A method for searching a backbone network framework of an urban power transmission network for extreme survival

Abstract

This invention relates to the field of urban power grid technology and discloses a method for searching the backbone network of an urban transmission network for extreme survival. The method includes: acquiring urban transmission network topology data, guaranteed power output prediction, and load prediction data; selecting the optimal power supply path for each important load node based on the above data to form a lifeline channel set; constructing and solving a backbone network search model to obtain an initial network structure with the goal of minimizing the network size, using the lifeline channel set as a mandatory constraint, and using an embedded source-grid-load-storage collaborative operation simulation as an operational constraint; performing connectivity judgment and repair on the initial network structure to obtain candidate network structures; and performing DC power flow verification, adjusting the power supply parameters and iteratively solving if overload occurs. This application achieves adaptive search of the backbone network structure in off-grid conditions through the embedded integration of mandatory lifeline channel constraints and source-grid-load-storage collaborative operation simulation, solving the problem that existing backbone networks cannot support independent operation under extreme survival conditions, and improving the power supply guarantee capability of urban power grids under extreme disasters.

Description

Technical Field

[0001] This invention relates to the technical field of urban power grids, and in particular to a method for searching the backbone network of urban power transmission networks for extreme survival. Background Technology

[0002] Currently, the frequency of major power outages is showing a significant upward trend. Extreme natural disasters, such as severe weather, have become the dominant factor triggering power system outages. Under the influence of extreme disasters, urban power grids may become detached from the upper-level grid and operate independently, relying on their own power generation, grid, load, and storage resources. As the main load center of the power system, urban power grids have high load density, concentrated important loads, and undertake multiple important functions, including economic ones. Their safe and stable operation is crucial to national economic development.

[0003] Therefore, conducting research on backbone grid structures adapted to off-grid operation of urban power grids and establishing a sound new power system security defense system for urban power grids are of great significance for the safe and stable operation of urban power grids in the face of extreme disasters.

[0004] In extreme scenarios, the existing backbone network cannot support the independent operation of local power grids and the upper-level main grid after they are disconnected from the grid. It is difficult to meet the differentiated power supply needs of loads. The intermittent power supply to loads of different importance will lead to the difficulty in selecting and protecting the power supply lifeline channels for some key loads that require uninterrupted power supply. The critical power supply path is difficult to lock under end disasters, so the power supply of important loads cannot be guaranteed. Summary of the Invention

[0005] To address the aforementioned technical problems, this application provides a method for searching urban power grid backbone structures for extreme survival.

[0006] Firstly, the aforementioned inventive objective of this application is achieved through the following technical solutions:

[0007] A method for searching the backbone network of an urban power transmission network under extreme survival requirements, the method comprising the following steps:

[0008] Acquire topology data of the urban power transmission network, extreme survivability output prediction data of backup power sources, and load prediction data of each load node;

[0009] Based on the topology data, the output prediction data, and the load prediction data, the optimal power supply path connecting each important load node to the backup power supply node is selected, and all selected optimal power supply paths are combined to form a set of lifeline channels that must be guaranteed.

[0010] With the goal of minimizing the scale of the grid structure, the lifeline channel set as a mandatory constraint, and the source-grid-load-storage collaborative operation simulation covering the entire extreme survival time period as an operational constraint, a backbone grid search model is constructed and solved to obtain the initial backbone grid structure.

[0011] The initial backbone network is assessed for connectivity. If there are disconnected areas, connecting lines are added to repair the connectivity, resulting in a candidate backbone network with connectivity.

[0012] The DC power flow is verified on the candidate backbone network. If there is no line overload in any time period, the candidate backbone network is determined as the final backbone network. If there is line overload, the load supply parameters in the operation constraints are adjusted, and the backbone network search model is reconstructed and solved.

[0013] By adopting the above technical solution, this application provides a complete data foundation for subsequent grid search by simultaneously acquiring three types of basic data: topology, power output prediction, and load prediction. On this basis, the optimal power supply path is quantitatively selected for each important load node, forming a lifeline channel set, which is incorporated into the search model as a mandatory constraint. This ensures from the source that the power supply path of important loads will not be cut off under any circumstances. Furthermore, the source-grid-load-storage coordinated operation simulation covering the entire extreme survival period is embedded as the operation constraint of the backbone grid search model. This makes the searched grid not only structurally simplified, but more importantly, it has the actual ability to operate independently and continuously in an off-grid state. This effectively solves the technical problem that existing backbone grids cannot support independent operation under extreme survival conditions. Through the iterative optimization mechanism of connectivity repair and DC power flow verification, the final output backbone grid is ensured to be both connected and reliable, achieving the optimal balance between minimizing grid size and maximizing survival guarantee capability. This significantly improves the resilience of urban power grids under extreme disasters and the power supply reliability of important loads.

[0014] In a preferred embodiment, this application can be further configured such that: the acquisition of urban power grid topology data, critical survivability power output prediction data for backup power sources, and load prediction data for each load node specifically includes:

[0015] The connection relationships between nodes and lines of the urban power transmission network, line impedance parameters, and line capacity parameters are obtained as the topology data.

[0016] The maximum available active power output of all backup power sources during each time period within the extreme survival time period is obtained and used as the output prediction data.

[0017] The predicted active power demand of each important load node and each ordinary load node in each time period during the extreme survival period is obtained as the load prediction data.

[0018] By adopting the above technical solutions, data acquisition is refined into three dimensions: topology, power output, and load demand. The physical connection characteristics of the power grid, the power generation capacity of the power sources, and the power consumption demand of the loads are extracted respectively. The line impedance parameters in the topology data provide an accurate mathematical model basis for subsequent electrical distance calculation and power flow analysis. The line capacity parameters provide a clear benchmark for judging line overload. The power output prediction data fully considers the output characteristics of guaranteed power sources such as coal-fired units and gas-fired units, providing realistic power source boundary conditions for the simulation of source-grid-load-storage coordinated operation. The load prediction data distinguishes between important loads and ordinary loads, laying the data foundation for subsequent differentiated supply guarantee strategies. The refined acquisition of these three types of data ensures that the entire backbone network search method is based on accurate and comprehensive data, avoiding search bias caused by missing or inaccurate data.

[0019] In a preferred embodiment, this application can be further configured as follows: Based on the topology data, the output prediction data, and the load prediction data, selecting the optimal power supply path connecting each critical load node to the guaranteed power supply node specifically includes:

[0020] For each critical load node, calculate its electrical distance index, geographical distance index, and average power flow load rate index between it and each backup power supply node;

[0021] The electrical distance index, the geographical distance index, and the average power flow load rate index are weighted and summed to obtain a comprehensive evaluation value between the important load node and the backup power supply node;

[0022] Select the reliable power supply node with the smallest comprehensive evaluation value corresponding to the important load node as the target power supply node, and determine the shortest electrical path connecting the important load node and the target power supply node as the optimal power supply path for the important load node.

[0023] The optimal power supply paths for all critical load nodes are aggregated to generate the lifeline channel set.

[0024] By adopting the above technical solution, and simultaneously considering three indicators—electrical distance, geographical distance, and power flow load rate—the quality of each power supply path was comprehensively evaluated from three dimensions: electrical performance, disaster risk, and operational margin. The electrical distance indicator ensures high power transmission efficiency and low loss for the selected path; the geographical distance indicator reduces the path's exposure risk in extreme disasters; and the power flow load rate indicator ensures that the path still has sufficient transmission margin under peak load. The entropy weight method is used to objectively assign weights to the three indicators, avoiding the bias of subjective judgment, making the comprehensive evaluation value more scientific and reasonable. This quantitative selection method changes the traditional practice of selecting power supply paths based on experience or simple rules, and achieves precise locking of lifeline channels, providing a reliable first constraint condition for subsequent backbone network search.

[0025] In a preferred embodiment, this application can be further configured as follows: The process of constructing and solving a backbone network search model with the goal of minimizing the network size, using the lifeline channel set as a mandatory constraint, and employing an embedded source-grid-load-storage collaborative operation simulation covering the entire extreme survival time period as an operational constraint, to obtain an initial backbone network, specifically includes:

[0026] The target layer of the backbone network search model is established with the objective function of minimizing the total number of lines contained in the backbone network.

[0027] All lines in the lifeline channel set are set as mandatory lines and added as the first constraint to the backbone network search model.

[0028] A source-grid-load-storage coordinated operation simulation covering all time periods within the extreme survival time period is constructed, and the feasibility of the operation simulation is added to the backbone network search model as a second constraint. The operation simulation includes: start-up and shutdown and output constraints of the guaranteed power supply, active power flow transmission constraints of the line, differentiated supply guarantee constraints of load nodes, and charging, discharging and state of charge constraints of the energy storage device.

[0029] The backbone network search model is solved using a mixed-integer linear programming algorithm to obtain the initial backbone network that satisfies all constraints and has the minimum number of lines.

[0030] By adopting the above technical solutions, with the goal of minimizing the total number of lines, the scale of the backbone network is effectively reduced, construction and maintenance costs are decreased, and the exposed area of ​​the network in extreme disasters is reduced. The set of lifeline channels is used as the first constraint condition to ensure that the power supply path of important loads is forcibly included in the network. Most importantly, the feasibility of source-grid-load-storage coordinated operation simulation is used as the second constraint condition, so that the searched network structure is not only structurally simplified, but also able to support the coordinated operation of source, grid, load and storage links throughout the entire extreme survival period. This design fundamentally solves the technical bottleneck that the existing backbone network cannot continue to operate independently after being disconnected from the grid. The mixed integer linear programming algorithm is used to solve the problem, ensuring the solution efficiency and global optimality of the results. An innovative backbone network search model is constructed, the core of which lies in the organic combination of the goal of "minimizing the scale of the network" and the constraint of "feasibility of extreme survival operation".

[0031] In a preferred embodiment, this application can be further configured such that the differentiated supply guarantee constraints of the load nodes specifically include:

[0032] For a node belonging to the set of critical load nodes, the active power it guarantees is equal to its predicted active power demand in each time period of the critical survival time.

[0033] For ordinary load nodes that do not belong to the set of important load nodes, the active power they are guaranteed to supply during each period of the extreme survival time shall not be less than their predicted active power demand multiplied by the preset guarantee rate coefficient.

[0034] By adopting the above technical solutions, a rigid supply guarantee strategy is used for critical load nodes, requiring that their guaranteed power supply strictly equal the predicted demand. This ensures that key users such as government agencies, hospitals, and data centers can have uninterrupted power supply under extreme disasters, fulfilling the social responsibility of the power grid company. For ordinary load nodes, a flexible supply guarantee strategy is adopted, allowing load reduction within a certain range. However, a minimum guarantee level is set through the supply guarantee rate coefficient λ, avoiding the social impact of a complete power outage. This differentiated supply guarantee strategy achieves the optimal allocation of limited resources under the constraints of limited power resources and grid capacity. It guarantees key areas while also taking into account general needs, and is a concrete manifestation of the concept of extreme survival operation.

[0035] In a preferred embodiment, this application can be further configured as follows: The initial backbone network is subjected to connectivity assessment; if disconnected regions exist, connecting lines are added to repair connectivity, resulting in a candidate connected backbone network. Specifically, this includes:

[0036] The Dijkstra algorithm is used to check and search all connected regions in the initial backbone network to determine whether the number of connected regions is equal to 1 and whether all important load nodes are in the same connected region.

[0037] If the number of connected regions is equal to 1 and all important load nodes are connected, the initial backbone network is determined to be connected, and the process proceeds directly to the DC power flow verification step; if the number of connected regions is not equal to 1 or there are important load nodes that are not connected, the initial backbone network is determined to be not connected, and all disconnected isolated subgraphs and the important load nodes contained therein are identified.

[0038] With the goal of minimizing the total length of the added tie lines, the minimum spanning tree algorithm is used to add tie lines from the original transmission network topology between the disconnected isolated subgraphs, so that all important load nodes are connected into a connected whole, wherein the intermediate nodes through which the tie lines pass include non-important nodes.

[0039] The initial backbone network after adding the connecting lines is determined as the candidate backbone network for connectivity.

[0040] By adopting the above technical solutions, this application addresses connectivity issues that may arise during backbone network search by employing a connectivity judgment and repair strategy centered on critical load nodes. Since the search model aims to minimize the total number of lines, some critical load nodes may lack direct connection paths, resulting in isolated subgraphs. By using Dijkstra's algorithm to quickly detect connected regions, connectivity judgment and shortest path identification can be completed simultaneously, improving detection efficiency. More importantly, this application optimizes the connectivity judgment standard from "all nodes connected" to "all critical load nodes connected," avoiding excessive addition of tie lines due to non-critical node disconnections. This maximizes control over the scale and economic cost of the backbone network while ensuring the power supply reliability of critical loads. When a disconnection is detected, a minimum spanning tree algorithm is used to repair it with the goal of minimizing the total tie line length, explicitly allowing tie lines to pass through non-critical nodes, further reducing the length of newly added lines and the risk of disaster exposure. This connectivity repair mechanism, guided by critical load nodes, ensures that the final backbone network is a complete network capable of interconnecting all critical loads, providing the necessary topological foundation for subsequent power flow verification and actual operation.

[0041] In a preferred embodiment, this application can be further configured as follows: The DC power flow verification of the connected candidate backbone network is performed. If there is no line overload in any time period, the connected candidate backbone network is determined as the final backbone network. If there is line overload, the load guarantee parameters in the operating constraints are adjusted, and the backbone network search model is reconstructed and solved. Specifically, this includes:

[0042] Based on the topology and line parameters of the connected candidate backbone network, and the power output and load levels of the source-grid-load-storage coordinated operation simulation at each time period, DC power flow calculation is performed time-by-time to obtain the active power flow value of each line at each time period.

[0043] The active power flow value of each line in each time period is compared with its corresponding line capacity parameter. If the active power flow value of all lines in all time periods is not greater than its corresponding line capacity parameter, the DC power flow verification is determined to be passed, and the connected candidate backbone network is output as the final backbone network.

[0044] If the active power flow value of any line in any time period is greater than its corresponding line capacity parameter, the DC power flow verification is determined to fail, and the preset supply guarantee rate coefficient is reduced by a preset step size. Then, the process returns to the step of constructing the backbone network search model and solving it.

[0045] By adopting the above technical solution and performing DC power flow calculations time by time, the load status of each line in the backbone network can be comprehensively examined throughout the entire extreme survival period. This avoids safety hazards caused by overload in a certain period. When an overload is detected, the total load demand is reduced by lowering the supply guarantee coefficient of ordinary loads. Then, the backbone network is searched again. This iterative optimization mechanism can gradually approach the optimal solution while meeting safety constraints. Through multiple iterations, the final backbone network can meet the requirements of extreme survival operation without overload, achieving a dual guarantee of safety and economy.

[0046] In a preferred embodiment, this application can be further configured such that the source-grid-load-storage collaborative operation simulation also includes network switching constraints, which are used to limit the upper limit of the number of lines that can be actively disconnected within the extreme survival time period.

[0047] By adopting the above technical solution, network switching constraints are introduced into the source-grid-load-storage coordinated operation simulation, providing appropriate topology adjustment flexibility for extreme survival operation. By introducing line operation state variables, it is allowed to actively disconnect some lines during operation according to changes in power output and load to optimize power flow distribution or isolate faults. However, at the same time, an upper limit is set on the number of lines that can be disconnected to prevent frequent or large-scale line disconnection from damaging the topology stability of the system and avoid cascading failures caused by excessive switching. This "limited flexibility" design concept enables the backbone network to make appropriate topology adjustments according to actual operating needs during extreme survival, while maintaining basic network structure stability, ensuring the continuity and reliability of power supply.

[0048] Secondly, the above-mentioned objective of this application is achieved through the following technical solution:

[0049] An electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the above-described method for searching the backbone of an urban power transmission network for extreme survival.

[0050] Thirdly, the above-mentioned objectives of this application are achieved through the following technical solutions:

[0051] A computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the above-described method for searching the backbone of an urban power transmission network for extreme survival.

[0052] In summary, this application includes at least one of the following beneficial technical effects:

[0053] 1. This application provides a complete data foundation for subsequent grid search by simultaneously acquiring three types of basic data: topology, power output prediction, and load prediction. Based on this, the optimal power supply path is quantitatively selected for each important load node and a lifeline channel set is formed. This set is incorporated into the search model as a mandatory constraint, ensuring from the source that the power supply path of important loads will not be cut off under any circumstances. Furthermore, the source-grid-load-storage collaborative operation simulation covering the entire extreme survival period is embedded as the operation constraint of the backbone grid search model. This makes the searched grid not only structurally streamlined, but more importantly, it has the actual ability to operate independently and continuously in an off-grid state. This effectively solves the technical problem that the existing backbone grid cannot support independent operation under extreme survival conditions. Through the iterative optimization mechanism of connectivity repair and DC power flow verification, the final output backbone grid is ensured to be both connected and reliable. It achieves the optimal balance between minimizing the grid size and maximizing survival guarantee capability, significantly improving the resilience of urban power grids under extreme disasters and the power supply reliability of important loads.

[0054] 2. By simultaneously considering three indicators—electrical distance, geographical distance, and power flow load rate—the quality of each power supply path was comprehensively evaluated from three dimensions: electrical performance, disaster risk, and operational margin. The electrical distance indicator ensured high power transmission efficiency and low loss for the selected path; the geographical distance indicator reduced the path's exposure risk in extreme disasters; and the power flow load rate indicator ensured that the path still had sufficient transmission margin under peak load. The entropy weight method was used to objectively assign weights to the three indicators, avoiding the bias of subjective judgment and making the comprehensive evaluation value more scientific and reasonable. This quantitative selection method changed the traditional practice of selecting power supply paths based on experience or simple rules, and achieved precise locking of lifeline channels, providing a reliable first constraint condition for subsequent backbone network search.

[0055] 3. Since the search model aims to minimize the total number of lines, it may lead to isolated subgraphs due to a lack of direct connection paths between some important load nodes. By employing Dijkstra's algorithm to quickly detect connected regions, connectivity judgment and shortest path identification can be completed simultaneously, improving detection efficiency. More importantly, this application optimizes the connectivity judgment standard from "all nodes are connected" to "all important load nodes are connected," avoiding the excessive addition of tie lines due to non-important nodes being disconnected. This maximizes the control of the backbone network size and economic cost while ensuring the power supply reliability of important loads. When a disconnection is detected, the minimum spanning tree algorithm is used to repair it with the goal of minimizing the total length of tie lines, and tie lines are explicitly allowed to pass through non-important nodes, further reducing the length of newly added lines and the risk of disaster exposure. This connectivity repair mechanism, guided by important load nodes, ensures that the final backbone network is a complete network that can guarantee the interconnection of all important loads, providing the necessary topological foundation for subsequent power flow verification and actual operation.

[0056] 4. By performing DC power flow calculations time-by-time, the load status of each line in the backbone network can be comprehensively examined throughout the entire extreme survival period, avoiding safety hazards caused by overload in a certain period. When an overload is detected, the total load demand is reduced by lowering the supply guarantee coefficient of ordinary loads, and then the backbone network is searched again. This iterative optimization mechanism can gradually approach the optimal solution while meeting safety constraints. Through multiple iterations, the final backbone network can meet the requirements of extreme survival operation without overload, achieving a dual guarantee of safety and economy. Attached Figure Description

[0057] Figure 1 This is a flowchart of a method for searching the backbone network of an urban power transmission network for extreme survival, according to one embodiment of this application.

[0058] Figure 2 This is a flowchart illustrating the implementation of step S10 in a method for searching the backbone network of an urban power transmission network for extreme survival, as described in one embodiment of this application.

[0059] Figure 3 This is a flowchart illustrating the implementation of step S20 in a method for searching the backbone network of an urban power transmission network for extreme survival, as described in one embodiment of this application.

[0060] Figure 4 This is a flowchart illustrating the implementation of step S30 in a method for searching the backbone network of an urban power transmission network for extreme survival, as described in one embodiment of this application.

[0061] Figure 5 This is a flowchart illustrating the implementation of step S33 in a method for searching the backbone network of an urban power transmission network for extreme survival, as described in one embodiment of this application.

[0062] Figure 6 This is a flowchart illustrating the implementation of step S40 in a method for searching the backbone network of an urban power transmission network for extreme survival, as described in one embodiment of this application.

[0063] Figure 7 This is a flowchart illustrating the implementation of step S50 in a method for searching the backbone network of an urban power transmission network for extreme survival, as described in one embodiment of this application.

[0064] Figure 8 This is a schematic diagram of an electronic device according to an embodiment of this application.

[0065] Figure 9 This is a modified IEEE 30-node test system diagram according to one embodiment of this application.

[0066] Figure 10 This is a comprehensive evaluation index diagram of the shortest power supply path for node pairs in one embodiment of this application.

[0067] Figure 11 This is a backbone network diagram at a specific moment in one embodiment of this application.

[0068] Figure 12 This is a diagram of the backbone network structure that ensures survival for one day under extreme conditions, according to one embodiment of this application.

[0069] Figure 13 This is a diagram of the backbone network structure that ensures survival for 3 days under extreme conditions, according to one embodiment of this application.

[0070] Figure 14 This is a diagram of the backbone structure that ensures survival for 5 days under extreme conditions, according to one embodiment of this application. Detailed Implementation

[0071] The present application will be further described in detail below with reference to the accompanying drawings.

[0072] In one embodiment, such as Figure 1 As shown, this application discloses a method for searching the backbone network of an urban power transmission network for extreme survival, which specifically includes the following steps:

[0073] S10: Obtain topology data of the urban power transmission network, predicted ultimate survivability output data of backup power sources, and predicted load data of each load node.

[0074] In this embodiment, topology data refers to a set of parameters describing the physical connections of the urban power grid, including the connection relationships between nodes and lines, line impedance parameters, and line capacity parameters; backup power sources refer to the main power sources that can continuously and stably output power under extreme disasters, mainly including coal-fired units, gas-fired units, etc.; extreme survival output prediction data refers to the maximum available active power output of all backup power sources in each time period within a preset extreme survival time period; load prediction data refers to the predicted active power demand of each important load node and ordinary load node in each time period within the extreme survival time period.

[0075] Specifically, the connection relationships between nodes and lines, line impedance parameters, and line capacity parameters of the urban power transmission network are acquired through power dispatch automation systems or SCADA systems as topology data. Simultaneously, based on weather forecast data and historical power output data, the maximum available active power output of all backup power sources during each time period within the extreme survival timeframe is predicted, generating power output forecast data. Furthermore, based on historical load data and user classification information, the predicted active power demand of each important load node and each ordinary load node during each time period within the extreme survival timeframe is obtained as load forecast data. These three types of data collectively form the basis for subsequent calculations.

[0076] S20: Based on the topology data, the output prediction data, and the load prediction data, select the optimal power supply path to connect to the backup power supply node for each important load node, and combine all the selected optimal power supply paths to form a set of lifeline channels that must be guaranteed.

[0077] In this embodiment, the lifeline channel refers to the critical power supply path between the backup power supply node and the important load node. These paths must be prioritized for protection under extreme disasters to ensure uninterrupted power supply to important loads. The selection of the optimal power supply path comprehensively considers three dimensions: electrical distance, geographical distance, and power flow load rate.

[0078] Specifically, for each critical load node, the electrical distance, geographical distance, and average power flow load rate of the lines along the path between it and each backup power source node are first calculated. The electrical distance reflects the impact of the line's reactance parameters on power flow, the geographical distance reflects the physical length of the line and the degree of exposure to disaster risks in the geographical environment, and the power flow load rate reflects the line's load level under peak load conditions. Next, the entropy weight method is used to objectively assign weights to these three indicators, calculating the weight coefficients for each indicator. Then, the electrical distance, geographical distance, and average power flow load rate are weighted and summed to obtain a comprehensive evaluation value between the critical load node and the backup power source node. The backup power source node with the smallest comprehensive evaluation value corresponding to the critical load node is selected as the target power source node, and the shortest electrical path connecting the critical load node and the target power source node is determined as the optimal power supply path for that critical load node. Finally, the optimal power supply paths for all critical load nodes are summarized to generate a lifeline channel set, which will serve as mandatory constraints in the subsequent backbone network search.

[0079] S30: With the goal of minimizing the scale of the grid structure, the lifeline channel set as a mandatory constraint, and the source-grid-load-storage collaborative operation simulation covering the entire extreme survival time period as an operational constraint, a backbone grid search model is constructed and solved to obtain the initial backbone grid structure.

[0080] In this embodiment, the backbone network search model is a mixed-integer linear programming model, the goal of which is to find the backbone network with the fewest lines while satisfying all constraints. The lifeline channel set serves as the first constraint, ensuring that the power supply paths of all important loads are forcibly included in the backbone network; the source-grid-load-storage coordinated operation simulation serves as the second constraint, ensuring that the searched backbone network can operate stably and continuously within the extreme survival time period.

[0081] Specifically, the target layer of the backbone network search model is first established with the objective function of minimizing the total number of lines in the backbone network. Then, all lines in the lifeline channel set are designated as mandatory lines and added to the model as the first constraint. Next, a source-grid-load-storage coordinated operation simulation covering all time periods within the extreme survival time is constructed, and the feasibility of the operation simulation is added to the model as the second constraint. This operation simulation comprehensively considers the start-up and shutdown constraints and output constraints of guaranteed power sources, the active power flow transmission constraints of lines, the differentiated supply constraints of load nodes, and the charging, discharging, and state of charge constraints of energy storage devices. The differentiated supply constraints mean that for important load nodes, the guaranteed active power supply equals their predicted active power demand in each time period of the extreme survival time; for ordinary load nodes, the guaranteed active power supply is not less than their predicted active power demand multiplied by a preset supply rate coefficient in each time period of the extreme survival time. In addition, the operation simulation also includes network switching constraints to limit the upper limit of the number of lines that can be actively disconnected within the extreme survival time to ensure the operational stability of the backbone network. Finally, the mixed-integer linear programming algorithm is used to solve the model, and the initial backbone network with the minimum number of lines that satisfies all constraints is obtained.

[0082] S40: Perform connectivity judgment on the initial backbone network. If there are disconnected areas, add connecting lines to repair connectivity and obtain a connected candidate backbone network.

[0083] In this embodiment, connectivity refers to whether there is at least one path connecting all nodes in the backbone network. Since the initial backbone network obtained by solving S30 may contain multiple isolated subgraphs that are not connected to each other, some nodes may not be able to exchange power with other nodes through the backbone network, so connectivity repair is required.

[0084] Specifically, Dijkstra's algorithm is used to examine all connected regions in the initial backbone network, determining whether the number of connected regions is equal to 1 and whether all important load nodes are within the same connected region. Dijkstra's algorithm can simultaneously perform connectivity detection and shortest path search, resulting in higher efficiency. If the number of connected regions is equal to 1 and all important load nodes are connected, the initial backbone network is determined to be connected, and the process proceeds directly to step S50 for DC power flow verification. If the number of connected regions is not equal to 1 or some important load nodes are not connected, the initial backbone network is determined to be disconnected, and all disconnected isolated subgraphs and their contained important load nodes are identified. Then, with the goal of minimizing the total length of added tie lines, the minimum spanning tree algorithm is used to add tie lines from the original transmission network topology between disconnected isolated subgraphs, connecting all important load nodes into a connected whole. Intermediate nodes along the tie lines can include non-important nodes. Finally, the initial backbone network after adding tie lines is determined as a candidate connected backbone network.

[0085] S50: Perform DC power flow verification on the connected candidate backbone network. If there is no line overload in any time period, the connected candidate backbone network is determined as the final backbone network. If there is line overload, adjust the load supply parameters in the operation constraints, and return to rebuild and solve the backbone network search model.

[0086] In this embodiment, DC power flow verification is a simplified power flow calculation method used to quickly determine whether the active power flow of each line in the backbone network exceeds its capacity limit. If overload exists, it indicates that the current backbone network cannot safely bear the power flow distribution in the extreme survival operation simulation, and it is necessary to reduce the load demand by adjusting the load supply parameters, and then re-search for the backbone network.

[0087] Specifically, based on the topology and line parameters of the connected candidate backbone network, and the power output and load levels of the source-grid-load-storage coordinated operation simulation at each time period, DC power flow calculations are performed time-by-time to obtain the active power flow value of each line at each time period. The active power flow value of each line at each time period is compared with its corresponding line capacity parameter. If the active power flow value of all lines at all time periods is not greater than its corresponding line capacity parameter, the DC power flow verification is deemed to have passed, and the connected candidate backbone network is output as the final backbone network. If the active power flow value of any line at any time period is greater than its corresponding line capacity parameter, the DC power flow verification is deemed to have failed, and the preset normal load supply guarantee rate coefficient is reduced by a preset step size. Then, the process returns to step S30 to reconstruct and solve the backbone network search model. Through this iterative optimization mechanism, a safe and reliable backbone network that can meet the requirements of extreme survival operation without overload is finally obtained.

[0088] In this embodiment, this application provides a complete data foundation for subsequent grid search by simultaneously acquiring three types of basic data: topology, power output prediction, and load prediction. Based on this, the optimal power supply path is quantitatively selected for each important load node, forming a lifeline channel set, which is incorporated into the search model as a mandatory constraint. This ensures from the source that the power supply path of important loads will not be cut off under any circumstances. Furthermore, the source-grid-load-storage coordinated operation simulation covering the entire extreme survival period is embedded as the operation constraint of the backbone grid search model. This makes the searched grid not only structurally streamlined, but more importantly, it has the actual ability to operate independently and continuously in an off-grid state. This effectively solves the technical problem that existing backbone grids cannot support independent operation under extreme survival conditions. Through the iterative optimization mechanism of connectivity repair and DC power flow verification, the final output backbone grid is ensured to be both connected and reliable, achieving the optimal balance between minimizing grid size and maximizing survival guarantee capability. This significantly improves the resilience of urban power grids under extreme disasters and the power supply reliability of important loads.

[0089] In one embodiment, such as Figure 2 As shown, in step S10, the topology data of the urban power transmission network, the ultimate survival output prediction data of the backup power supply, and the load prediction data of each load node are obtained, specifically including:

[0090] S11: Obtain the connection relationship between nodes and lines of the urban power transmission network, line impedance parameters, and line capacity parameters as the topology data.

[0091] In this embodiment, the connection relationship between nodes and lines describes the topology of the power grid, that is, which nodes are connected by lines; the line impedance parameters include resistance and reactance, which determine the electrical distance and power flow distribution characteristics of the line; the line capacity parameter refers to the maximum allowable transmission power of the line, which is the basis for judging whether the line is overloaded.

[0092] Specifically, the network topology model file of the urban power transmission network is exported through the Energy Management System (EMS) or Supervisory Control System (SCADA) of the power dispatch center. This file contains the numbers and connections of all substation nodes and transmission lines, as well as impedance parameters such as resistance, reactance, and susceptance for each line. Simultaneously, the long-term allowable current carrying capacity of each line is obtained as the line capacity parameter. After formatting this data, a standard topology dataset is generated for subsequent electrical distance calculations and power flow analysis.

[0093] S12: Obtain the maximum available active power output of all guaranteed power sources in each time period within the extreme survival time period, and use it as the output prediction data.

[0094] In this embodiment, the backup power source refers to the main power source that can continuously and stably output power under extreme disaster conditions, mainly including coal-fired units and gas-fired units. This type of power source is not affected by weather conditions and can provide reliable basic power support. The extreme survival time period refers to the length of time that the urban power grid needs to operate independently after being disconnected from the upper-level main grid, usually set to 1 day, 3 days or 5 days, etc.

[0095] Specifically, based on weather forecast data, fuel supply security plans, and unit maintenance plans, the maximum available active power output of each backup power source is predicted within a preset extreme survival period, typically 15 minutes or 1 hour per dispatch period. For coal-fired units, their minimum stable output, maximum technical output, and ramp rate limitations are considered; for gas-fired units, their fuel supply constraints and rapid start-up and shutdown characteristics are considered. These predicted data are organized into a time series to generate an output prediction data matrix, where rows represent different time periods and columns represent different backup power source nodes.

[0096] S13: Obtain the predicted active power demand of each important load node and each ordinary load node in each time period within the extreme survival time period, as the load prediction data.

[0097] In this embodiment, critical load nodes refer to user nodes that undertake important political, economic, and livelihood functions, such as government agencies, hospitals, and data centers. These loads must be supplied with uninterrupted power during extreme survival periods. Ordinary load nodes refer to nodes where general industrial, commercial, and residential users are located, and power supply may be appropriately reduced in emergency situations.

[0098] Specifically, based on historical load data, meteorological factors, holiday information, and user classification tags, time series forecasting methods or machine learning algorithms are used to predict the active power demand of each important load node and each ordinary load node in each time period within the extreme survival period. For important load nodes, the predicted value is considered a rigid demand that must be met; for ordinary load nodes, the predicted value is used as a benchmark reference, and subsequent flexible adjustments will be made through the supply guarantee rate coefficient. The predicted data of the two types of load nodes are stored separately to form a load forecasting dataset.

[0099] In one embodiment, such as Figure 3 As shown, in step S20, based on topology data, output forecast data, and load forecast data, the optimal power supply path connecting each important load node to the guaranteed power supply node is selected, specifically including:

[0100] S21: For each critical load node, calculate its electrical distance index, geographical distance index, and average power flow load rate index of the lines along the path between it and each guaranteed power supply node.

[0101] In this embodiment, the electrical distance index reflects the resistance encountered when current flows on the transmission line, which is usually measured by the line's reactance value; the geographical distance index reflects the actual physical length of the line, and the longer the line, the greater the risk of damage from natural disasters; the power flow load rate index reflects the load level of the line under normal operating conditions, and the higher the load rate, the smaller the line's reserve capacity.

[0102] Specifically, for each pair of guaranteed power supply nodes and critical load nodes, the Dijkstra algorithm or the Floyd algorithm is used to find all feasible power supply paths between them. For each path, the electrical distance of each line is calculated, i.e., the sum of the line reactance values, as the electrical distance index of the path; the sum of the geographical lengths of each line is calculated as the geographical distance index of the path; and the average power flow load rate of each line under peak load is calculated as the average power flow load rate index of the path. These three indices evaluate the quality of the power supply path from three dimensions: electrical characteristics, physical characteristics, and operational characteristics.

[0103] S22: The electrical distance index, the geographical distance index, and the average power flow load rate index are weighted and summed to obtain a comprehensive evaluation value between the important load node and the backup power supply node.

[0104] In this embodiment, the weight coefficients of the weighted summation are calculated using the entropy weighting method, which is an objective weighting method that can automatically determine the weights based on the dispersion of each indicator data, avoiding bias caused by subjective judgment.

[0105] Specifically, we first construct the original indicator data matrix A, where the rows of the matrix represent different lines, and the columns represent three indicators: electrical distance, geographical distance, and power flow load rate.

[0106] In the formula: M is the number of lines in the power grid;

[0107] Then, utility theory was used to normalize the values ​​of each indicator to eliminate the influence of different dimensions and orders of magnitude:

[0108] In the formula: Xm, Ym, Zm are the normalized electrical distance, geographical distance and power load rate of the m-th line, and 0≤Xm, Ym, Zm≤1.

[0109] Next, the entropy and information redundancy of each indicator are calculated based on the normalized data. The smaller the entropy value, the greater the data dispersion of that indicator, and the greater its contribution to the evaluation results; therefore, it is assigned a higher weight.

[0110] (1)

[0111] (2)

[0112] (3)

[0113] e x The entropy value of the electrical distance index; d x This is for information redundancy. Therefore, the weight value κ1 of the electrical distance can be obtained. Similarly, the weight values ​​κ2 and κ3 are calculated, which will not be elaborated here.

[0114] Finally, the information redundancy is normalized to obtain the weight coefficients of each indicator. The normalized values ​​of the three indicators are multiplied by their corresponding weight coefficients and then summed to obtain the comprehensive evaluation value ζωij of each power supply path between the important load node and the guaranteed power supply node.

[0115]

[0116]

[0117] In the formula: Let ω be the number of lines contained in the ω-th power supply path; Let x be the electrical distance of line m. m Geographical distance y m and peak load factor z m The weighted sum of indicators, load and network flow are constantly changing, and the power flow load rate under peak load is selected for research during the insurance period; and These are designated as either the guaranteed power supply set or the critical load set. Guaranteed power supplies mainly include coal-fired units, gas-fired units, and other main power sources capable of providing stable and continuous output. κ1, κ2, and κ3 represent the weights for electrical distance, geographical distance, and power flow load rate, respectively.

[0118] S23: Select the reliable power supply node with the smallest comprehensive evaluation value corresponding to the important load node as the target power supply node, and determine the shortest electrical path connecting the important load node and the target power supply node as the optimal power supply path for the important load node.

[0119] In this embodiment, the power supply path with the smallest comprehensive evaluation value means that the optimal balance has been achieved in the three dimensions of electrical distance, geographical distance and power flow load rate, which ensures good electrical performance, reduces disaster risk, and leaves sufficient transmission margin.

[0120] Specifically, for each critical load node, all power supply paths between it and all backup power supply nodes are traversed, the comprehensive evaluation values ​​of each path are compared, and the path with the smallest comprehensive evaluation value is selected as the optimal power supply path for that critical load node.

[0121]

[0122] П j This serves as the final lifeline channel for critical load node j;

[0123] The backup power node corresponding to this path is the target power node for this critical load node. If the comprehensive evaluation values ​​of multiple paths are the same, the path with fewer lines and a simpler topology is preferred.

[0124] S24: Summarize the optimal power supply paths for all important load nodes to generate the lifeline channel set.

[0125] In this embodiment, the lifeline channel set is a mandatory constraint for the subsequent backbone network search. All lines in the set must be retained in the final backbone network to ensure that the power supply path for critical loads is not cut off under any circumstances.

[0126] Specifically, all lines included in the optimal power supply path selected for each important load node in step S23 are summarized, duplicate lines are removed, and a line set is generated. Each line in this set is marked as a "required line". In the subsequent backbone network search model, the state variables of these lines will be forced to be set to 1, that is, they must be selected.

[0127] In one embodiment, such as Figure 4 As shown, in step S30, with the goal of minimizing the network size, the lifeline channel set as a mandatory constraint, and the source-grid-load-storage collaborative operation simulation covering the entire extreme survival time period as an operational constraint, a backbone network search model is constructed and solved to obtain the initial backbone network, specifically including:

[0128] S31: Using minimizing the total number of lines contained in the backbone network as the objective function, establish the target layer of the backbone network search model.

[0129] In this embodiment, minimizing the grid size means reducing the number of lines contained in the backbone grid as much as possible while satisfying all constraints. The benefits of doing so are: on the one hand, it can reduce construction and maintenance costs, and on the other hand, it can reduce the exposed area of ​​the grid in extreme disasters and increase the probability of survival.

[0130] Specifically, define the decision variable ξ. m Indicates whether line m has been selected for inclusion in the backbone network:

[0131]

[0132] ξ m =1 indicates that line m is selected, ξ m=0 indicates that line m is not selected, and the objective function is ξ of all lines. m Minimize the sum of, i.e.

[0133]

[0134] The objective function is a linear expression, which makes it easier to solve using a mixed-integer linear programming algorithm.

[0135] S32: Set all lines in the lifeline channel set as mandatory lines and add them to the backbone network search model as the first constraint condition.

[0136] In this embodiment, the first constraint ensures that the power supply path of all critical loads is forcibly incorporated into the backbone network, which is the fundamental guarantee for achieving uninterrupted power supply to critical loads.

[0137] Specifically, for each line m in the lifeline channel set, add the constraint ξ. m =1, which forces the line to be selected. These constraints are added to the model in the form of equality constraints and will not be relaxed or violated during the solution process.

[0138] S23: Construct a source-grid-load-storage coordinated operation simulation covering all time periods within the extreme survival time period, and add the feasibility of the operation simulation as a second constraint to the backbone network search model. The operation simulation includes: start-up and shutdown and output constraints of the guaranteed power supply, active power flow transmission constraints of the line, differentiated supply guarantee constraints of load nodes, and charging, discharging and state of charge constraints of the energy storage device.

[0139] In this embodiment, the source-grid-load-storage coordinated operation simulation refers to the joint optimization and scheduling of the four links of power source, grid, load and energy storage during the extreme survival period to ensure that the system can continue to operate stably without external power support. The feasibility of this operation simulation is used as the second constraint, which means that only those grid structures that can support the entire extreme survival process are feasible.

[0140] Specifically, a mathematical constraint system for the operational simulation is constructed, including: source-side constraints, such as upper and lower limits of unit output, minimum start-up and shutdown time constraints, and ramp-up rate constraints; grid-side constraints, such as node power balance constraints, line power flow transmission constraints, and voltage phase angle constraints; load-side constraints, such as full supply guarantee constraints for important loads and supply guarantee constraints for ordinary loads based on the supply guarantee rate; and storage-side constraints, such as energy storage charging and discharging power constraints, upper and lower limits of state of charge constraints, and energy conservation constraints. These constraints are then incorporated into the backbone network search model in the form of linear inequalities or equations.

[0141] S34: The backbone network search model is solved using a mixed integer linear programming algorithm to obtain the initial backbone network that satisfies all constraints and has the minimum number of lines.

[0142] In this embodiment, the mixed-integer linear programming algorithm is a mathematical programming method that can handle optimization problems that simultaneously contain continuous and integer variables, and is applicable to the binary decision variables and continuous power flow variables contained in this model.

[0143] Specifically, the constructed mixed-integer linear programming model is solved. During the solution process, the solver searches for the combination of decision variables that minimizes the objective function while satisfying all constraints. After the solution is completed, all ξ values ​​are output. m The lines with a value of 1 form the initial backbone network.

[0144] In one embodiment, such as Figure 5 As shown, in step S33, the differentiated supply guarantee constraints for load nodes specifically include:

[0145] S331: For a node belonging to the set of important load nodes, the active power it guarantees is equal to its predicted active power demand in each time period of the extreme survival time period.

[0146] In this embodiment, uninterrupted power supply to critical load nodes is one of the core objectives of extreme survival. Therefore, the power supply to these nodes must be strictly equal to their predicted demand, and no reduction is allowed.

[0147] Specifically, for each important load node d∈ In each time interval t∈ of the limit survival time interval Add constraints:

[0148]

[0149] in The actual power supply of critical load nodes at time t. The constraint, which forecasts active power demand, ensures that critical loads are not cut at any time.

[0150] S332: For ordinary load nodes that do not belong to the set of important load nodes, during each period of the extreme survival time period, the active power they are guaranteed to supply shall not be less than their predicted active power demand multiplied by the preset guarantee rate coefficient.

[0151] In this embodiment, the normal load is allowed to be reduced within a certain range, but in order to ensure basic socio-economic operation, a minimum supply guarantee rate needs to be set. The supply guarantee rate coefficient λ is a parameter between 0 and 1, which can be adjusted according to the actual situation.

[0152] Specifically, for each ordinary load node d∈ In each time interval t∈ of the limit survival time interval Add constraints ≥ λ × ,in Let t be the actual power supply of a normal load node. The predicted active power demand is given, where λ is a preset supply guarantee rate coefficient. Additionally, to ensure that the total reduction in ordinary load does not exceed a reasonable range, a cumulative supply guarantee constraint needs to be added.

[0153]

[0154] That is, during the entire extreme survival period, the total power supply for ordinary loads shall not be less than λ times their total demand, but not more than their total demand.

[0155] In one embodiment, such as Figure 6 As shown, in step S40, the initial backbone network is subjected to connectivity assessment. If disconnected regions exist, connecting lines are added to repair connectivity, resulting in a candidate connected backbone network. Specifically, this includes:

[0156] S41: Based on Dijkstra's algorithm, check all connected regions in the initial backbone network to determine whether the number of connected regions is equal to 1 and whether all important load nodes are in the same connected region.

[0157] In this embodiment, Dijkstra's algorithm is a classic shortest path algorithm that can traverse all reachable nodes and record the shortest distance from a specified starting point. This step utilizes the traversal characteristic of Dijkstra's algorithm to detect connected regions, and simultaneously provides shortest path information for subsequent connectivity repair, achieving two goals at once. A connected region refers to a group of nodes connected by lines, and there is at least one path between any two nodes within the same connected region. Critical load nodes refer to key user nodes that require uninterrupted power supply, such as government agencies, hospitals, and data centers.

[0158] Specifically, the initial backbone network is abstracted as an undirected graph, where nodes represent substations and lines represent edges. Starting from any guaranteed power supply node or critical load node, Dijkstra's algorithm is used to traverse all reachable nodes, recording the set of visited nodes to form a connected region. Then, it is checked whether this connected region contains all critical load nodes. If there are unvisited critical load nodes, Dijkstra's algorithm is executed again starting from the unvisited critical load node to form a second connected region. This process is repeated until all critical load nodes are assigned to a connected region. The total number of connected regions is counted, and a list of critical load nodes contained in each connected region is recorded.

[0159] S42: If the number of connected regions is equal to 1 and all important load nodes are connected, the initial backbone network is determined to be connected, and the process proceeds directly to the DC power flow verification step. If the number of connected regions is not equal to 1 or there are important load nodes that are not connected, the initial backbone network is determined to be not connected, and all disconnected isolated subgraphs and the important load nodes contained therein are identified.

[0160] In this embodiment, the judgment criterion for this step focuses on the connectivity of critical load nodes, rather than the connectivity of all nodes. This is because the core objective of extreme survival scenarios is to ensure the power supply to critical loads. As long as all critical load nodes can be interconnected through the backbone network, even if some non-critical nodes are isolated, it will not affect the power supply reliability of critical loads.

[0161] Specifically, the judgment is made based on the traversal results of step S41. If the number of connected regions is exactly one, and this connected region contains all important load nodes, it indicates that the initial backbone network can already meet the interconnection requirements of important loads, and no connectivity repair is needed. The process can proceed directly to step S50 for DC power flow verification. Conversely, if the number of connected regions is greater than one, or although the number of connected regions is one, it fails to cover all important load nodes, the initial backbone network is determined to be disconnected. In this case, it is necessary to identify all disconnected isolated subgraphs and record the list of important load nodes contained in each isolated subgraph to provide a basis for subsequent connectivity repair.

[0162] S43: With the goal of minimizing the total length of the added tie lines, the minimum spanning tree algorithm is used to add tie lines from the original transmission network topology between the disconnected isolated subgraphs, so that all important load nodes are connected into a connected whole, wherein the intermediate nodes through which the tie lines pass include non-important nodes.

[0163] In this embodiment, the minimum spanning tree algorithm is an algorithm that finds the minimum weighted edge set connecting all vertices in a weighted graph, where the weights are the geographical lengths of the connecting lines. The goal of this step is not to connect all isolated subgraphs, but to connect all important load nodes as the ultimate objective, allowing connecting lines to pass through non-important nodes as intermediate stepping stones, thereby minimizing the total length and cost of new lines while ensuring connectivity.

[0164] Specifically, each isolated subgraph containing critical load nodes is treated as a supernode. For any two such isolated subgraphs, all feasible tie paths between them are found in the original transmission network topology, and the geographical length of each path is calculated. It's important to note that these paths may pass through other isolated subgraphs or non-critical nodes, as long as they connect the two supernodes. A complete graph is constructed with supernodes as vertices and path lengths as weights. Then, Prim's algorithm or Kruskal's algorithm is used to find the minimum spanning tree of this complete graph. Each edge in the minimum spanning tree corresponds to a tie path that needs to be added. Adding all lines on these paths to the initial backbone network connects all critical load nodes into a connected whole.

[0165] S44: The initial backbone network after adding the connecting lines is determined as the candidate backbone network for connectivity.

[0166] In this embodiment, the connected candidate backbone network is a connected network formed by adding necessary tie lines on the basis of the initial backbone network, and it is a prerequisite for performing DC power flow verification.

[0167] Specifically, all the tie lines added in step S43 are merged with the lines in the initial backbone network to generate a new set of lines. The network corresponding to this set of lines is the candidate backbone network with connectivity, where all nodes are in the same connected subgraph, allowing for subsequent power flow calculations.

[0168] In one embodiment, such as Figure 7 As shown, in step S50, DC power flow verification is performed on the candidate backbone network. If there is no line overload in any time period, the candidate backbone network is determined as the final backbone network. If there is line overload, the load supply parameters in the operation constraints are adjusted, and the process returns to reconstruct and solve the backbone network search model. Specifically, this includes:

[0169] S51: Based on the topology and line parameters of the connected candidate backbone network, and the power output and load levels of the source-grid-load-storage coordinated operation simulation at each time period, DC power flow calculation is performed time-by-time to obtain the active power flow value of each line at each time period.

[0170] In this embodiment, DC power flow calculation is a simplified power flow calculation method that ignores reactive power and voltage amplitude. It linearizes the AC power flow equation, making the calculation fast and the convergence good, and is suitable for rapid verification of large-scale power grids.

[0171] Specifically, for each time period within the extreme survival timeframe, based on the scheduling results of the source-grid-load-storage coordinated operation simulation during that time period, the active power output of each power source node and the active power demand of each load node are obtained. Based on the topology of the connected candidate backbone network and line reactance parameters, a node admittance matrix and a node injected power vector are constructed. The DC power flow calculation formula P = Bθ is used to solve for the voltage phase angle of each node, thereby calculating the active power flow value of each line. This process is repeated to obtain the active power flow value matrix for all lines in all time periods.

[0172] S52: Compare the active power flow value of each line in each time period with its corresponding line capacity parameter. If the active power flow value of all lines in all time periods is not greater than its corresponding line capacity parameter, the DC power flow verification is determined to be passed, and the connected candidate backbone network is output as the final backbone network.

[0173] In this embodiment, the line capacity parameter is the maximum allowable transmission power of the line. Exceeding this value will cause the line to overheat, increase sag, or even break. Therefore, overload must not occur at any time.

[0174] Specifically, for each line m and each time period t, the active power flow value |P{m,t}| is compared with the line capacity parameter Cap. m Compare them. If for all m and all t, |P{m,t}| ≤ Cap m If the DC power flow verification is passed, the candidate backbone network can safely bear the power flow distribution in the extreme survival operation simulation. It is then determined as the final backbone network and output.

[0175] S53: If the active power flow value of any line in any time period is greater than its corresponding line capacity parameter, the DC power flow verification is determined to be unsuccessful, and the preset power supply guarantee rate coefficient is reduced by a preset step size. Then, the process returns to the step of constructing the backbone network search model and solving it.

[0176] In this embodiment, when a line overload occurs, it indicates that the current backbone network cannot operate safely. It is necessary to reduce the total load demand by lowering the supply rate of ordinary loads, thereby reducing the line power flow, and then re-search the backbone network.

[0177] Specifically, if there exist lines m and time periods t such that |P_{m,t}| > Cap mIf the DC power flow verification fails, the preset normal load supply rate coefficient λ is reduced by a preset step size Δλ, for example, 0.05, i.e., λnew = λold - Δλ. Then, the updated λ is substituted into the load differentiation supply constraint in step S33, and the process returns to step S31 to reconstruct the backbone network search model and solve it. This process is repeated until the DC power flow verification passes or the supply rate coefficient is reduced to the preset lower limit. The final output backbone network is a safe and reliable network that can meet the extreme survival operation requirements without overload.

[0178] In one embodiment, step S30, namely the source-grid-load-storage coordinated operation simulation, further includes network switching constraints, which are used to limit the upper limit of the number of lines that can be actively disconnected within the extreme survival time period.

[0179] In this embodiment, network switching constraints refer to the ability to proactively disconnect some lines during extreme survival operations to flexibly adjust the electrical connections between sources, grids, loads, and storage based on actual operating conditions. However, the number of disconnected lines cannot exceed a preset upper limit. This is because frequent or large-scale line disconnections can disrupt the system's topology stability, increase the risk of power flow shifting, and even trigger cascading failures; therefore, this constraint must be limited. This constraint ensures that the backbone network has both a certain degree of topology adjustment flexibility and maintains basic network structure stability during operation, avoiding a decrease in power supply reliability due to excessive switching.

[0180] Specifically, when constructing a simulation of coordinated operation of power generation, grid, load, and storage, line operation state variables are introduced. This indicates whether line m is in a closed operating state at time t. =1 indicates that the circuit is closed and energized. =0 indicates that the line is disconnected. First, the operational status of the line must conform to the topology selection result of the backbone network; that is, only lines selected for the backbone network can be put into operation. Therefore, constraints are added.

[0181]

[0182] Where ξm is the decision variable for whether line m is selected into the backbone network. Secondly, at each time t within the extreme survival time period, the number of lines that can be actively disconnected is limited; that is, the difference between the number of closed lines at the current time and the number of lines selected at the initial time cannot exceed the maximum allowed disconnection number δt, expressed as...

[0183]

[0184] δt can be set according to system scale and operating experience, for example, it can be taken as 5% to 10% of the total number of lines, and different values ​​can be set at different times to adapt to operating needs. By introducing network switching constraints, the backbone network can adjust the network topology in a timely manner according to power output and load changes during extreme survival processes, while avoiding the operational risks caused by excessive switching and ensuring the safe and stable operation of the system.

[0185] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0186] In one embodiment, combined with Figure 9-14 The backbone network search method for urban power transmission networks designed for extreme survival was applied to perform a backbone network search on the IEEE 30-node test system:

[0187] Consider a typical microgrid system comprising wind power, solar power, energy storage batteries, and loads, connected to the main grid. Microgrid operation data is based on 96 time points per day (one point every 15 minutes), with virtual data simulating a real-world scenario: solar power output is higher during the day, wind power fluctuates significantly, and loads have morning and evening peaks. Traditional energy storage benefit calculation methods only calculate peak-valley differences based on purchased electricity prices, ignoring the zero-cost nature and self-sufficiency of green electricity (wind and solar) within the microgrid, leading to inaccurate calculations. This case study applies the method of this invention, optimizing energy storage operation by dynamically calculating cost-benefit coefficients.

[0188] like Figure 9 As shown, a modified IEEE 30-node system is used, incorporating various power sources including coal-fired power plants, gas-fired power plants, wind and solar power, and energy storage, with loads of varying importance. The modified IEEE 30-node system has 41 lines. Nodes 13 and 23 each connect to a coal-fired power plant, nodes 1 and 27 each connect to a gas-fired power plant, nodes 2 and 22 each connect to a wind farm and a solar power plant, and node 6 connects to an energy storage device. The peak load is 189.2 MW, of which 81.9 MW is critical load and 107.3 MW is general load. Nodes 1, 13, 23, and 27 are designated as backup power sources, while nodes 2, 3, 10, 12, 17, 18, 21, 24, and 29 are designated as critical loads. The battery energy storage device has parameters of 20 MW / 80 MWh. The ramp rates for the coal-fired power plants, gas-fired power plants, and energy storage device are 1.5%, 8%, and 100% of rated capacity per minute, respectively. The normal load supply guarantee rate is taken as λ=30%.

[0189] The above-mentioned search method for urban power grid backbone structures oriented towards extreme survival is applied step by step in the calculation:

[0190] Step 1: Obtain Data

[0191] The peak load is 189.2MW, of which critical load is 81.9MW and ordinary load is 107.3MW. The backup power nodes are nodes 1, 13, 23, and 27, and the critical load nodes are nodes 2, 3, 10, 12, 17, 18, 21, 24, and 29. The battery energy storage equipment has parameters of 20MW / 80MWh. The ramp rates for coal-fired units, gas-fired units, and energy storage equipment are 1.5%, 8%, and 100% of rated capacity per minute, respectively. The ordinary load supply guarantee rate is taken as λ=30%.

[0192] Step 2: Select the lifeline channel connecting the backup power supply node and the critical load node.

[0193] According to formulas (1) to (3) in step S22, calculate the weight values ​​of electrical distance, geographical distance and power flow load rate of the power grid line respectively:

[0194] Table 1 Sub-indicator Weight Values

[0195] electrical distance 0.336 Geographical distance 0.326 Current load rate 0.338

[0196] like Figure 10 As shown, by calculating the comprehensive evaluation index of the power supply path between different backup power sources and important load nodes, alternative lifeline channels between nodes are obtained.

[0197] According to the formula The power supply path corresponding to the minimum comprehensive evaluation index is selected as the final lifeline channel:

[0198] Table 2 Selection of the final lifeline channel

[0199] 2 1 Line 1 0.267 3 1 Line 2 0.373 10 1 Lines 14, 11, 6, 1 0.623 12 13 Line 16 0.052 17 13 Routes 21, 19, and 16 0.839 18 23 Lines 22 and 30 0.406 21 23 Routes 29, 31, and 32 0.987 24 23 Line 32 0.246 29 27 Line 37 0.635

[0200] Steps 3-5: Search for backbone networks with connectivity and perform DC power flow verification.

[0201] like Figure 11-14 As shown, the backbone network to be searched must meet the safe operation requirements at a specific moment or meet the requirements of safety, resilience, and economy. After the urban power grid is disconnected from the grid, it should also meet the differentiated load supply requirements within a continuous extreme survival time. In addition to setting the backbone network search scenario at a specific moment, four scenarios with extreme survival times of 1 day, 3 days, and 5 days are also set.

[0202] The backbone grid structure differs under specific timeframes and in different scenarios requiring 1, 3, and 5 days of extreme survival. This is because the backbone grid structure is related to the extreme survival time and the source-load fluctuation characteristics during that period. When the required extreme survival time is different, the identified backbone grid structure will also be different. The backbone grid structure identified under 1, 3, and 5 days of extreme survival can meet the differentiated load supply requirements. Based on the backbone grid structure identified under specific timeframes, the load supply rates for different importance under 1, 3, and 5 days of extreme survival are shown in Table 3.

[0203] Table 3 Supply retention rate under different extreme survival times

[0204] One day of extreme survival 90.81% 40.06% Survival in extreme conditions for 3 days 89.95% 40.01% Survival in extreme conditions for 5 days 89.51% 40.06%

[0205] Therefore, the backbone grid at a specific moment is difficult to adapt to the differentiated load supply requirements under different extreme survival times, while the backbone grid searched by the method proposed in this paper can meet the supply requirements.

[0206] In one embodiment, an electronic device is provided, which may be a server, and its internal structure diagram may be as follows: Figure 8 As shown, this electronic device includes a processor, memory, network interface, and database connected via a system bus. The processor provides computing and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores the operating system, computer programs, and database. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage medium. The database stores the database. The network interface is used for communication with external terminals via a network connection. When the computer program is executed by the processor, it implements a method for searching the backbone network of an urban power transmission grid with extreme survivability.

[0207] In one embodiment, an electronic device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to perform the following steps:

[0208] Acquire topology data of the urban power transmission network, extreme survivability output prediction data of backup power sources, and load prediction data of each load node;

[0209] Based on the topology data, the output prediction data, and the load prediction data, the optimal power supply path connecting each important load node to the backup power supply node is selected, and all selected optimal power supply paths are combined to form a set of lifeline channels that must be guaranteed.

[0210] With the goal of minimizing the scale of the grid structure, the lifeline channel set as a mandatory constraint, and the source-grid-load-storage collaborative operation simulation covering the entire extreme survival time period as an operational constraint, a backbone grid search model is constructed and solved to obtain the initial backbone grid structure.

[0211] The initial backbone network is assessed for connectivity. If there are disconnected areas, connecting lines are added to repair the connectivity, resulting in a candidate backbone network with connectivity.

[0212] The DC power flow is verified on the candidate backbone network. If there is no line overload in any time period, the candidate backbone network is determined as the final backbone network. If there is line overload, the load supply parameters in the operation constraints are adjusted, and the backbone network search model is reconstructed and solved.

[0213] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, the computer program performing the following steps when executed by a processor:

[0214] Acquire topology data of the urban power transmission network, extreme survivability output prediction data of backup power sources, and load prediction data of each load node;

[0215] Based on the topology data, the output prediction data, and the load prediction data, the optimal power supply path connecting each important load node to the backup power supply node is selected, and all selected optimal power supply paths are combined to form a set of lifeline channels that must be guaranteed.

[0216] With the goal of minimizing the scale of the grid structure, the lifeline channel set as a mandatory constraint, and the source-grid-load-storage collaborative operation simulation covering the entire extreme survival time period as an operational constraint, a backbone grid search model is constructed and solved to obtain the initial backbone grid structure.

[0217] The initial backbone network is assessed for connectivity. If there are disconnected areas, connecting lines are added to repair the connectivity, resulting in a candidate backbone network with connectivity.

[0218] The DC power flow is verified on the candidate backbone network. If there is no line overload in any time period, the candidate backbone network is determined as the final backbone network. If there is line overload, the load supply parameters in the operation constraints are adjusted, and the backbone network search model is reconstructed and solved.

[0219] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0220] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is used as an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

[0221] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A method for searching the backbone network of an urban power transmission network for extreme survival, characterized in that, The method for searching urban power grid backbone structures for extreme survival includes the following steps: Acquire topology data of the urban power transmission network, extreme survivability output prediction data of backup power sources, and load prediction data of each load node; Based on the topology data, the output prediction data, and the load prediction data, the optimal power supply path connecting each important load node to the backup power supply node is selected, and all selected optimal power supply paths are combined to form a set of lifeline channels that must be guaranteed. With the goal of minimizing the scale of the grid structure, the lifeline channel set as a mandatory constraint, and the source-grid-load-storage collaborative operation simulation covering the entire extreme survival time period as an operational constraint, a backbone grid search model is constructed and solved to obtain the initial backbone grid structure. The initial backbone network is assessed for connectivity. If there are disconnected areas, connecting lines are added to repair the connectivity, resulting in a candidate backbone network with connectivity. The DC power flow is verified on the candidate backbone network. If there is no line overload in any time period, the candidate backbone network is determined as the final backbone network. If there is line overload, the load supply parameters in the operation constraints are adjusted, and the backbone network search model is reconstructed and solved.

2. The urban power transmission network backbone search method for extreme survival as described in claim 1, characterized in that, The acquisition of urban power grid topology data, critical survivability power output prediction data for backup power sources, and load prediction data for each load node specifically includes: The connection relationships between nodes and lines of the urban power transmission network, line impedance parameters, and line capacity parameters are obtained as the topology data. The maximum available active power output of all backup power sources during each time period within the extreme survival time period is obtained and used as the output prediction data. The predicted active power demand of each important load node and each ordinary load node in each time period during the extreme survival period is obtained as the load prediction data.

3. The urban power transmission network backbone search method for extreme survival as described in claim 1, characterized in that, The step of selecting the optimal power supply path connecting each critical load node to a backup power supply node based on the topology data, the output prediction data, and the load prediction data specifically includes: For each critical load node, calculate its electrical distance index, geographical distance index, and average power flow load rate index between it and each backup power supply node; The electrical distance index, the geographical distance index, and the average power flow load rate index are weighted and summed to obtain a comprehensive evaluation value between the important load node and the backup power supply node; Select the reliable power supply node with the smallest comprehensive evaluation value corresponding to the important load node as the target power supply node, and determine the shortest electrical path connecting the important load node and the target power supply node as the optimal power supply path for the important load node. The optimal power supply paths for all critical load nodes are aggregated to generate the lifeline channel set.

4. The urban power transmission network backbone search method for extreme survival as described in claim 1, characterized in that, The process aims to minimize the network structure size, uses the lifeline channel set as a mandatory constraint, and employs an embedded source-grid-load-storage collaborative operation simulation covering the entire extreme survival time as an operational constraint. A backbone network search model is constructed and solved to obtain the initial backbone network structure, specifically including: The target layer of the backbone network search model is established with the objective function of minimizing the total number of lines contained in the backbone network. All lines in the lifeline channel set are set as mandatory lines and added to the backbone network search model as the first constraint condition. A source-grid-load-storage coordinated operation simulation covering all time periods within the extreme survival time period is constructed, and the feasibility of the operation simulation is added to the backbone network search model as a second constraint. The operation simulation includes: start-up and shutdown and output constraints of the guaranteed power supply, active power flow transmission constraints of the line, differentiated supply guarantee constraints of load nodes, and charging, discharging and state of charge constraints of the energy storage device. The backbone network search model is solved using a mixed-integer linear programming algorithm to obtain the initial backbone network that satisfies all constraints and has the minimum number of lines.

5. The urban power transmission network backbone search method for extreme survival as described in claim 4, characterized in that, The differentiated supply guarantee constraints for the load nodes specifically include: For a node belonging to the set of critical load nodes, the active power it guarantees is equal to its predicted active power demand in each time period of the extreme survival time period. For ordinary load nodes that do not belong to the set of important load nodes, the active power they are guaranteed to supply during each period of the extreme survival time shall not be less than their predicted active power demand multiplied by the preset guarantee rate coefficient.

6. The urban power transmission network backbone search method for extreme survival as described in claim 1, characterized in that, The process of determining the connectivity of the initial backbone network and, if any disconnected areas exist, adding connecting lines to repair connectivity and obtaining a candidate connected backbone network specifically includes: The Dijkstra algorithm is used to check and search all connected regions in the initial backbone network to determine whether the number of connected regions is equal to 1 and whether all important load nodes are in the same connected region. If the number of connected regions is equal to 1 and all important load nodes are connected, the initial backbone network is determined to be connected, and the process proceeds directly to the DC power flow verification step; if the number of connected regions is not equal to 1 or there are important load nodes that are not connected, the initial backbone network is determined to be not connected, and all disconnected isolated subgraphs and the important load nodes contained therein are identified. With the goal of minimizing the total length of the added tie lines, the minimum spanning tree algorithm is used to add tie lines from the original transmission network topology between the disconnected isolated subgraphs, so that all important load nodes are connected into a connected whole, wherein the intermediate nodes through which the tie lines pass include non-important nodes. The initial backbone network after adding the connecting lines is determined as the candidate backbone network for connectivity.

7. The urban power transmission network backbone search method for extreme survival as described in claim 1, characterized in that, The DC power flow verification of the connected candidate backbone network is performed. If there is no line overload in any time period, the connected candidate backbone network is determined as the final backbone network. If line overload exists, adjust the load guarantee parameters in the operational constraints, and return to reconstruct and solve the backbone network search model, specifically including: Based on the topology and line parameters of the connected candidate backbone network, and the power output and load levels of the source-grid-load-storage coordinated operation simulation at each time period, DC power flow calculation is performed time-by-time to obtain the active power flow value of each line at each time period. The active power flow value of each line in each time period is compared with its corresponding line capacity parameter. If the active power flow value of all lines in all time periods is not greater than its corresponding line capacity parameter, the DC power flow verification is determined to be passed, and the connected candidate backbone network is output as the final backbone network. If the active power flow value of any line in any time period is greater than its corresponding line capacity parameter, the DC power flow verification is determined to fail, and the preset supply guarantee rate coefficient is reduced by a preset step size. Then, the process returns to the step of constructing the backbone network search model and solving it.

8. The urban power transmission network backbone search method for extreme survival as described in claim 4, characterized in that, The source-grid-load-storage coordinated operation simulation also includes network switching constraints, which limit the maximum number of lines that can be actively disconnected within the extreme survival time period.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the urban power transmission network backbone search method as described in any one of claims 1 to 8.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the urban power grid backbone network search method for extreme survival as described in any one of claims 1 to 8.

Images