A fault self-healing oriented active power distribution network comprehensive planning method and system

Abstract

This invention discloses a comprehensive planning method and system for active distribution networks oriented towards fault self-healing, relating to the field of active distribution network comprehensive planning technology. The method includes: acquiring active distribution network topology data, controllable power supply distribution data, and load distribution data to construct an initial distribution network structure; analyzing the fault propagation process under different fault scenarios to determine the hierarchical fault influence domain; determining node recovery priorities based on the hierarchical fault influence domain and establishing a collaborative constraint relationship between primary equipment configuration requirements and secondary automation configuration requirements; generating a set of candidate equipment configuration schemes based on the collaborative constraint relationship and evaluating the power supply recovery capability of each scheme; and performing multi-objective optimization by combining equipment configuration cost and power supply recovery coverage to determine the optimal equipment configuration scheme, thereby achieving collaborative configuration of the active distribution network structure and equipment resources, improving fault self-healing capability and planning reliability.

Description

Technical Field

[0001] This invention relates to the field of active distribution network integrated planning technology, specifically to an active distribution network integrated planning method and system oriented towards fault self-healing. Background Technology

[0002] Against the backdrop of the continuous advancement of new power system construction, active distribution networks, due to their integration of various flexible resources such as distributed power sources, energy storage devices, and controllable loads, have become crucial infrastructure for improving power supply reliability, absorbing new energy sources, and supporting fault self-healing. As the operational structure of distribution networks becomes increasingly complex and the coordination relationships between different types of power sources and loads continue to strengthen, traditional planning methods focused on power supply reliability or equipment configuration are no longer sufficient to meet the comprehensive needs of active distribution networks for rapid fault recovery, enhanced self-healing capabilities, and coordinated resource allocation. Therefore, conducting comprehensive planning for active distribution networks oriented towards fault self-healing has become an important research direction in the field of distribution network planning.

[0003] However, existing integrated planning methods for active distribution networks typically only optimize network topology or equipment configuration, lacking hierarchical analysis of fault propagation processes and impact ranges under different fault scenarios. They struggle to establish node recovery priorities based on fault impact levels and fail to establish collaborative constraints between primary equipment configuration requirements and secondary automation configuration requirements. This results in equipment configuration schemes that cannot simultaneously address fault recovery and automation control needs. Furthermore, the lack of power supply recovery capability evaluation based on fault-level impact domains and multi-objective integrated optimization mechanisms hinders the collaborative planning of network structure and equipment resources, impacting the fault self-healing capability and overall planning effectiveness of active distribution networks. Therefore, a fault-self-healing-oriented integrated planning method and system for active distribution networks is needed to address the aforementioned problems. Summary of the Invention

[0004] To address the aforementioned technical problems, this paper provides a comprehensive planning method and system for active distribution networks oriented towards fault self-healing. This solution resolves the issues raised in the background section regarding existing comprehensive planning methods for active distribution networks, which typically only optimize network topology or equipment configuration. These methods lack hierarchical analysis of fault propagation processes and impact ranges under different fault scenarios, making it difficult to establish node recovery priorities based on the degree of fault impact. Furthermore, they fail to establish a collaborative constraint relationship between primary equipment configuration requirements and secondary automation configuration requirements, resulting in equipment configuration schemes that cannot simultaneously meet fault recovery and automation control needs. Additionally, the lack of a power supply recovery capability evaluation based on fault-level impact domains and a multi-objective comprehensive optimization mechanism hinders the collaborative planning of network structure and equipment resources, thus affecting the fault self-healing capability and comprehensive planning effectiveness of active distribution networks.

[0005] To achieve the above objectives, the technical solution adopted by the present invention is as follows: A comprehensive planning method for active distribution networks oriented towards fault self-healing includes: Acquire topology data, controllable power supply distribution data, and load distribution data of the active power distribution network, and construct the initial power distribution network structure; Based on the initial power distribution network structure, the fault propagation process under different fault scenarios is analyzed to determine the graded fault influence domain corresponding to each fault scenario. Based on the hierarchical fault impact domain, the node recovery priority under different impact levels is determined, a set of node hierarchical recovery constraint relationships is generated, and a collaborative constraint relationship between primary equipment configuration requirements and secondary automation configuration requirements is established. Based on the collaborative constraint relationship between primary equipment configuration requirements and secondary automation configuration requirements, and combined with adjustable power distribution data and load distribution data, a set of candidate equipment configuration schemes is generated. Based on the set of candidate device configuration schemes, the power supply recovery capability of each configuration scheme is evaluated under the constraints of different fault impact domains, and self-healing capability evaluation results are generated. Based on the self-healing capability evaluation results, multi-objective comprehensive optimization is carried out by combining equipment configuration costs and power supply restoration coverage to generate the optimal equipment configuration scheme. The grid structure and equipment resources of the active distribution network are coordinated and configured to output comprehensive planning results for fault-level self-healing.

[0006] Furthermore, a fault-healing-oriented active distribution network integrated planning system is proposed to implement the above-mentioned active distribution network integrated planning method, including: The data acquisition module is used to acquire topology data, controllable power distribution data, and load distribution data of the active power distribution network, and to construct an initial power distribution network structure. The fault analysis module is used to analyze the fault propagation process under different fault scenarios based on the initial power distribution network structure, and determine the graded fault influence domain corresponding to each fault scenario. The recovery constraint module is used to determine the node recovery priority under different impact levels based on the hierarchical fault impact domain, generate a set of node hierarchical recovery constraint relationships, and establish a collaborative constraint relationship between primary equipment configuration requirements and secondary automation configuration requirements. The scheme generation module is used to generate a set of candidate equipment configuration schemes based on the collaborative constraint relationship between primary equipment configuration requirements and secondary automation configuration requirements, combined with adjustable power distribution data and load distribution data. The capability evaluation module is used to evaluate the power supply recovery capability of each configuration scheme based on a set of candidate equipment configuration schemes under the constraints of different graded fault influence domains, and generate self-healing capability evaluation results. The optimization configuration module is used to perform multi-objective comprehensive optimization based on the self-healing capability evaluation results, combined with equipment configuration costs and power supply restoration coverage, to generate the optimal equipment configuration scheme, and to perform coordinated configuration of the grid structure and equipment resources of the active distribution network, outputting comprehensive planning results oriented towards fault-level self-healing.

[0007] Compared with the prior art, the beneficial effects of the present invention are: This proposal presents a comprehensive planning method and system for active distribution networks oriented towards fault self-healing. By acquiring topology data, controllable power source distribution data, and load distribution data of the active distribution network, it constructs an initial distribution network structure that includes power source access relationships and load distribution relationships. Based on different fault scenarios, it analyzes the fault propagation process and determines the corresponding hierarchical fault impact domains. This allows the fault impact range to be hierarchically characterized in conjunction with network structure characteristics, improving the accuracy of fault scenario analysis and providing a reliable basis for subsequent recovery planning. This proposal presents a comprehensive planning method and system for active distribution networks oriented towards fault self-healing. By analyzing the node recovery requirements corresponding to each impact level based on the hierarchical fault impact domain, the node recovery priority is determined. Furthermore, a collaborative constraint relationship is established between the primary equipment configuration requirements and the secondary automation configuration requirements based on the node recovery sequence. This enables the equipment planning process to match the fault recovery requirements, avoiding the problem of insufficient recovery capability caused by the independent configuration of primary equipment and automation equipment in traditional planning methods, and improving the synergy of equipment configuration. This proposal presents a comprehensive planning method and system for active distribution networks oriented towards fault self-healing. By combining the coordinated constraint relationship between primary equipment configuration requirements and secondary automation configuration requirements with controllable power distribution data and load distribution data, a set of candidate equipment configuration schemes is generated. The power access relationship, load coverage and equipment configuration of each configuration scheme are analyzed so that the generated configuration scheme can simultaneously meet the power restoration requirements and resource allocation constraints, thereby improving the adaptability of the candidate planning schemes. This proposal presents a comprehensive planning method and system for active distribution networks oriented towards fault self-healing. By evaluating the power supply recovery capability of candidate equipment configuration schemes under the constraints of different fault impact domains, the self-healing capability evaluation results are generated by combining the number of nodes covered by the recovery, the recovery load capacity, and the recovery completion time. Furthermore, multi-objective optimization is performed by combining equipment configuration cost and power supply recovery coverage to determine the optimal equipment configuration scheme, thereby achieving coordinated configuration of network structure and equipment resources, and improving the fault self-healing capability and comprehensive planning effect of active distribution networks. Attached Figure Description

[0008] Figure 1 This is a flowchart of an active distribution network integrated planning method for fault self-healing proposed in this invention; Figure 2 This is a flowchart of the fault influence domain analysis in this invention; Figure 3 This is a flowchart of the optimized device configuration scheme in this invention; Figure 4 This is a system framework diagram of an active distribution network integrated planning system for fault self-healing proposed in this invention. Detailed Implementation

[0009] The following description is intended to disclose the invention and enable those skilled in the art to implement it. The preferred embodiments described below are merely examples, and other obvious variations will occur to those skilled in the art.

[0010] Reference Figure 1 - Figure 4 As shown, an active distribution network integrated planning method for fault self-healing in an embodiment of the present invention includes: Acquire topology data, controllable power supply distribution data, and load distribution data of the active power distribution network, and construct the initial power distribution network structure; Specifically, in the planning scenario of an active distribution network including distributed photovoltaic (PV), energy storage devices, and adjustable loads, the basic operational data of the target distribution area is first collected. Topology data is acquired through distribution automation systems, equipment asset management databases, and geographic information platforms. The collected data includes distribution node numbers, substation outgoing line information, feeder line connections, line affiliations, and switchgear connection locations, describing the actual connection relationships between different nodes in the distribution network. Adjustable power distribution data is acquired through distributed power management platforms, energy storage operation monitoring platforms, and microgrid management platforms. The collected data includes access nodes, equipment types, rated capacities, operating status, and adjustable power ranges for adjustable power sources such as PV and energy storage, reflecting the power support capabilities of different distribution nodes. Load distribution data is acquired through smart meter acquisition systems, electricity consumption information acquisition platforms, and historical load databases. The collected data includes connection nodes, load capacity, electricity consumption curves, load types, and importance for each load node, representing the load demand for different nodes. For example, during the planning of a regional power distribution network, the connection relationships of feeder lines and the connection locations of switching equipment can be obtained through the power distribution automation system, the information of energy storage devices of connected nodes can be obtained through the distributed power management platform, and the power supply demand corresponding to each load node can be determined by combining the load capacity and electricity consumption curve data collected by smart meters, thereby obtaining a data foundation that matches the subsequent network topology construction, power mapping and load association processing.

[0011] Based on the initial power distribution network structure, the fault propagation process under different fault scenarios is analyzed to determine the graded fault influence domain corresponding to each fault scenario. Based on the hierarchical fault impact domain, the node recovery priority under different impact levels is determined, a set of node hierarchical recovery constraint relationships is generated, and a collaborative constraint relationship between primary equipment configuration requirements and secondary automation configuration requirements is established. Based on the collaborative constraint relationship between primary equipment configuration requirements and secondary automation configuration requirements, and combined with adjustable power distribution data and load distribution data, a set of candidate equipment configuration schemes is generated. Based on the set of candidate device configuration schemes, the power supply recovery capability of each configuration scheme is evaluated under the constraints of different fault impact domains, and self-healing capability evaluation results are generated. Based on the self-healing capability evaluation results, multi-objective comprehensive optimization is carried out by combining equipment configuration costs and power supply restoration coverage to generate the optimal equipment configuration scheme. The grid structure and equipment resources of the active distribution network are coordinated and configured to output comprehensive planning results for fault-level self-healing.

[0012] Furthermore, based on the initial power distribution network structure, the fault propagation process under different fault scenarios is analyzed to determine the hierarchical fault influence domain corresponding to each fault scenario, specifically including: Based on the topology data, extract the connection relationships of each power distribution node and line to generate a basic network topology model; Based on the adjustable power distribution data, each adjustable power source is mapped to the corresponding node in the basic network topology model to generate a power configuration network. Based on the load distribution data, each load node is mapped to the power configuration network to generate the load configuration network; Based on the power configuration network and load configuration network, the power supply connection relationships of each node are associated and integrated to generate the initial power distribution network structure. Based on the initial power distribution network structure, the topological relationships between each node are extracted to generate a power distribution network topology graph. Specifically, after acquiring the topology data, the system reads the node numbers, line numbers, equipment connection relationships, and information on the first and last nodes of each line. Based on the correspondence between the first and last nodes, it establishes the connection relationships between each distribution node and records the connection direction between nodes according to the actual power supply direction of the line. When a line has multiple branches, the system continues to establish the connection relationships between the corresponding branch nodes based on the branch locations, ensuring that each distribution node can form a continuous power supply network according to the actual line connection method. For example, by reading the start and end node information in the line data, the direct connection relationships between different distribution nodes are determined. For lines with branch connections, corresponding node associations are added based on the branch locations, thus forming a network topology model that reflects the basic connection relationships of the distribution network.

[0013] After establishing the connection relationships between distribution nodes, the corresponding distribution nodes are located in the network topology model based on the power access location, equipment type, capacity parameters, and operating status information in the adjustable power distribution data. Information such as power capacity, adjustable power range, and operating status are then associated with the corresponding nodes, ensuring that nodes with adjustable power access also possess the corresponding power support attributes. For example, for an energy storage device connected to a line node, its access node number is matched to associate the energy storage device with the corresponding distribution node; for distributed photovoltaic power sources, the corresponding location association is completed based on its grid connection node, thus forming a power configuration network containing adjustable power information.

[0014] Subsequently, based on the load node location, load capacity, load type, and power supply relationship information in the load distribution data, the corresponding load nodes are located in the power configuration network. Then, according to the correspondence between load access lines, user metering points, and distribution nodes, the load demand information is associated with the corresponding nodes, ensuring that each node simultaneously possesses the load attributes of its corresponding area. For example, by analyzing the power supply node corresponding to the smart meter, user-side load data is associated with the corresponding distribution node, thereby generating a load configuration network containing load distribution information.

[0015] After associating power and load information, the corresponding nodes in the power configuration network and load configuration network are matched according to their node numbers. Line connection relationships, power support capacity, and load demand information are integrated into a single node attribute, and the power supply relationships between nodes are updated synchronously. This ensures that each distribution node simultaneously contains line attributes, power attributes, and load attributes, forming a complete initial distribution network structure. This processing method unifies the previously scattered line data, power data, and load data into a single network model, providing a complete data foundation for subsequent fault propagation analysis.

[0016] Finally, each distribution node in the initial distribution network structure is used as a node element in the topology graph, and the line connections between nodes are used as association edges. The connection direction and sequence of each association edge are recorded according to the actual power supply direction. When there is a direct line connection between two distribution nodes, a corresponding node association is established; when two nodes are connected through multiple intermediate nodes, the connection relationships between each intermediate node are recorded sequentially to form a complete node association path. This results in a distribution network topology graph that reflects the actual topology of the distribution network. Since the topology graph already stores the connection direction, connection sequence, and complete connection path between nodes, the corresponding upstream and downstream connection paths can be directly extracted based on the connection relationships between each node, providing basic data for fault propagation analysis.

[0017] Based on the power distribution network topology diagram, extract the upstream and downstream connection paths corresponding to each node to generate a set of node propagation paths; Based on the set of node propagation paths, and combined with the fault occurrence nodes corresponding to different fault scenarios, an initial fault propagation path is generated. Based on the initial fault propagation path, the fault propagation direction is analyzed layer by layer to generate a set of fault propagation chains; Based on the set of fault propagation chains, the propagation path depth and propagation number of each node are statistically analyzed to generate node propagation impact characteristics. The propagation path depth is the number of connection layers traversed by the fault source node along the fault propagation chain to the target node. Based on the set of fault propagation chains, the propagation connection relationships between each node are statistically analyzed to generate a node propagation connection matrix. Based on the node propagation connection matrix, the fault propagation range is clustered and divided to generate a set of fault-affected areas. Based on the set of affected areas, the number of propagation coverage nodes, maximum propagation path depth, and cumulative propagation times for each area are statistically analyzed. Based on the statistical results, the impact level is classified to generate a graded fault impact domain.

[0018] Understandably, after obtaining the distribution network topology diagram, the possible propagation paths of a fault along the lines are determined by analyzing the connections between the nodes. First, each distribution node serves as the starting point for path analysis. Based on the line connection direction between nodes, upstream and downstream nodes associated with them are searched layer by layer, and the nodes traversed during the search and their corresponding connecting edges are recorded, thus forming the propagation path relationships between different nodes. When different fault locations exist, based on the corresponding position of the faulty node in the topology diagram, paths related to the faulty node are selected from the established propagation path relationships. The faulty node is used as the propagation starting point, and nodes connected to it are used as the initial affected nodes, thus obtaining the initial propagation range for that fault scenario. For example, when a fault occurs at a node in the middle of a feeder, searches can be performed along the upstream power supply path and the downstream load path corresponding to that node to determine the power supply nodes and load nodes that the fault may affect.

[0019] After determining the initial range of fault propagation, the fault's impact range continues to expand outward along the actual connections of the distribution network. During this expansion, the currently identified affected nodes are used as new search nodes to find their unrecorded associated nodes, and these new nodes are added to the fault propagation path. Simultaneously, the sequential connections between nodes are preserved until no new expandable nodes remain. Through continuous iterative expansion, multiple fault propagation chains extending outward from the fault source node can be obtained, describing the possible propagation processes of faults under different line structures. For example, in a distribution network with multiple branch lines, the fault may propagate in multiple directions along different branches. By independently recording each branch path, corresponding multi-directional propagation chains can be formed.

[0020] After obtaining the set of fault propagation chains, the node arrangement relationships in each propagation chain are traversed to extract consecutive node connections, and the propagation chains formed under different fault scenarios are summarized and analyzed. For example, a fault propagation chain represents a path where the fault source node propagates outward through multiple distribution nodes. Adjacent nodes in the path are recorded as a group of propagation association nodes. When two nodes have consecutive connections in multiple propagation chains, the corresponding node association information is retained; when there is no consecutive propagation relationship between two nodes, no corresponding association is established. By statistically analyzing the node connections in all fault propagation chains, the actual node association relationships during fault propagation are converted into a matrix form. Each distribution node is used as a row and column of the matrix. When there is a consecutive fault propagation connection between a corresponding row node and a column node, the existence of a propagation connection is recorded at the corresponding matrix element position; when there is no consecutive fault propagation connection between two nodes, the corresponding matrix element records the absence of a propagation connection, thus forming a node propagation connection matrix. Based on the node propagation connection matrix, clustering is performed according to the node propagation connection relationships recorded in the matrix. Nodes with propagation connections, located on the same feeder branch, and with the same propagation direction are grouped into the same cluster, and each cluster is considered a fault impact region. Nodes without propagation connections or with different propagation directions are assigned to different fault impact regions. For example, if multiple distribution nodes are affected by the same fault source along the same line branch propagation path under different fault scenarios, they can be grouped into the same fault impact region based on their common propagation relationship, while nodes with unrelated propagation paths are assigned to different regions, ultimately forming a set of fault impact regions for subsequent impact level analysis.

[0021] After forming a set of affected areas, the number of affected nodes, the maximum propagation path depth corresponding to each node within the affected area, and the cumulative number of times each node appears in the entire fault propagation chain are statistically analyzed. These statistical results are used as the basis for classifying the impact level of the affected areas. The number of affected nodes represents the node range covered by the fault propagation; the maximum propagation path depth represents the connection levels traversed by the fault to reach the affected area; and the cumulative number of appearances represents the repeated associations of nodes in the affected area across different propagation paths. For different affected areas, the three statistical indicators—the number of affected nodes, the maximum propagation path depth, and the cumulative number of appearances—are ranked according to the corresponding indicators for all affected areas, and the impact level is classified based on the ranking results. Specifically, when all three statistical indicators for a given area are in the top 30% of the ranking, the area is classified as a high-impact affected area; when all three statistical indicators are in the bottom 30% of the ranking, the area is classified as a low-impact affected area; and the remaining affected areas are classified as medium-impact affected areas. The division ratio of the top 30% and bottom 30% can be pre-set according to the scale of the power distribution network, the number of fault scenarios, and planning requirements. This embodiment uses 30% as a preferred implementation method to ensure good differentiation between areas with different impact levels. For example, in multiple fault-affected areas, after statistically ranking the number of affected nodes, maximum propagation path depth, and cumulative occurrence frequency of each area, a certain feeder trunk area is classified as a high-impact fault area because it covers a large number of nodes, has a deep fault propagation level, and the nodes repeatedly appear in multiple fault propagation chains, with all three statistical indicators ranking in the top 30%. On the other hand, a certain terminal branch area contains only a small number of affected nodes, has a short propagation path, and a low cumulative occurrence frequency of nodes, with all three statistical indicators ranking in the bottom 30%. Therefore, it is classified as a low-impact fault area, and the remaining areas are classified as medium-impact fault areas.

[0022] Furthermore, based on the graded fault impact domain, the node recovery priority under different impact levels is determined, a set of node graded recovery constraint relationships is generated, and a collaborative constraint relationship between primary equipment configuration requirements and secondary automation configuration requirements is established, specifically including: Based on the graded fault impact domain, extract the impact level and regional coverage nodes corresponding to each fault impact area, and generate a set of regional nodes. Based on the set of regional nodes, the impact level of each affected area is associated with the regional coverage nodes to generate a set of node recovery levels. Based on the node recovery level set, the node recovery priority is determined according to the impact level of each node, and a node recovery priority sequence is generated. In the same region, the recovery order of each node is determined according to the propagation path depth of the corresponding node. Based on the node recovery priority sequence, the preceding recovery nodes are sequentially associated with their corresponding subsequent recovery nodes to generate a set of node hierarchical recovery constraint relationships; Based on the set of node hierarchical recovery constraint relationships, analyze the power supply recovery mode corresponding to each recovery node, and generate a set of node recovery methods; Based on the set of node recovery methods, each recovery method is mapped to a device type to generate a set of device type requirements; Based on the set of equipment type requirements and the distribution of recovery nodes, the configuration quantity and location of each primary device are planned to generate a set of primary device configuration requirements. Based on the set of primary equipment configuration requirements, determine the control objects corresponding to the primary equipment, generate a set of secondary automation control objects, and configure the corresponding automation terminal types to obtain the set of secondary automation configuration requirements. By associating the set of primary equipment configuration requirements with the set of secondary automation configuration requirements, a collaborative constraint relationship is established between the primary equipment configuration requirements and the secondary automation configuration requirements.

[0023] Understandably, after obtaining the graded fault impact domain, the regional impact results formed during the fault propagation analysis phase are used as the basis for recovery planning, and the recovery order analysis is performed on nodes within different fault impact regions. First, according to the fault impact region division results, nodes belonging to the same region are grouped together, and the impact level information corresponding to the region is retained, so that each node can be associated with the fault impact degree of its region. In actual processing, a correspondence table between regions and nodes can be established to associate and store the fault impact region number, impact level, and node number within the region, allowing subsequent recovery analysis to directly obtain the region to which different nodes belong and their corresponding impact level. For nodes in different impact regions, the recovery level is determined first based on the regional impact level; for multiple nodes within the same region, the propagation path depth recorded during the fault propagation process is further used for sorting, placing nodes with fewer connection levels from the fault source in the propagation path first, and nodes with more connection levels in the propagation path later, so that the recovery process can prioritize nodes with a critical impact on fault propagation. When multiple nodes within the same area have the same propagation path depth, the restoration order is determined by their sequential connection in the power supply direction. If the connection relationship still cannot distinguish the restoration order, the nodes are sorted according to their numerical order to ensure that each restored node has a unique restoration order. For example, in a high-impact area, if node A and node B both belong to the same fault-affected area, with node A having a propagation path depth of 1 and node B having a propagation path depth of 3, then node A is restored first. If node C and node D have the same propagation path depth, the restoration order is determined by their sequential connection in the power supply direction. If the connection relationship still cannot distinguish the restoration order, the node numerical order is used.

[0024] After establishing the node recovery sequence, the ordered node relationships are converted into sequential constraints during the recovery process. When the recovery of a node can relieve the power supply limitations of subsequent nodes or provide recovery conditions for subsequent nodes, that node is designated as the preceding recovery node, and nodes dependent on the recovery result are recorded as subsequent recovery nodes. By organizing the dependencies between different recovery nodes, the node recovery process within the fault area is no longer executed independently, but rather forms a recovery sequence with sequential constraints. For example, for a fault area extending from the main line to multiple branches, the main nodes can be recovered first, and then downstream nodes can be recovered sequentially according to the branch connection relationships, thus forming a recovery process that conforms to the fault propagation law.

[0025] After obtaining the node recovery constraints, the power restoration methods required during node recovery are analyzed based on the recovery sequence of each node and the power supply connections in the distribution network. Specifically, the recovery method is determined by assessing whether the node has a backup power supply path that can connect to other normal power lines, whether there is an adjustable power source to provide power support, and whether the line connection status needs to be changed during the recovery process. For nodes that can regain power by changing line connections, the path adjustment recovery method is used; for nodes where the original power supply path cannot meet the recovery requirements but there is an adjustable power source nearby, the power support recovery method is used; for areas involving both line adjustment and power support, the corresponding recovery method is determined based on the equipment requirements during the recovery process. Through this method, different nodes can match the appropriate recovery method according to their own power supply conditions.

[0026] After determining the node restoration method, the functional requirements involved in the restoration process are converted into equipment configuration requirements. For restoration methods that require changing the line connection status, the corresponding primary equipment type and configuration requirements are determined; for equipment that requires remote control and operation status acquisition, the corresponding secondary automation configuration requirements are further determined. For example, when a pole-mounted sectionalizing switch needs to be configured at a certain line location to achieve power supply path switching during the restoration process, the pole-mounted sectionalizing switch is used as the corresponding control object, and a feeder automation terminal is configured to enable it to have remote control, remote signaling, and operation status feedback capabilities; when a tie switch is configured to achieve backup power switching during the restoration process, the corresponding automation terminal is configured simultaneously so that the tie switch can perform remote opening and closing control according to the restoration strategy.

[0027] Subsequently, based on the equipment type, deployment location, and control requirements of the primary equipment, the corresponding control objects are determined, and the types of automation terminals are matched to establish a correspondence between primary and secondary automation equipment. For example, if a primary switchgear for power supply switching is configured at a certain line location, then that switchgear is designated as the control object, and a corresponding automation terminal is configured to enable it to have remote control and status feedback capabilities. By associating and organizing the primary equipment requirements and secondary automation requirements corresponding to each recovery node, each type of primary equipment corresponds to a corresponding automation control device, maintaining a one-to-one correspondence between deployment location, control object, and control function. When the configuration of primary equipment is adjusted, its corresponding secondary automation configuration is adjusted synchronously, ensuring that the equipment configuration not only meets the fault recovery function requirements but also guarantees that the primary equipment has the corresponding automation control capabilities, ultimately forming a collaborative constraint relationship between the primary equipment configuration requirements and the secondary automation configuration requirements.

[0028] Furthermore, based on the collaborative constraints between primary equipment configuration requirements and secondary automation configuration requirements, and combined with adjustable power distribution data and load distribution data, a set of candidate equipment configuration schemes is generated, specifically including: Based on the collaborative constraint relationship between primary equipment configuration requirements and secondary automation configuration requirements, primary equipment configuration requirements and secondary automation configuration requirements with corresponding relationships are combined to generate a set of equipment collaborative configuration combinations. Based on the set of device collaborative configuration combinations, the adjustable power supply is mapped to the power supply node in the corresponding device collaborative configuration combination to generate a power access configuration set; Based on the power access configuration set, each load node is associated with the corresponding adjustable power source to generate a load power supply configuration set; Based on the load power supply configuration set, the primary equipment and secondary automation equipment that the power supply path passes through are associated to generate the equipment power supply configuration set; Based on the set of equipment power supply configurations, the number of adjustable power supply accesses and the number of load coverage nodes corresponding to each equipment configuration combination are counted to generate a set of equipment power supply configuration features. Based on the set of equipment configuration features, equipment configuration combinations with fewer load coverage nodes than the preset number of coverage nodes are eliminated, and a set of candidate equipment configuration schemes is generated.

[0029] Specifically, after establishing the collaborative constraint relationship between primary equipment configuration requirements and secondary automation configuration requirements, the two types of configuration requirements are matched according to the control object relationship corresponding to the primary equipment. Each type of primary equipment is associated with its corresponding secondary automation equipment, and the equipment type, deployment location, and corresponding control relationship are combined to form a collaborative equipment configuration group. Each configuration group includes the corresponding primary equipment configuration content and the matching secondary automation configuration content. For example, when a line switch is planned for a certain location, the corresponding automation terminal configuration requirement is also included in the same configuration group, ensuring consistency between equipment configuration and automation control capabilities, thus forming a collaborative equipment configuration group. After the combination is completed, each adjustable power supply is mapped to the power supply node in each collaborative equipment configuration group according to the access node, ensuring that different configuration groups can obtain the corresponding power supply access relationship. For example, when the power supply nodes covered by a certain equipment collaborative configuration combination have been connected to energy storage devices or distributed photovoltaics, the corresponding controllable power supply is mapped to the configuration combination to form a power supply access configuration. Subsequently, taking the power supply node corresponding to the controllable power supply as the starting point, along the actual power supply connection relationship of the distribution network, starting from the power access node, the load nodes that maintain a power supply connection with it are searched node by node according to the line connection direction. When there are switching devices between the lines, it is simultaneously determined whether the corresponding switch is in a connection state that can form a power supply path. Load nodes that meet the power supply conditions are all associated with the corresponding controllable power supply, so that each load node corresponds to its power supply source, forming a load power supply configuration.

[0030] After determining the load power supply relationship, the equipment configuration along the power supply path is organized, associating the primary equipment along the path with the corresponding secondary automation equipment to form a complete equipment configuration relationship for each power supply path. For example, when a load node is restored to power by a power supply path, the switching equipment along that path and its corresponding automation terminal are included in the same power supply configuration, forming an equipment power supply configuration. Subsequently, the configuration of each equipment configuration combination is statistically analyzed, calculating the number of controllable power supplies that each configuration combination can connect to and the number of load nodes that it can cover, forming the equipment power supply configuration characteristics. The number of controllable power supplies connected reflects the power supply support capacity of the corresponding configuration combination, and the number of load-covered nodes reflects the range of nodes that the corresponding configuration combination can restore power to. Finally, equipment configuration combinations are screened based on the equipment power supply configuration characteristics. The preset number of covered nodes can be predetermined based on the statistical results of the number of load-covered nodes corresponding to each equipment configuration combination within the planning area. For example, the average number of load-covered nodes corresponding to all equipment configuration combinations, the minimum number of covered nodes required by the plan, or historical planning experience values ​​can be used as the preset number of covered nodes. In practical applications, the corresponding preset method can be selected according to the scale of the planning area, the load distribution, and the planning coverage requirements. When the number of load coverage nodes corresponding to a certain equipment configuration combination is lower than the preset number of coverage nodes, it is considered that the range of nodes that the configuration combination can restore power to is insufficient and will not participate in subsequent scheme optimization; the remaining equipment configuration combinations that meet the coverage requirements are retained as a set of candidate equipment configuration schemes to provide a configuration basis for subsequent power restoration capability evaluation.

[0031] Furthermore, based on the set of candidate device configuration schemes, the power supply recovery capability of each configuration scheme is evaluated under the constraints of different fault impact domains, generating self-healing capability evaluation results, specifically including: Based on the candidate equipment configuration scheme set, obtain the primary equipment configuration and secondary automation configuration corresponding to each candidate equipment configuration scheme, and generate a scheme configuration set; Based on the scheme configuration set, each candidate equipment configuration scheme is mapped to each level of fault influence domain to generate a scheme fault mapping set; Based on the fault mapping set of the schemes, determine the power restoration path of each candidate equipment configuration scheme in the corresponding fault-affected area, and generate a set of scheme recovery paths; Based on the set of recovery paths, identify the power supply nodes and power loads corresponding to each recovery path, and generate a set of recovery features. Based on the recovery feature set of the scheme, the number of recovery coverage nodes, recovery load capacity and recovery completion time corresponding to each candidate equipment configuration scheme are statistically analyzed. Based on the fault mapping set of the scheme, the total number of power outage nodes and the total power outage load capacity corresponding to the graded fault impact domain are obtained. Combined with the number of restored coverage nodes and the restored load capacity, the restored node coverage rate and the restored load coverage rate are calculated. The restoration completion time is then reverse-normalized to generate a power restoration capability feature set. Based on the power restoration capability feature set, the power restoration capability corresponding to each candidate equipment configuration scheme is quantitatively calculated to generate a power restoration capability evaluation value. Based on the power restoration capability evaluation value, generate the self-healing capability evaluation results of the corresponding candidate equipment configuration scheme; Specifically, after candidate equipment configuration schemes are formed, the equipment deployment relationships in each scheme are used as the analysis object for fault recovery capability. Corresponding fault scenarios are loaded for different fault-affected areas, and the primary equipment operation process and power supply path switching process are simulated according to the equipment deployment relationships in the candidate equipment configuration schemes to analyze the recovery status of each power supply node after a fault occurs. First, the primary equipment locations, equipment types, and corresponding secondary automation control relationships configured in the schemes are associated with the affected nodes in the fault-affected domain, ensuring that each candidate scheme corresponds to a complete fault recovery execution condition. During the analysis, based on the affected range of nodes within the fault-affected area and the equipment distribution in the candidate schemes, the power supply range that the scheme can participate in for recovery is determined. Furthermore, according to the connection relationships between devices, a new power supply path that can reach the affected nodes is searched along the line connection direction starting from the normal power supply node after fault isolation. When all primary equipment on the searched path meets the operation conditions, the path is determined as a recoverable power supply path. For example, when a line fault occurs in a certain fault-affected area, based on the switch equipment locations configured in the candidate schemes, it is determined whether a new power supply connection relationship can be formed through equipment switching. Simultaneously, the control capabilities of the automation equipment are considered to determine whether the recovery process has the conditions for remote execution.

[0032] After determining the recoverable power supply paths corresponding to different candidate schemes, the set of recovery paths for each candidate scheme is used as the analysis object. Distribution nodes capable of regaining power are sequentially located along the power supply direction of each recovery path, and load information on the corresponding nodes is associated to identify the recovery power supply nodes and loads corresponding to each recovery path. If multiple recovery paths exist for the same candidate scheme in different fault-affected areas, the recovery power supply nodes and loads corresponding to each recovery path are extracted separately and summarized according to the candidate scheme. This ensures that recovery nodes and loads belonging to the same candidate scheme are uniformly grouped, forming a scheme recovery feature set for the corresponding candidate scheme. For example, if a candidate scheme isolates the fault area by adding switching equipment, two recovery power supply paths can be formed by adjacent power lines. The recovery of distribution nodes and load information corresponding to the two recovery paths is extracted separately and summarized to form a scheme recovery feature set for the candidate scheme. This provides a data foundation for subsequent statistics on the number of restored covered nodes, restored load capacity, and restoration completion time.

[0033] After obtaining the recovery characteristics corresponding to each candidate equipment configuration scheme, a recovery process analysis is performed on each candidate equipment configuration scheme to extract the number of recovery coverage nodes, recovery load capacity, and recovery completion time corresponding to each scheme. Specifically, the number of recovery coverage nodes is obtained by counting the number of nodes that complete power restoration during the recovery process; the recovery load capacity is obtained by summing the load capacities corresponding to the nodes whose power has been restored; and the recovery completion time is obtained by recording the time interval between the time of the fault occurrence and the time when all recovery paths have completed power restoration, with the time corresponding to the last node to complete power restoration being used as the recovery completion time of the candidate scheme. Subsequently, the hierarchical fault influence domain corresponding to each candidate equipment configuration scheme is determined using the scheme fault mapping set, and the number of power-out nodes and power-out load capacity corresponding to each hierarchical fault influence domain are obtained. The total number of power-out nodes is obtained by counting the number of power-out nodes, and the total power-out load capacity is obtained by summing the power-out load capacities. The recovery node coverage rate is calculated by comparing the ratio of the number of restored coverage nodes to the total number of power-out nodes; the recovery load coverage rate is calculated by comparing the ratio of the restored load capacity to the total power-out load capacity; simultaneously, a normalization range is determined based on the maximum and minimum recovery completion times corresponding to all candidate equipment configuration schemes, and the recovery completion time is normalized by range normalization. Since a shorter recovery completion time indicates higher recovery efficiency, a reverse normalization process is applied to the recovery completion time index, ensuring that all normalized evaluation indicators satisfy the evaluation direction that a larger value indicates better scheme performance, resulting in the reverse-normalized recovery time index.

[0034] After processing the above indicators, the recovery node coverage rate, recovery load coverage rate, and the reverse-normalized recovery time indicator are combined to form a power supply recovery capability feature set. Since the recovery node coverage rate and recovery load coverage rate reflect the recovery range of nodes and the degree of load recovery within the fault impact domain, respectively, while the recovery time indicator reflects the timeliness of the recovery process, by converting different types of evaluation indicators into dimensionless feature data, the recovery capabilities corresponding to each candidate equipment configuration scheme can be compared under a unified evaluation scale. Furthermore, the power supply recovery capability feature set is input into the power supply recovery capability evaluation calculation process to calculate the power supply recovery capability evaluation value corresponding to each candidate equipment configuration scheme. Each candidate equipment configuration scheme is then associated with its corresponding power supply recovery capability evaluation value to form a self-healing capability evaluation result. This self-healing capability evaluation result is used to characterize the comprehensive recovery effect of different equipment configuration schemes under the corresponding fault impact level. A higher evaluation value indicates that the scheme can cover more power-loss nodes, restore more load capacity, and has higher recovery efficiency.

[0035] The formula for calculating the power restoration capability evaluation value is as follows: In the formula, This is the evaluation value for power restoration capability. To restore node coverage, To restore load coverage, The recovery completion time after reverse normalization. This represents the imbalance penalty coefficient.

[0036] Understandably, this calculation method comprehensively considers the restoration coverage, load support capacity, and restoration timeliness to quantitatively evaluate the power restoration capabilities of different candidate equipment configuration schemes. Specifically, the restoration node coverage rate reflects the degree of restoration coverage of power-loss nodes within the fault impact domain, demonstrating the scheme's restoration capability for the affected area's node range; the restoration load coverage rate reflects the degree of restoration of power-loss load capacity, demonstrating the scheme's support capability for actual power supply needs; and the inversely normalized restoration time index reflects the efficiency of power restoration after a fault occurs, demonstrating the scheme's rapid restoration capability. Since these evaluation indicators respectively characterize the node restoration range, load restoration scale, and restoration timeliness, they have different evaluation focuses. Inverse normalization of the restoration completion time before calculation ensures that shorter restoration times correspond to higher index values, thus ensuring that each evaluation indicator satisfies the evaluation direction that larger values ​​indicate stronger restoration capabilities, guaranteeing that different candidate equipment configuration schemes can be compared under a unified evaluation scale.

[0037] This calculation relationship does not simply superimpose evaluation indicators. Instead, it comprehensively characterizes the synergistic effect between the restoration range and the restoration power supply scale through the product relationship between the restored node coverage rate and the restored load coverage rate. This ensures that configuration schemes with only a large node restoration range but insufficient load restoration capacity, or those that only restore large load capacity but have limited node coverage, will not receive excessively high evaluations. Simultaneously, the denominator in the calculation relationship reflects the constraint effect of restoration timeliness on the power supply restoration capability evaluation results. This denominator is formed by superimposing a basic constraint term and a restoration timeliness deviation term. The basic constraint term ensures that the evaluation calculation process remains within a stable range, while the restoration timeliness deviation term characterizes the degree of deviation of the current candidate equipment configuration scheme's restoration efficiency from the rapid restoration state. Since the restoration completion time, after inverse normalization, has a larger value, it indicates a faster restoration speed. Therefore, the smaller the restoration timeliness deviation, the weaker the corresponding timeliness constraint effect; conversely, the larger the restoration timeliness deviation, the stronger the corresponding timeliness constraint effect. Thus, by setting the superposition relationship between the basic constraint term and the timeliness deviation term, the denominator can increase as the restoration efficiency decreases, thereby reducing the evaluation value of the corresponding candidate equipment configuration scheme. When a candidate solution can quickly restore power, the recovery time index after inverse normalization approaches its maximum value. At this point, the corresponding timeliness deviation term approaches zero, and the denominator approaches 1, indicating a weak attenuation effect of recovery timeliness on the evaluation results. Conversely, when the recovery time is longer, the inverse normalized recovery time index decreases, the corresponding timeliness deviation term increases, and the denominator increases accordingly, constraining the evaluation results due to recovery timeliness. Furthermore, an exponential adjustment term constrains the difference between the restoration node coverage rate and the restoration load coverage rate. When a large deviation exists, an imbalance penalty coefficient enhances the attenuation effect on the mismatch, ensuring the evaluation results reflect the degree of matching between the restoration node range and the restoration load scale. Specifically, the absolute difference between the restoration node coverage rate and the restoration load coverage rate is calculated. This absolute difference is used as the degree of restoration imbalance for the current candidate equipment configuration scheme. A value of 1 is used as the base penalty constant, and this degree of restoration imbalance is superimposed to obtain the imbalance penalty coefficient corresponding to the current candidate equipment configuration scheme. Since both the restored node coverage rate and the restored load coverage rate fall within the range of 0 to 1, the absolute difference between them varies with the degree of matching between the two indicators. When the restored node coverage rate and the restored load coverage rate are equal, the absolute difference between them is 0, the imbalance penalty coefficient is 1, and the index adjustment term is 1, which does not have a decay effect on the evaluation result. As the difference between the restored node coverage rate and the restored load coverage rate increases, the imbalance penalty coefficient increases synchronously, which strengthens the decay effect of the index adjustment term, thereby reducing the evaluation value of candidate equipment configuration schemes where the scope of restored nodes and the scale of restored loads are significantly mismatched.Therefore, this calculation relationship can comprehensively characterize the power supply recovery capability of candidate equipment configuration schemes from multiple dimensions such as recovery coverage capability, load recovery capability, recovery efficiency, and the degree of matching of recovery capability, providing a quantitative basis for subsequent optimization and selection of equipment configuration schemes.

[0038] Traditional evaluation methods typically assess fault recovery effectiveness based on a single indicator, such as the number of restored nodes, restored load capacity, or restoration time. This approach fails to simultaneously consider the restoration scope, load restoration scale, restoration speed, and the coordination between node and load restoration. This calculation method addresses the mismatch between restoration scope and load support capacity in active distribution network fault recovery. Through the coordinated calculation of restored node coverage and restored load coverage, along with restoration time constraints and restoration capacity matching corrections, it achieves an accurate evaluation of the comprehensive restoration capabilities of different candidate equipment configurations. Each calculation step corresponds to actual fault recovery evaluation needs, avoiding biased evaluation results due to reliance on a single restoration indicator.

[0039] Furthermore, based on the self-healing capability evaluation results, multi-objective comprehensive optimization is performed by combining equipment configuration costs and power restoration coverage to generate the optimal equipment configuration scheme. The network structure and equipment resources of the active distribution network are then coordinated for configuration, outputting comprehensive planning results oriented towards fault-level self-healing, specifically including: Based on the self-healing capability evaluation results, the power supply recovery capability evaluation value corresponding to each candidate equipment configuration scheme is extracted, and then associated and integrated with the corresponding candidate equipment configuration scheme to generate a scheme evaluation set. Based on the scheme evaluation set, the equipment configuration costs corresponding to each candidate equipment configuration scheme are matched to generate a scheme cost set; Based on the cost set of the solutions, the power restoration coverage corresponding to each candidate equipment configuration solution is merged to generate a comprehensive evaluation set of solutions; Based on the comprehensive evaluation set of the schemes, a Pareto multi-objective optimization model is established with the optimization objectives of maximizing the evaluation value of power restoration capability, minimizing equipment configuration cost, and maximizing the coverage of power restoration. Based on the Pareto multi-objective optimization model, the Pareto optimal solution set is obtained by screening, and the optimal equipment configuration scheme is determined from it; Understandably, after obtaining the self-healing capability evaluation results for each candidate equipment configuration scheme, the power supply recovery capability evaluation values ​​of different candidate schemes are correlated and organized with the equipment configuration content of the scheme itself. Specifically, each candidate equipment configuration scheme is treated as an independent evaluation object, and its corresponding power supply recovery capability evaluation value, formed by the number of restored coverage nodes, restored load capacity, and restoration completion time, is bound to the primary equipment type, configuration quantity, deployment location, and secondary automation configuration relationship contained in the scheme, so that each candidate scheme has corresponding recovery capability characteristics and equipment configuration characteristics. For example, if a candidate scheme configures multiple line switching devices and corresponding automation terminals, and can restore multiple load nodes in a certain fault-affected area, then the equipment configuration relationship of the scheme is associated and stored with the corresponding power supply recovery capability evaluation value to form a scheme evaluation set for comprehensive optimization analysis.

[0040] Based on the evaluation set of solutions, the equipment configuration cost information corresponding to each candidate equipment configuration scheme is further introduced to uniformly organize the construction investment of different schemes. Specifically, according to the configuration of primary equipment and secondary automation equipment in the candidate schemes, the quantity and configuration cost of each type of equipment are statistically analyzed, and the cost information generated by equipment procurement, deployment, etc., is matched with the corresponding candidate schemes, so that each candidate scheme corresponds to an equipment configuration cost. Through this process, different candidate schemes can simultaneously reflect fault recovery capabilities and construction investment, forming a scheme cost set. For example, for candidate schemes that include more switching equipment and automation terminals, their equipment configuration costs will increase accordingly, while simplified configuration schemes that only cover some nodes correspond to lower costs.

[0041] Subsequently, a correlation analysis was conducted between the equipment configuration costs and power restoration coverage of each candidate scheme, incorporating the number of nodes that can be restored, the load coverage area, and the corresponding construction costs into the scheme evaluation process. The power restoration coverage, determined by the aforementioned restoration path analysis, represents the affected area that a candidate scheme can restore power in the event of a fault; the equipment configuration cost represents the investment required for scheme implementation. By unifying the restoration capability evaluation value, equipment configuration cost, and power restoration coverage, a comprehensive scheme evaluation set reflecting the overall performance of the scheme is formed. For example, a candidate scheme with higher configuration costs may have a higher overall evaluation result if it covers more fault-affected nodes and possesses higher restoration capabilities, thus distinguishing it from a low-cost scheme with a limited restoration range.

[0042] Based on a comprehensive evaluation set of solutions, a multi-objective optimization model for fault self-healing planning is established. During the optimization process, the power restoration capability evaluation value is used as the improvement objective, ensuring that the selected solutions possess strong fault recovery capabilities; equipment configuration cost is used as the minimization objective, enabling solutions to meet restoration requirements while reducing equipment investment; and power restoration coverage is used as the improvement objective, allowing solutions to cover more power nodes affected by faults. Since there are interdependent relationships between the power restoration capability evaluation value, equipment configuration cost, and power restoration coverage—for example, increasing the number of devices can improve power restoration capability and coverage, but also increases equipment configuration costs—a Pareto multi-objective optimization approach is adopted to comprehensively compare different candidate solutions. This ensures that multiple optimization objectives achieve a balanced result under mutual constraints, avoiding a decline in overall planning effectiveness due to optimizing only a single objective.

[0043] In the specific optimization process, firstly, an initial scheme population is established based on the candidate equipment configuration schemes in the comprehensive evaluation set of schemes. Each candidate equipment configuration scheme is encoded as an independent scheme individual, and the equipment type, configuration quantity, and deployment location in the scheme are used as scheme encoding parameters. Subsequently, objective functions are calculated for each scheme individual in the initial scheme population to obtain the target values ​​corresponding to the power restoration capability evaluation value, equipment configuration cost, and power restoration coverage area. The schemes are then sorted according to the target dominance relationship between different schemes to obtain the current non-dominated scheme set.

[0044] After completing the initial scheme evaluation, the individual schemes in the current non-dominated scheme set are iteratively updated. Specifically, based on the comprehensive scheme evaluation results, schemes with better recovery capabilities, cost control capabilities, and coverage are selected as the basis for updates. Equipment configuration parameters between different schemes are combined and adjusted to generate new candidate equipment configuration schemes. Simultaneously, the number of equipment configurations, equipment deployment locations, or equipment combination relationships in some schemes are randomly adjusted to form new schemes, thereby expanding the scheme search scope. The corresponding power supply recovery capability evaluation value, equipment configuration cost, and power supply recovery coverage are recalculated for the updated candidate equipment configuration schemes, and the current non-dominated scheme set is updated according to the multi-objective dominance relationship. The scheme selection, scheme combination adjustment, scheme parameter update, and non-dominated scheme screening processes are repeated to gradually optimize the candidate equipment configuration schemes towards higher recovery capabilities, lower configuration costs, and wider coverage. When the preset number of iterations is reached, or when the non-dominated scheme set no longer changes significantly during multiple consecutive iterations, the optimization process ends, and the finally retained non-dominated schemes are taken as the Pareto optimal solution set.

[0045] Subsequently, the final equipment configuration scheme is determined from the Pareto optimal solution set based on planning requirements. When there are multiple candidate schemes in the Pareto optimal solution set, the scheme with the highest power restoration capability evaluation value can be selected first; when the power restoration capability evaluation values ​​of multiple schemes are the same, the scheme with the lowest equipment configuration cost is further selected; if the power restoration capability evaluation value and equipment configuration cost are the same, the scheme with the largest power restoration coverage is selected, thus obtaining the optimal equipment configuration scheme that balances restoration effect and configuration cost.

[0046] As a preferred embodiment of the present invention, the above-mentioned Pareto multi-objective optimization solution process can be implemented using an improved non-dominated sorting genetic algorithm. The specific preferred parameter configurations are as follows: the initial population size is set to 100, and the maximum number of iterations is 120; the iteration crossover probability is preferably 0.8, with a possible range of 0.7 to 0.9. This crossover probability is used to control the combination and adjustment of equipment configuration parameters among different scheme individuals described above; the iteration mutation probability is preferably 0.15, with a possible range of 0.05 to 0.2. This mutation probability is used to control the random perturbation update operation of the equipment configuration parameters of the scheme individuals described above; the convergence criterion is set as follows: when the set of non-dominated schemes no longer updates in 15 consecutive iterations, the algorithm is considered converged, and the iteration process can be terminated early. The above parameter configuration is suitable for the application scenario of distribution network fault self-healing equipment configuration planning. It can ensure both the accuracy of scheme optimization and the diversity of population individuals, while also taking into account the efficiency of optimization solution computation, avoiding insufficient optimization due to too few iterations or unnecessary computational consumption due to too many iterations.

[0047] Based on the optimal equipment configuration scheme and the corresponding equipment configuration location, the power supply connection path between each power supply node is adjusted to generate an optimized grid structure; Based on the optimized network structure and the optimal equipment configuration scheme, primary equipment and secondary automation equipment are deployed to the corresponding nodes and line locations to generate equipment resource configuration results; Based on the equipment resource configuration results, the optimized grid structure is updated to generate the collaborative configuration results of the grid structure and equipment resources; Based on the collaborative configuration results of the grid structure and equipment resources, a comprehensive planning result oriented towards fault classification and self-healing is output.

[0048] Specifically, after determining the optimal equipment configuration scheme, the primary equipment and secondary automation equipment configuration locations determined in the scheme are applied to the power distribution network. Existing power supply connections are adjusted to ensure the power supply paths of the distribution network meet fault self-healing requirements. Specifically, based on the deployment nodes and line locations corresponding to each device in the optimal equipment configuration scheme, the connection relationships between power supply nodes are re-analyzed. When the original power supply path cannot meet fault recovery requirements, a new power supply connection path is established according to the equipment locations planned in the optimal equipment configuration scheme. When the original power supply path can be retained, the original connection relationships remain unchanged, and only the power supply connections corresponding to the newly added equipment are supplemented, thus forming an optimized network structure. For example, for areas where the original end load nodes only have a single path of power supply, a backup power supply connection path can be added based on the deployment location of the newly added line switches, so that power supply to the area can be restored by adjusting the power supply path after a fault occurs. For areas that already have reliable power supply capabilities, the existing network structure is maintained, and only the corresponding equipment configuration is supplemented according to the planning results.

[0049] After forming the optimized network structure, primary equipment and secondary automation equipment are deployed to corresponding nodes and line locations according to the optimal equipment configuration scheme. Primary equipment is configured according to the planned locations to meet functional requirements such as fault isolation, power supply path adjustment, and power restoration. Secondary automation equipment is configured according to corresponding controlled objects, ensuring that each primary device requiring control actions corresponds to a specific automation terminal, enabling equipment status acquisition, control command execution, and operational information feedback. During equipment deployment, the locations of primary equipment and corresponding secondary automation equipment maintain a correspondence, ensuring that the equipment configuration not only meets power supply structure requirements but also possesses automation control capabilities. For example, when a line switch is configured at a certain line location, an automation terminal is simultaneously configured at the corresponding location, enabling the line switch to perform remote control and operational status feedback according to the fault recovery strategy, thus forming the equipment resource configuration result.

[0050] After equipment deployment is completed, the deployment results are synchronously updated to the optimized distribution network structure, and the relationships between power supply nodes, lines, and equipment are reorganized. During the update process, the correspondence between equipment and nodes / lines is re-established based on the actual deployment locations of primary equipment, and equipment control relationships are established based on the control objects corresponding to secondary automation equipment. Simultaneously, the power supply relationships between nodes are updated according to the adjusted power supply connection paths, ensuring consistency in equipment configuration, control, and power supply connection relationships, resulting in a coordinated configuration of the network structure and equipment resources. For example, after a new line switch is deployed, it is connected to the corresponding line and node, and its control relationship with the automation terminal is established. The adjusted power supply connection path for that line is also updated, ensuring that the updated distribution network structure accurately reflects the planned equipment deployment, providing a unified network foundation for subsequent fault analysis, power restoration, and self-healing control.

[0051] After completing the coordinated configuration of the power grid structure and equipment resources, the final planning results are compiled and output. The optimized power grid structure, primary equipment configuration results, secondary automation equipment configuration results, and the deployment locations of each device are unified into a comprehensive planning result oriented towards fault-based self-healing. This planning result can intuitively reflect the configuration locations of different devices in the power distribution network, the connection relationships between power supply nodes, and the configuration of automation control, providing a planning basis for subsequent active power distribution network construction, equipment deployment, and fault-based self-healing operation.

[0052] Furthermore, a fault-healing-oriented active distribution network integrated planning system is proposed to implement any of the above-mentioned active distribution network integrated planning methods, including: The data acquisition module is used to acquire topology data, controllable power distribution data, and load distribution data of the active power distribution network, and to construct the initial power distribution network structure. The fault analysis module is used to analyze the fault propagation process under different fault scenarios based on the initial power distribution network structure, and determine the graded fault influence domain corresponding to each fault scenario. The recovery constraint module is used to determine the node recovery priority under different impact levels based on the hierarchical fault impact domain, generate a set of node hierarchical recovery constraint relationships, and establish a collaborative constraint relationship between primary equipment configuration requirements and secondary automation configuration requirements. The scheme generation module is used to generate a set of candidate equipment configuration schemes based on the collaborative constraint relationship between primary equipment configuration requirements and secondary automation configuration requirements, combined with adjustable power distribution data and load distribution data. The capability evaluation module is used to evaluate the power supply recovery capability of each configuration scheme based on a set of candidate equipment configuration schemes under the constraints of different fault influence domains, and generate self-healing capability evaluation results. The optimization configuration module is used to perform multi-objective comprehensive optimization based on the self-healing capability evaluation results, combined with equipment configuration costs and power restoration coverage, to generate the optimal equipment configuration scheme, and to perform coordinated configuration of the grid structure and equipment resources of the active distribution network, outputting comprehensive planning results oriented towards fault-level self-healing.

[0053] Furthermore, the fault analysis module includes: The topology building unit is used to extract the connection relationships of each distribution node and line based on the topology data, generate a basic network topology model, and generate an initial distribution network structure based on the controllable power distribution data and load distribution data. The path analysis unit is used to extract the topological relationships between nodes based on the initial power distribution network structure, generate a power distribution network topology graph, and extract the set of node propagation paths based on the power distribution network topology graph. The fault propagation unit is used to generate an initial fault propagation path based on the set of node propagation paths and the fault occurrence nodes corresponding to different fault scenarios, and to generate a set of fault propagation chains based on the initial fault propagation path. The feature extraction unit is used to statistically analyze the propagation path depth and propagation number of each node based on the fault propagation chain set, generate node propagation impact features, and generate a node propagation connection matrix. The level division unit is used to cluster and divide the fault propagation range based on the node propagation connection matrix, generate a set of fault-affected areas, and divide the impact level according to the statistical results to generate a graded fault impact domain.

[0054] Furthermore, the solution generation module includes: The configuration combination unit is used to combine the primary equipment configuration requirements and secondary automation configuration requirements that have corresponding relationships according to the collaborative constraint relationship between the primary equipment configuration requirements and the secondary automation configuration requirements, and generate a set of equipment collaborative configuration combinations. The power matching unit is used to map the adjustable power supply to the power supply node in the corresponding device collaborative configuration combination based on the device collaborative configuration combination set, and generate a power access configuration set. The load association unit is used to associate each load node with the corresponding adjustable power source according to the power access configuration set, and generate a load power supply configuration set. The equipment configuration unit is used to associate the primary equipment and secondary automation equipment that the power supply path passes through according to the load power supply configuration set, generate the equipment power supply configuration set, and count the number of adjustable power supply access and the number of load coverage nodes corresponding to each equipment configuration combination, and generate the equipment power supply configuration feature set. The scheme filtering unit is used to eliminate equipment configuration combinations with fewer than the preset number of load coverage nodes based on the equipment power supply configuration feature set, and generate a set of candidate equipment configuration schemes.

[0055] Furthermore, the optimized configuration module includes: The capability evaluation unit is used to calculate the number of restoration coverage nodes, restoration load capacity and restoration completion time corresponding to each candidate equipment configuration scheme based on the scheme restoration feature set, calculate the restoration node coverage rate and restoration load coverage rate, and perform reverse normalization on the restoration completion time to generate a power supply restoration capability evaluation value. The scheme integration unit is used to generate the self-healing capability evaluation result of the corresponding candidate equipment configuration scheme based on the power supply restoration capability evaluation value, and to associate and integrate it with the corresponding candidate equipment configuration scheme to generate a scheme evaluation set. The optimization solution unit is used to establish a Pareto multi-objective optimization model based on the comprehensive evaluation set of the scheme, with the optimization objectives of maximizing the evaluation value of power restoration capability, minimizing equipment configuration cost, and maximizing the power restoration coverage. The Pareto optimal solution set is then selected and the optimal equipment configuration scheme is determined from it. The space frame adjustment unit is used to adjust the power supply connection path between each power supply node based on the optimal equipment configuration scheme and the corresponding equipment configuration position, so as to generate an optimized space frame structure. The resource allocation unit is used to deploy primary equipment and secondary automation equipment to corresponding nodes and line locations based on the optimized network structure and the optimal equipment configuration scheme, generate equipment resource allocation results, and output comprehensive planning results oriented towards fault classification and self-healing.

[0056] In summary, the advantages of this invention are as follows: By acquiring topology data, controllable power distribution data, and load distribution data of the active power distribution network, an initial power distribution network structure including network connection relationships, power distribution relationships, and load distribution relationships is constructed. Furthermore, the fault propagation process under different fault scenarios is analyzed to determine the fault impact range and corresponding graded fault impact domains, thus achieving a graded representation of the fault propagation process. By combining the graded fault impact domains to determine the node recovery priority under different impact levels, a set of node graded recovery constraint relationships is generated. Finally, based on the node recovery requirements, a collaborative constraint relationship is established between primary equipment configuration requirements and secondary automation configuration requirements, enabling the equipment configuration process to be synchronized with the fault recovery process. Matching improves the configuration coordination between primary and secondary automation equipment; by combining adjustable power distribution data and load distribution data to generate a set of candidate equipment configuration schemes, and evaluating the power supply recovery capability of each configuration scheme under the constraints of different fault impact domains, the self-healing capability evaluation results are generated by comprehensively considering the number of recovery coverage nodes, recovery load capacity, and recovery completion time, so that the planning scheme can accurately reflect the power supply recovery capability under fault conditions; at the same time, by combining equipment configuration cost and power supply recovery coverage for multi-objective comprehensive optimization, the optimal equipment configuration scheme is determined, and the active distribution network structure and equipment resources are coordinated to improve the fault self-healing capability and comprehensive planning effect of the active distribution network.

[0057] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the claimed invention. The scope of protection claimed by the appended claims and their equivalents is defined.

Claims

1. A comprehensive planning method for active distribution networks oriented towards fault self-healing, characterized in that, include: Acquire topology data, controllable power supply distribution data, and load distribution data of the active power distribution network, and construct the initial power distribution network structure; Based on the initial power distribution network structure, the fault propagation process under different fault scenarios is analyzed to determine the graded fault influence domain corresponding to each fault scenario. Based on the hierarchical fault impact domain, the node recovery priority under different impact levels is determined, a set of node hierarchical recovery constraint relationships is generated, and a collaborative constraint relationship between primary equipment configuration requirements and secondary automation configuration requirements is established. Based on the collaborative constraint relationship between primary equipment configuration requirements and secondary automation configuration requirements, and combined with adjustable power distribution data and load distribution data, a set of candidate equipment configuration schemes is generated. Based on the set of candidate device configuration schemes, the power supply recovery capability of each configuration scheme is evaluated under the constraints of different fault impact domains, and self-healing capability evaluation results are generated. Based on the self-healing capability evaluation results, multi-objective comprehensive optimization is carried out by combining equipment configuration costs and power supply restoration coverage to generate the optimal equipment configuration scheme. The grid structure and equipment resources of the active distribution network are coordinated and configured to output comprehensive planning results for fault-level self-healing.

2. The active distribution network integrated planning method for fault self-healing as described in claim 1, characterized in that, The analysis of the fault propagation process under different fault scenarios based on the initial power distribution network structure, and the determination of the hierarchical fault influence domain corresponding to each fault scenario, specifically includes: Based on the topology data, extract the connection relationships of each power distribution node and line to generate a basic network topology model; Based on the adjustable power distribution data, each adjustable power source is mapped to the corresponding node in the basic network topology model to generate a power configuration network. Based on the load distribution data, each load node is mapped to the power configuration network to generate the load configuration network; Based on the power configuration network and load configuration network, the power supply connection relationships of each node are associated and integrated to generate the initial power distribution network structure. Based on the initial power distribution network structure, the topological relationships between each node are extracted to generate a power distribution network topology graph. Based on the power distribution network topology diagram, extract the upstream and downstream connection paths corresponding to each node to generate a set of node propagation paths; Based on the set of node propagation paths, and combined with the fault occurrence nodes corresponding to different fault scenarios, an initial fault propagation path is generated. Based on the initial fault propagation path, the fault propagation direction is analyzed layer by layer to generate a set of fault propagation chains; Based on the set of fault propagation chains, the propagation path depth and propagation number of each node are statistically analyzed to generate node propagation impact characteristics. The propagation path depth is the number of connection layers traversed by the fault source node along the fault propagation chain to the target node. Based on the set of fault propagation chains, the propagation connection relationships between each node are statistically analyzed to generate a node propagation connection matrix. Based on the node propagation connection matrix, the fault propagation range is clustered and divided to generate a set of fault-affected areas. Based on the set of affected areas, the number of propagation coverage nodes, maximum propagation path depth, and cumulative propagation times for each area are statistically analyzed. Based on the statistical results, the impact level is classified to generate a graded fault impact domain.

3. The active distribution network integrated planning method for fault self-healing as described in claim 1, characterized in that, Based on the hierarchical fault impact domain, the node recovery priority under different impact levels is determined, a set of node hierarchical recovery constraint relationships is generated, and a collaborative constraint relationship between primary equipment configuration requirements and secondary automation configuration requirements is established, specifically including: Based on the graded fault impact domain, extract the impact level and regional coverage nodes corresponding to each fault impact area, and generate a set of regional nodes. Based on the set of regional nodes, the impact level of each affected area is associated with the regional coverage nodes to generate a set of node recovery levels. Based on the node recovery level set, the node recovery priority is determined according to the impact level of each node, and a node recovery priority sequence is generated. In the same region, the recovery order of each node is determined according to the propagation path depth of the corresponding node. Based on the node recovery priority sequence, the preceding recovery nodes are sequentially associated with their corresponding subsequent recovery nodes to generate a set of node hierarchical recovery constraint relationships; Based on the set of node hierarchical recovery constraint relationships, analyze the power supply recovery mode corresponding to each recovery node, and generate a set of node recovery methods; Based on the set of node recovery methods, each recovery method is mapped to a device type to generate a set of device type requirements; Based on the set of equipment type requirements and the distribution of recovery nodes, the configuration quantity and location of each primary device are planned to generate a set of primary device configuration requirements. Based on the set of primary equipment configuration requirements, determine the control objects corresponding to the primary equipment, generate a set of secondary automation control objects, and configure the corresponding automation terminal types to obtain the set of secondary automation configuration requirements. By associating the set of primary equipment configuration requirements with the set of secondary automation configuration requirements, a collaborative constraint relationship is established between the primary equipment configuration requirements and the secondary automation configuration requirements.

4. The active distribution network integrated planning method for fault self-healing as described in claim 1, characterized in that, The process involves generating a set of candidate equipment configuration schemes based on the collaborative constraint relationship between primary equipment configuration requirements and secondary automation configuration requirements, combined with adjustable power distribution data and load distribution data. Specifically, this includes: Based on the collaborative constraint relationship between primary equipment configuration requirements and secondary automation configuration requirements, primary equipment configuration requirements and secondary automation configuration requirements with corresponding relationships are combined to generate a set of equipment collaborative configuration combinations. Based on the set of device collaborative configuration combinations, the adjustable power supply is mapped to the power supply node in the corresponding device collaborative configuration combination to generate a power access configuration set; Based on the power access configuration set, each load node is associated with the corresponding adjustable power source to generate a load power supply configuration set; Based on the load power supply configuration set, the primary equipment and secondary automation equipment that the power supply path passes through are associated to generate the equipment power supply configuration set; Based on the set of equipment power supply configurations, the number of adjustable power supply accesses and the number of load coverage nodes corresponding to each equipment configuration combination are counted to generate a set of equipment power supply configuration features. Based on the set of equipment configuration features, equipment configuration combinations with fewer load coverage nodes than the preset number of coverage nodes are eliminated, and a set of candidate equipment configuration schemes is generated.

5. The active distribution network integrated planning method for fault self-healing as described in claim 1, characterized in that, The power supply recovery capability of each configuration scheme is evaluated under the constraints of different fault influence domains based on the candidate device configuration scheme set, and self-healing capability evaluation results are generated, specifically including: Based on the candidate equipment configuration scheme set, obtain the primary equipment configuration and secondary automation configuration corresponding to each candidate equipment configuration scheme, and generate a scheme configuration set; Based on the scheme configuration set, each candidate equipment configuration scheme is mapped to each level of fault influence domain to generate a scheme fault mapping set; Based on the fault mapping set of the schemes, determine the power restoration path of each candidate equipment configuration scheme in the corresponding fault-affected area, and generate a set of scheme recovery paths; Based on the set of recovery paths, identify the power supply nodes and power supply loads corresponding to each recovery path, and generate a set of recovery features. Based on the recovery feature set of the scheme, the number of recovery coverage nodes, recovery load capacity and recovery completion time corresponding to each candidate equipment configuration scheme are statistically analyzed. Based on the fault mapping set of the scheme, the total number of power outage nodes and the total power outage load capacity corresponding to the graded fault impact domain are obtained. Combined with the number of restored coverage nodes and the restored load capacity, the restored node coverage rate and the restored load coverage rate are calculated. The restoration completion time is then reverse-normalized to generate a power supply restoration capability feature set. Based on the power restoration capability feature set, the power restoration capability corresponding to each candidate equipment configuration scheme is quantitatively calculated to generate a power restoration capability evaluation value. Based on the power restoration capability evaluation value, generate the self-healing capability evaluation results of the corresponding candidate equipment configuration scheme; The formula for calculating the power restoration capability evaluation value is as follows: In the formula, This is the evaluation value for power restoration capability. To restore node coverage, To restore load coverage, The recovery completion time after reverse normalization. This represents the imbalance penalty coefficient.

6. The active distribution network integrated planning method for fault self-healing as described in claim 1, characterized in that, Based on the self-healing capability evaluation results, multi-objective comprehensive optimization is performed by combining equipment configuration costs and power restoration coverage to generate the optimal equipment configuration scheme. Furthermore, the active power distribution network undergoes coordinated configuration of its grid structure and equipment resources, outputting comprehensive planning results oriented towards fault-level self-healing. Specifically, this includes: Based on the self-healing capability evaluation results, the power supply recovery capability evaluation value corresponding to each candidate equipment configuration scheme is extracted, and then associated and integrated with the corresponding candidate equipment configuration scheme to generate a scheme evaluation set. Based on the scheme evaluation set, the equipment configuration costs corresponding to each candidate equipment configuration scheme are matched to generate a scheme cost set; Based on the cost set of the solutions, the power restoration coverage corresponding to each candidate equipment configuration solution is merged to generate a comprehensive evaluation set of solutions; Based on the comprehensive evaluation set of the schemes, a Pareto multi-objective optimization model is established with the optimization objectives of maximizing the evaluation value of power restoration capability, minimizing equipment configuration cost, and maximizing the coverage of power restoration. Based on the Pareto multi-objective optimization model, the Pareto optimal solution set is obtained by screening, and the optimal equipment configuration scheme is determined from it; Based on the optimal equipment configuration scheme and the corresponding equipment configuration location, the power supply connection path between each power supply node is adjusted to generate an optimized grid structure; Based on the optimized network structure and the optimal equipment configuration scheme, primary equipment and secondary automation equipment are deployed to the corresponding nodes and line locations to generate equipment resource configuration results; Based on the equipment resource configuration results, the optimized grid structure is updated to generate the collaborative configuration results of the grid structure and equipment resources; Based on the collaborative configuration results of the grid structure and equipment resources, a comprehensive planning result oriented towards fault classification and self-healing is output.

7. A fault-healing-oriented active distribution network integrated planning system, used to implement the active distribution network integrated planning method as described in any one of claims 1-6, characterized in that, include: The data acquisition module is used to acquire topology data, controllable power distribution data, and load distribution data of the active power distribution network, and to construct an initial power distribution network structure. The fault analysis module is used to analyze the fault propagation process under different fault scenarios based on the initial power distribution network structure, and determine the graded fault influence domain corresponding to each fault scenario. The recovery constraint module is used to determine the node recovery priority under different impact levels based on the hierarchical fault impact domain, generate a set of node hierarchical recovery constraint relationships, and establish a collaborative constraint relationship between primary equipment configuration requirements and secondary automation configuration requirements. The scheme generation module is used to generate a set of candidate equipment configuration schemes based on the collaborative constraint relationship between primary equipment configuration requirements and secondary automation configuration requirements, combined with adjustable power distribution data and load distribution data. The capability evaluation module is used to evaluate the power supply recovery capability of each configuration scheme based on a set of candidate equipment configuration schemes under the constraints of different graded fault influence domains, and generate self-healing capability evaluation results. The optimization configuration module is used to perform multi-objective comprehensive optimization based on the self-healing capability evaluation results, combined with equipment configuration costs and power supply restoration coverage, to generate the optimal equipment configuration scheme, and to perform coordinated configuration of the grid structure and equipment resources of the active distribution network, outputting comprehensive planning results oriented towards fault-level self-healing.

8. The active distribution network integrated planning system for fault self-healing according to claim 7, characterized in that, The fault analysis module includes: The topology building unit is used to extract the connection relationships of each distribution node and line based on the topology data, generate a basic network topology model, and generate an initial distribution network structure based on the controllable power distribution data and load distribution data. The path analysis unit is used to extract the topological relationships between nodes based on the initial power distribution network structure, generate a power distribution network topology graph, and extract a set of node propagation paths based on the power distribution network topology graph. The fault propagation unit is used to generate an initial fault propagation path based on the set of node propagation paths and the fault occurrence nodes corresponding to different fault scenarios, and to generate a set of fault propagation chains based on the initial fault propagation path. The feature extraction unit is used to statistically analyze the propagation path depth and propagation number of each node according to the fault propagation chain set, generate node propagation impact features, and generate a node propagation connection matrix. The level division unit is used to cluster and divide the fault propagation range based on the node propagation connection matrix, generate a set of fault-affected areas, and divide the impact level according to the statistical results to generate a graded fault impact domain.

9. A fault-healing-oriented active distribution network integrated planning system according to claim 7, characterized in that, The scheme generation module includes: A configuration combination unit is configured to combine primary equipment configuration requirements and secondary automation configuration requirements that have corresponding relationships according to the collaborative constraint relationship between primary equipment configuration requirements and secondary automation configuration requirements, and generate a set of equipment collaborative configuration combinations. A power matching unit is used to map an adjustable power supply to a power supply node in a corresponding device collaborative configuration combination based on a set of device collaborative configuration combinations, thereby generating a power access configuration set. A load association unit is used to associate each load node with a corresponding adjustable power source according to a power access configuration set, thereby generating a load power supply configuration set. The equipment configuration unit is used to associate the primary equipment and secondary automation equipment that the power supply path passes through according to the load power supply configuration set, generate the equipment power supply configuration set, and count the number of adjustable power supply access and the number of load coverage nodes corresponding to each equipment configuration combination, and generate the equipment power supply configuration feature set. The scheme filtering unit is used to generate a candidate equipment configuration scheme set by eliminating equipment configuration combinations with fewer than a preset number of load coverage nodes based on the equipment power supply configuration feature set.

10. A fault-healing-oriented active distribution network integrated planning system according to claim 7, characterized in that, The optimized configuration module includes: The capability evaluation unit is used to calculate the number of recovery coverage nodes, recovery load capacity and recovery completion time corresponding to each candidate equipment configuration scheme according to the scheme recovery feature set, calculate the recovery node coverage rate and recovery load coverage rate, and perform reverse normalization on the recovery completion time to generate a power supply recovery capability evaluation value. The scheme integration unit is used to generate the self-healing capability evaluation result of the corresponding candidate equipment configuration scheme based on the power supply restoration capability evaluation value, and to associate and integrate it with the corresponding candidate equipment configuration scheme to generate a scheme evaluation set. The optimization solution unit is used to establish a Pareto multi-objective optimization model based on the comprehensive evaluation set of the schemes, with the optimization objectives of maximizing the evaluation value of power restoration capability, minimizing equipment configuration cost, and maximizing the coverage of power restoration as the optimization objectives, to screen out the Pareto optimal solution set and determine the optimal equipment configuration scheme from it. A grid adjustment unit is used to adjust the power supply connection path between each power supply node based on the optimal equipment configuration scheme and the corresponding equipment configuration position, so as to generate an optimized grid structure. The resource configuration unit is used to deploy primary equipment and secondary automation equipment to corresponding nodes and line locations based on the optimized network structure and the optimal equipment configuration scheme, generate equipment resource configuration results, and output comprehensive planning results oriented towards fault classification and self-healing.

Images