the verification results, target evaluation is performed, and parameter correction targets are set. Local iterative search is conducted on the topology optimization parameters according to the parameter correction targets to obtain local parameter information to be corrected. Global iterative search is conducted on the topology optimization parameters according to the parameter correction targets to obtain global parameter information to be corrected. Based on the local and global parameter information to be corrected, the topology optimization parameters are corrected and analyzed to determine parameter correction information, which includes correction step size and correction direction information. Multiple rounds of iterative optimization are performed on the topology optimization parameters according to the correction direction information and the correction step size information to obtain optimized parameters, which include parameter correction effects. The parameter correction effects are evaluated in real time until the optimized parameters meet preset comprehensive indicators, generating a topology optimization scheme.

[0065] In this embodiment, when evaluating targets based on verification results, key performance indicators from the verification results are extracted and compared to obtain deviation information for parameters such as node voltage deviation, current utilization rate deviation, active power loss variation, and renewable energy utilization rate variation. These deviations are then compared item by item with preset technical indicators, such as specifying that voltage deviation does not exceed 5% of the rated value and branch current utilization rate does not exceed 90% of the rated current. During the comparison process, performance indicators that do not meet or are close to the upper limit are marked, and parameter correction targets are set based on the technical requirements corresponding to these indicators.

[0066] Subsequently, during the local iterative search of topology optimization parameters according to the parameter correction target, a neighborhood perturbation method is employed. Parameters directly related to the parameter correction target are adjusted slightly around their current values. For example, reactive power compensation capacity, distributed power generation reactive power output settings, or switching states are increased or decreased by 1% to 2%. After each parameter adjustment, power flow calculations are performed, and voltage deviation, current utilization, and loss changes are recorded. Parameter adjustments that can reduce deviations are extracted as local parameters to be corrected.

[0067] Next, when performing a global iterative search for topology optimization parameters according to the parameter correction target, a multi-point sampling method is adopted. Several discrete value points are selected within the allowable parameter range, for example, 20 or 30 values ​​are selected at equal intervals. Power flow calculations are performed on each set of values, and the corresponding performance indicators are recorded. By comparing the performance of all candidate value points, parameter value regions that can significantly improve voltage deviation, current utilization, or loss level globally are identified, and the parameter values ​​within these regions are used as global parameters to be corrected.

[0068] Subsequently, when performing correction analysis based on local and global parameter information, a sensitivity analysis method is adopted. By comparing the impact of parameter changes on indicators such as voltage deviation, current utilization rate, and loss in the local and global ranges, the correction direction of each parameter is determined, such as increasing or decreasing the value of a certain parameter. The correction step size is set according to the performance improvement rate. For example, a smaller step size is used for parameters with high sensitivity, and a larger step size is used for parameters with low sensitivity, thereby forming parameter correction information that includes correction direction and correction step size.

[0069] Next, when performing multiple rounds of iterative optimization of the topology parameters according to the correction direction and correction step size, a round-by-round update method is adopted. In each round, the parameter values ​​are adjusted according to the correction direction, and the adjustment range is controlled according to the correction step size. After each round of adjustment, power flow calculation is performed to record performance indicators such as voltage deviation, current utilization rate, and loss, thereby obtaining the parameter correction effect. In multiple rounds of iteration, when the improvement of various performance indicators gradually tends to stabilize, it can be determined that the parameters are approaching the optimal value range.

[0070] Finally, when evaluating the effect of parameter correction in real time and determining whether the optimized parameters meet the preset comprehensive indicators, a threshold verification method is used to verify the consistency between the performance indicators after each iteration and the preset comprehensive indicators. For example, it checks whether the node voltage is within the specified range, whether the branch current utilization rate is lower than the allowable upper limit, whether the network active power loss is lower than the predetermined proportion of the baseline scheme, and whether the renewable energy utilization rate reaches the predetermined level. When the optimized parameters of a certain round meet the preset requirements in all comprehensive indicators, they are confirmed as optimized parameters, and a topology optimization scheme is generated accordingly. If the requirements are not met, the iteration steps continue until the preset comprehensive indicators are met, and a topology optimization scheme is generated.

[0071] In summary, the embodiments of this application have at least the following technical effects: This application performs composite operation analysis based on a distribution network and microgrid, setting composite constraints; retrieves target topology data of the distribution network and microgrid, and performs multi-objective collaborative optimization on the target topology data according to the composite constraints to obtain topology optimization parameters; performs grid-connected / off-grid switching analysis based on the distribution network and microgrid to construct transient stability control data; controls the distribution network and microgrid according to the topology optimization parameters and the transient stability control data to obtain operational stability parameters; verifies the topology optimization parameters through grid simulation testing based on the operational stability parameters, iteratively corrects the topology optimization parameters according to the verification results until preset indicators are met, generates the topology optimization scheme, and distributes it to the distribution network and microgrid for control execution. This invention solves the technical problem of existing technologies being unable to perform high-reliability, multi-objective collaborative optimization of distribution network and microgrid topology. Through a comprehensive process of composite operation analysis with constraint setting, multi-objective collaborative optimization, grid-connected / off-grid switching analysis, hierarchical collaborative control, and iterative verification and correction based on digital twins, it achieves the technical effect of obtaining a topology optimization scheme that can dynamically adapt to changes in operating conditions and realizing reliable execution control.

[0072] Example 2, based on the same inventive concept as the distribution network-microgrid topology optimization method in the foregoing examples, such as... Figure 2 As shown, this application provides a topology optimization device for a distribution network-microgrid. The device and method embodiments in this application are based on the same inventive concept. The device includes: The constraint setting module 11 is used to perform composite operation analysis based on the distribution network-microgrid and set composite constraints; the collaborative optimization module 12 is used to retrieve the target topology data of the distribution network-microgrid, and perform multi-objective collaborative optimization on the target topology data according to the composite constraints to obtain topology optimization parameters; the switching analysis module 13 is used to perform grid-connected and off-grid switching analysis based on the distribution network-microgrid and construct transient stability control data; the control module 14 is used to control the distribution network-microgrid according to the topology optimization parameters and the transient stability control data to obtain operating stability parameters; the correction module 15 is used to verify the topology optimization parameters by performing grid simulation tests based on the operating stability parameters, iteratively correct the topology optimization parameters according to the verification results until the preset indicators are met, generate the topology optimization scheme, and send it to the distribution network-microgrid for control execution.

[0073] Furthermore, the device is also used to perform the following functions: Multi-timescale operation analysis is performed on the distribution network-microgrid system to obtain multi-scale operation modes. Topology analysis is then performed on the distribution network-microgrid system according to these multi-scale operation modes, setting topology constraints. Operational analysis is also performed on the distribution network-microgrid system according to these multi-scale operation modes, setting operational constraints. Power analysis is then performed on the distribution network-microgrid system according to these multi-scale operation modes, setting power constraints. Based on the topology constraints, operational constraints, and power constraints, stochastic chance constraint programming is performed to construct initial composite constraints. Multi-scenario adaptability analysis is then performed on the initial composite constraints to generate multiple adaptation coefficients. Finally, the initial composite constraints are identified based on these multiple adaptation coefficients to obtain the composite constraints.

[0074] Furthermore, the device is also used to perform the following functions: A topology database of the distribution network-microgrid is constructed, and the target topology data of the distribution network-microgrid is retrieved by traversing the topology database. Multi-objective analysis is performed based on the composite constraints and the target topology data to determine multiple optimization objectives. Based on the composite constraints and the multiple optimization objectives, the target topology data is collaboratively optimized to obtain a combination of topology optimization parameters. The performance of the topology optimization parameter combination is evaluated in multiple scenarios, and the impact of parameter fluctuations is analyzed based on the evaluation results to determine the topology optimization parameters.

[0075] Furthermore, the device is also used to perform the following functions: A topology analysis is performed on the distribution network-microgrid system. Based on the analysis results, the nodes are grouped and identified to determine multiple network nodes. Connection analysis is then performed on these nodes to determine multiple branch connection relationships. These multiple network nodes are then integrated using graph theory to construct a network topology graph. Based on this graph, node-branch association analysis is performed to construct a node-branch association matrix. The distribution network-microgrid system is then matched for operation based on this association matrix, and topology structure selection criteria are set. Finally, a depth-first search is performed on the topology database based on these selection criteria to determine the target topology data for the distribution network-microgrid system.

[0076] Furthermore, the device is also used to perform the following functions: Cluster analysis is performed on the target topology data according to the multiple optimization objectives to determine multiple topology clusters; dimensionality reduction is performed on the multiple topology clusters according to the composite constraints to generate multiple representative parameter solutions; entropy weighting is applied to the multiple representative parameter solutions for multi-attribute evaluation to obtain bidirectional evaluation information of parameter attributes, which includes parameter attribute advantage information and parameter attribute disadvantage information; historical operation logs of the distribution network-microgrid are retrieved for preference analysis to obtain operation preference data; weighted analysis is performed on the multiple representative parameter solutions according to the operation preference data combined with the parameter attribute advantage information and the parameter attribute disadvantage information to determine multiple selected solutions; environmental sensitivity verification is performed on the multiple selected solutions to obtain the topology optimization parameter combination.

[0077] Furthermore, the device is also used to perform the following functions: Historical operating scenarios of the distribution network-microgrid are retrieved to construct a multi-scenario test set; the topology optimization parameter combination is simulated according to the multi-scenario test set to obtain multi-scenario simulation data; multi-scenario performance evaluation is performed based on the multi-scenario simulation data to generate performance evaluation results; global sensitivity analysis is performed based on the performance evaluation results to generate multiple sensitivity indices for interactive response, constructing a parameter-performance interactive response surface; the topology optimization parameter combination is sorted in descending order according to the parameter-performance interactive response surface, key parameters and minor parameters are identified, the minor parameters are filtered out, and the key parameters are output as topology optimization parameters.

[0078] Furthermore, the device is also used to perform the following functions: The topology optimization parameters are configured to the distribution network-microgrid to generate a grid configuration result; multiple control parameters are tuned based on the transient stability control data; multi-level collaborative control is performed according to the grid configuration result using the multiple control parameters: S1: The multi-level distribution network-microgrid includes a grid equipment layer, a grid unit layer, and a grid system layer, wherein the grid equipment layer, the grid unit layer, and the grid system layer have an information interaction mechanism; S2: Based on the grid equipment layer, the inverter is activated to perform local fast control on the distribution network-microgrid to obtain a first control result; S3: Based on the grid unit layer, the microgrid central controller is activated to perform unit coordination control to obtain a second control result; S4: Based on the grid system layer, the distribution network dispatch controller is activated to perform optimized control to obtain a third control result; S5: Based on the first control result, the second control result, and the third control result, collaborative control analysis is performed through the information interaction mechanism to obtain a collaborative control result; the distribution network-microgrid is monitored in real time according to the collaborative control result to generate the operating stability parameters.

[0079] Furthermore, the device is also used to perform the following functions: A digital twin test platform is constructed based on the aforementioned stable operating parameters. The topology optimization parameters are synchronized to the digital twin test platform to perform power grid simulation tests and generate simulation test results. The effectiveness of the topology optimization parameters is evaluated based on the simulation test results: transient stability analysis is performed on the topology optimization parameters based on the simulation test results to obtain a first analysis result; steady-state operation analysis is performed on the topology optimization parameters based on the simulation test results to obtain a second analysis result; reliability verification is performed based on the first and second analysis results to obtain parameter analysis results; parameter verification is performed on the topology optimization parameters according to the parameter analysis results to generate verification results; an adaptive iterative correction algorithm is used to optimize and adjust the topology optimization parameters based on the verification results. When the optimized parameters meet preset indicators, a topology optimization scheme is generated, and the topology optimization scheme is distributed to the distribution network-microgrid system for execution through a security verification mechanism.

[0080] Furthermore, the device is also used to perform the following functions: Based on the verification results, target evaluation is performed, and parameter correction targets are set. Local iterative search is conducted on the topology optimization parameters according to the parameter correction targets to obtain local parameter information to be corrected. Global iterative search is conducted on the topology optimization parameters according to the parameter correction targets to obtain global parameter information to be corrected. Based on the local and global parameter information to be corrected, the topology optimization parameters are corrected and analyzed to determine parameter correction information, which includes correction step size and correction direction information. Multiple rounds of iterative optimization are performed on the topology optimization parameters according to the correction direction information and the correction step size information to obtain optimized parameters, which include parameter correction effects. The parameter correction effects are evaluated in real time until the optimized parameters meet preset comprehensive indicators, generating a topology optimization scheme.

[0081] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, the above description focuses on specific embodiments of this specification. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some implementations, multitasking and parallel processing are possible or may be advantageous.

[0082] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any modifications, equivalent changes, and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A method for optimizing the topology of a distribution network-microgrid, characterized in that, The method includes: Based on the distribution network-microgrid composite operation analysis, composite constraints are set; Retrieve the target topology data of the distribution network-microgrid, and perform multi-objective collaborative optimization on the target topology data according to the composite constraints to obtain the topology optimization parameters; Based on the grid-to-off-grid switching analysis of the distribution network and microgrid, transient stability control data is constructed. The distribution network-microgrid is controlled according to the topology optimization parameters and the transient stability control data to obtain the operating stability parameters; Based on the aforementioned stable operating parameters, a power grid simulation test is conducted to verify the topology optimization parameters. The topology optimization parameters are iteratively corrected according to the verification results until the preset indicators are met. The topology optimization scheme is then generated and sent to the distribution network-microgrid for control and execution. The process includes: verifying the topology optimization parameters through power grid simulation testing based on the operational stability parameters; iteratively correcting the topology optimization parameters according to the verification results until preset indicators are met; generating the topology optimization scheme; and distributing it to the distribution network-microgrid for control execution. A digital twin test platform is constructed based on the aforementioned stable operating parameters. The topology optimization parameters are synchronized to the digital twin test platform to perform power grid simulation tests and generate simulation test results. The effectiveness of the topology optimization parameters is evaluated based on the simulation test results. Based on the simulation test results, transient stability analysis is performed on the topology optimization parameters to obtain the first analysis result; Based on the simulation test results, a steady-state operation analysis of the topology optimization parameters is performed to obtain a second analysis result; Based on the first analysis result and the second analysis result, a reliability test is performed to obtain the parameter analysis result; Based on the parameter analysis results, the topology optimization parameters are validated, and validation results are generated. Based on the verification results, an adaptive iterative correction algorithm is used to optimize and adjust the topology optimization parameters. When the optimization parameters meet the preset indicators, a topology optimization scheme is generated and the topology optimization scheme is sent to the distribution network-microgrid system for execution through a security verification mechanism.

2. The method for optimizing the topology of a distribution network-microgrid as described in claim 1, characterized in that, Based on the distribution network-microgrid composite operation analysis, composite constraints are set, and the methods include: Multi-time-scale operation analysis is performed based on the distribution network-microgrid to obtain multi-scale operation modes; Topology analysis of the distribution network-microgrid is performed according to the multi-scale operation mode, and topology constraints are set. The operation of the distribution network-microgrid is analyzed according to the multi-scale operation mode, and operation constraints are set. Power analysis of the distribution network-microgrid is performed according to the multi-scale operation mode, and power constraints are set. Based on the topological constraints, operational constraints, and power constraints, stochastic chance constraint programming is performed to construct initial composite constraints. Perform multi-scenario adaptive analysis on the initial composite constraints to generate multiple adaptation coefficients; The initial composite constraint is identified according to the multiple fitness coefficients to obtain the composite constraint.

3. The method for optimizing the topology of a distribution network-microgrid as described in claim 1, characterized in that, The method involves retrieving target topology data of a distribution network-microgrid, performing multi-objective collaborative optimization on the target topology data according to the aforementioned composite constraints, and obtaining topology optimization parameters. Construct a topology database of distribution network-microgrid, and traverse the topology database to retrieve the target topology data of distribution network-microgrid; A multi-objective analysis is performed based on the composite constraints and the target topology data to determine multiple objectives to be optimized. Based on the composite constraints, the target topology data is solved collaboratively according to the multiple optimization objectives to obtain the topology optimization parameter combination. The topology optimization parameter combination is evaluated in multiple scenarios, and the impact of parameter fluctuations is analyzed based on the evaluation results to determine the topology optimization parameters.

4. The method for optimizing the topology of a distribution network-microgrid as described in claim 3, characterized in that, Constructing a distribution network-microgrid topology database, and retrieving target topology data of the distribution network-microgrid by traversing the topology database, includes the following methods: The distribution network-microgrid topology analysis is performed, and the analysis results are grouped and identified to determine multiple network nodes. Then, the connection analysis is performed on the multiple network nodes to determine the connection relationship of multiple branches. Based on the aforementioned branch connection relationships, the multiple network nodes are integrated using graph theory to construct a network topology graph; Based on the network topology graph, a node-branch correlation analysis is performed to construct a node-branch correlation matrix of the network topology; The distribution network and microgrid are matched for operation according to the node-branch correlation matrix, and the topology screening conditions are set. Based on the topology filtering criteria, a depth-first search is performed to traverse the topology database to determine the target topology data for the distribution network-microgrid.

5. The method for optimizing the topology of a distribution network-microgrid as described in claim 3, characterized in that, Based on the composite constraints, the target topology data is collaboratively optimized according to the multiple optimization objectives to obtain a combination of topology optimization parameters. The method includes: Cluster analysis is performed on the target topology data according to the multiple optimization objectives to determine multiple topology clusters; The multiple topological clusters are dimensionality-reduced according to the composite constraints to generate multiple representative parameter solutions. By traversing the multiple representative parameter solutions, entropy weight is applied to perform multi-attribute evaluation to obtain bidirectional evaluation information of parameter attributes. The bidirectional evaluation information includes parameter attribute advantage information and parameter attribute disadvantage information. Historical operation logs of the distribution network and microgrid are retrieved for preference analysis to obtain operation preference data; Based on the operational preference data, combined with the parameter attribute advantage information and the parameter attribute disadvantage information, a weighted analysis is performed on the multiple representative parameter solutions to determine multiple selected solutions; The selected solutions are subjected to environmental sensitivity verification to obtain the optimal combination of topology parameters.

6. The method for optimizing the topology of a distribution network-microgrid as described in claim 3, characterized in that, The topology optimization parameter combination is subjected to multi-scenario performance evaluation, and the parameter fluctuation impact is analyzed based on the evaluation results to determine the topology optimization parameters. The method includes: Retrieve historical operating scenarios of the distribution network and microgrid to construct a multi-scenario test set; The topology optimization parameter combination is simulated according to the multi-scenario test set to obtain multi-scenario simulation data; Based on the multi-scenario simulation data, multi-scenario performance evaluation is performed, and performance evaluation results are generated. Based on the performance evaluation results, a global sensitivity analysis is performed to generate multiple sensitivity indices for interactive response, and a parameter-performance interactive response surface is constructed. The topology optimization parameter combinations are sorted in descending order according to the parameter-performance interaction response surface. Key parameters and minor parameters are identified, the minor parameters are filtered out, and the key parameters are output as topology optimization parameters.

7. The method for optimizing the topology of a distribution network-microgrid as described in claim 1, characterized in that, The distribution network-microgrid is controlled according to the topology optimization parameters and the transient stability control data to obtain operational stability parameters. The method includes: The topology optimization parameters are configured to the distribution network-microgrid to generate the power grid configuration results. Multiple control parameters are tuned based on the transient stability control data; Multi-level coordinated control is performed by executing the multiple control parameters according to the power grid configuration results: S1: The multi-level distribution network-microgrid includes a power grid equipment layer, a power grid unit layer, and a power grid system layer, wherein the power grid equipment layer, the power grid unit layer, and the power grid system layer have an information interaction mechanism; S2: Based on the power grid equipment layer, activate the inverter to perform local fast control on the distribution network-microgrid and obtain the first control result; S3: Based on the power grid unit layer, activate the microgrid central controller to perform unit coordination control and obtain the second control result; S4: Based on the power grid system layer, activate the distribution network dispatch controller to perform optimized control and obtain the third control result; S5: Based on the first control result, the second control result, and the third control result, perform collaborative control analysis through an information interaction mechanism to obtain collaborative control results; Based on the results of the coordinated control, the distribution network-microgrid is monitored in real time to generate the operational stability parameters.

8. The method for optimizing the topology of a distribution network-microgrid as described in claim 1, characterized in that, Based on the verification results, an adaptive iterative correction algorithm is used to optimize and adjust the topology optimization parameters. When the optimization parameters meet the preset indicators, a topology optimization scheme is generated. The method includes: Based on the verification results, evaluate the target and set parameters to correct the target. According to the parameter correction target, a local iterative search is performed on the topology optimization parameters to obtain local parameter information to be corrected. A global iterative search is performed on the topology optimization parameters according to the parameter correction target to obtain global parameter information to be corrected. Based on the local parameter information to be corrected and the global parameter information to be corrected, the topology optimization parameters are corrected and analyzed to determine the parameter correction information, which includes correction step information and correction direction information. The topology optimization parameters are iteratively optimized in multiple rounds according to the correction direction information and the correction step size information to obtain optimization parameters, which include the parameter correction effect. The effect of the parameter correction is evaluated in real time until the optimized parameters meet the preset comprehensive index, and a topology optimization scheme is generated.

9. A topology optimization device for a distribution network-microgrid, characterized in that, The apparatus is used to execute the topology optimization method for a distribution network-microgrid as described in any one of claims 1-8, and the apparatus comprises: The constraint setting module is used to set composite constraint conditions for combined operation analysis based on distribution network-microgrid; The collaborative optimization module is used to retrieve the target topology data of the distribution network-microgrid, and perform multi-objective collaborative optimization on the target topology data according to the composite constraints to obtain the topology optimization parameters. The switching analysis module is used to perform grid-connected and off-grid switching analysis based on the distribution network-microgrid and to construct transient stability control data; The control module is used to control the distribution network-microgrid according to the topology optimization parameters and the transient stability control data to obtain the operating stability parameters; The correction module is used to verify the topology optimization parameters by conducting power grid simulation tests based on the operating stability parameters, iteratively correct the topology optimization parameters according to the verification results until the preset indicators are met, generate the topology optimization scheme, and send it to the distribution network-microgrid for control execution.

Images