Three-network coordination power grid recovery method and system based on complex network theory

Abstract

The application provides a three-network coordination power grid recovery method and system based on complex network theory, and belongs to the technical field of power systems.The scheme effectively solves the problem that the traditional scheme only focuses on the fault propagation within a single network, and the simulation result and the actual physical condition have a large difference; and due to the single structure of the current urban power station, the existing power grid recovery strategy mostly only considers the simple traditional power station, has poor recovery ability in response to disasters, weak disaster resistance, and insufficient flexibility of the recovery strategy.

Description

Technical Field

[0001] This invention belongs to the field of power system technology, and in particular relates to a method and system for the restoration of a three-network coordinated power grid based on complex network theory. Background Technology

[0002] The statements in this section are merely background information related to the present invention and do not necessarily constitute prior art.

[0003] Currently, most research on power grid recovery focuses only on the power grid itself, or considers a two-layer network structure by incorporating communication networks, but lacks simulation studies on the integration of power grid, communication network, and transportation network. Given the deep coupling between power grid and information and communication networks today, defects in the communication network structure or deliberate attacks can cause major power outages. Similarly, congestion in the transportation network can significantly delay the power grid's recovery rate. Therefore, modeling only single-layer or two-layer networks is not accurate enough for simulating real-world physical conditions.

[0004] Meanwhile, existing solutions treat all nodes equally and restore them simultaneously when restoring failed nodes in the power grid, which is not only inefficient but also fails to effectively guarantee social benefits. Due to the lack of consideration for the combined impact of the three networks, traditional cascading failure models often only focus on the propagation of faults within a single network in terms of fault simulation. The simulation results differ significantly from the actual physical situation, meaning that the simulation results are not accurate enough and have large errors. Because urban power plants currently have a simple structure, existing power grid restoration strategies only consider traditional power plants, resulting in poor disaster recovery capabilities, weak disaster resistance, and inflexible strategies. Summary of the Invention

[0005] To overcome the shortcomings of the existing technologies, this invention provides a three-network coordinated power grid restoration method and system based on complex network theory. This addresses the problem that traditional solutions only focus on fault propagation within a single network, resulting in significant discrepancies between simulation results and actual physical conditions. Furthermore, due to the simple structure of current urban power plants, most existing power grid restoration strategies only consider traditional power plants, leading to poor disaster recovery capabilities, weak disaster resistance, and inflexible restoration strategies.

[0006] According to a first aspect of the present invention, a three-network coordinated power grid restoration method based on complex network theory is provided, comprising:

[0007] Acquire power grid site data, as well as corresponding transportation site data and communication site data within a preset area, and construct a three-network coupling network of power grid, transportation network and communication network;

[0008] For nodes in a three-network coupled network, a pre-constructed three-network collaborative node importance evaluation system is used to obtain the node importance score for each node. Specifically, the three-network collaborative node importance evaluation system involves calculating the overall power grid importance of the corresponding power grid station, the overall transportation network importance of the corresponding transportation station, and the overall communication network importance of the corresponding communication station. The node importance score is obtained by weighted summation of the overall power grid importance, the overall transportation network importance, and the overall communication network importance.

[0009] In response to the occurrence of power grid site faults, identify the fault nodes in the three-network coupled network and the fault nodes caused by cascading faults;

[0010] For a identified fault node, the power grid fault node is restored based on a preset cascaded failure recovery strategy. Specifically, the cascaded failure recovery strategy is as follows: based on the node importance score of the fault node, important nodes are restored first, and the restoration process is achieved in coordination by generator sets, distributed power sources and mobile emergency power sources.

[0011] Furthermore, the calculation of the overall importance of the power grid station corresponding to the node, the overall importance of the transportation network station corresponding to the node, and the overall importance of the communication network station corresponding to the node are specifically as follows: the overall importance of the power grid station is determined based on the station level, station size, load level, and load size of the power grid station; the overall importance of the communication network station is determined based on the degree centrality, proximity centrality, and betweenness centrality of the communication network station; and the overall importance of the transportation network station is determined based on the degree centrality, proximity centrality, station proximity, passenger flow, and population size of the transportation network station.

[0012] Furthermore, the comprehensive importance of the power grid, the comprehensive importance of the transportation network, and the comprehensive importance of the communication network are specifically expressed as follows:

[0013] Overall importance of the power grid:

[0014] Overall Importance of Communication Networks: P c (v i ) = DC i +CC i ;

[0015] Overall Importance of Transportation Network:

[0016] Overall importance of nodes in the three networks:

[0017] Among them, P e (v i ) is the network node v iThe overall importance of the power grid at the site, d(v i CC represents the influence value of the influencing factors for each site. i For node proximity centrality, DC i For the degree centrality of nodes, P i To assess the overall importance of nodes in the three networks, and These are the weighting coefficients for different transportation networks. and LL represents the weighting coefficients for the power grid, communication network, and transportation network. i Let M be the node topology metrics for a given site. i For a given station, N represents its own attribute indicators. i For a specific site, the regional environmental indicators and...

[0018] Furthermore, in response to the occurrence of a power grid site fault, determining the fault node in the three-network coupled network and the fault node caused by cascading faults specifically involves: determining the corresponding node in the three-network coupled network based on the location of the faulty power grid site; and simulating cascading failure of the coupled network in the three-network coupled network based on a preset load redistribution strategy to determine the fault node caused by cascading faults.

[0019] Furthermore, the load redistribution strategy specifically involves: constructing the initial load of each node based on the node importance score of each node in the three-network coupled network; determining the load capacity of the node based on the initial load; and for a faulty node, transferring its load to a node adjacent to the current faulty node according to a preset load distribution rule, and removing the current faulty node and its corresponding edge.

[0020] Furthermore, the process of coordinated restoration using generator sets, distributed power sources, and mobile emergency power sources is as follows:

[0021] Node priority allocation principle: The priority of different types of power sources is determined based on the characteristics of different power sources and the comprehensive importance score of the three networks;

[0022] By combining the power supply's response speed and operating characteristics, a three-stage coordinated mechanism of "emergency-transition-stability" is constructed.

[0023] Integrate real-time data from the power grid, communication network, and transportation network, and determine power dispatching schemes based on the linkage of data from these three networks;

[0024] Optimize the total lifecycle cost for different power supplies;

[0025] The above four strategies are combined to form a closed-loop strategy of "dynamic assessment - tiered response - flexible adjustment" for coordinated recovery.

[0026] Furthermore, the node priority classification method for different types of power sources is as follows: mobile emergency power sources are prioritized for deployment to the nodes with the highest scores in the three networks; distributed power sources provide support to the next more important nodes nearby to form local microgrids; and traditional generator sets are centrally used to restore the backbone network.

[0027] According to a second aspect of the present invention, a three-network coordinated power grid restoration system based on complex network theory is provided, comprising:

[0028] The data acquisition unit is used to acquire power grid station data, as well as corresponding transportation station data and communication station data within a preset area, and to construct a three-network coupling network of power grid, transportation network and communication network.

[0029] An importance scoring unit is used to obtain the node importance score for each node in a three-network coupled network using a pre-constructed three-network collaborative node importance evaluation system. Specifically, the three-network collaborative node importance evaluation system involves calculating the overall power grid importance of the corresponding power grid station, the overall transportation network importance of the corresponding transportation station, and the overall communication network importance of the corresponding communication station. The node importance score is obtained by weighted summation of these three overall importance scores.

[0030] The fault node determination unit is used to determine the fault nodes in the three-network coupled network and the fault nodes caused by cascading faults in response to the occurrence of power grid site faults.

[0031] The power grid restoration unit is used to restore the power grid fault nodes based on a preset cascaded failure restoration strategy for a given fault node. The cascaded failure restoration strategy specifically involves prioritizing the restoration of important nodes based on their node importance scores. The restoration process is achieved through the coordinated use of generator sets, distributed power sources, and mobile emergency power sources.

[0032] According to a third aspect of the present invention, an electronic device is provided, including a memory, a processor, and a computer program stored in the memory and running on the memory, wherein the processor executes the program to implement the aforementioned three-network coordinated power grid restoration method based on complex network theory.

[0033] According to a fourth aspect of the present invention, a non-transitory computer-readable storage medium is provided, on which a computer program is stored, which, when executed by a processor, implements the aforementioned three-network coordinated power grid restoration method based on complex network theory.

[0034] The above one or more technical solutions have the following beneficial effects:

[0035] This invention provides a method and system for power grid restoration based on complex network theory. The scheme constructs an index system for evaluating the importance of nodes based on complex network theory, providing a scientific evaluation strategy for nodes in the power grid, communication network, and transportation network, thereby ensuring the priority restoration of important nodes. Through complex network theory, the scheme in this embodiment can quantify the importance of nodes in different networks (power grid, communication network, and transportation network), using multiple indicators to evaluate the role of nodes in the network, especially those nodes that may become critical hubs or bottlenecks. This evaluation method helps identify key nodes in the network and ensures their priority restoration. Simultaneously, considering the complex interdependencies between the power grid, communication network, and transportation network, the scheme in this embodiment constructs a multi-layer network (i.e., a three-network coupled network). This network enables comprehensive analysis of the interactions between different networks, providing more comprehensive information for the identification of key nodes.

[0036] The present invention provides a cascading failure simulation strategy that can accurately predict how a single node failure propagates between different networks and triggers a chain reaction. By simulating the failure propagation path and the scope of impact, it can effectively identify the faulty nodes caused by cascading failures and effectively predict potential large-scale network collapses.

[0037] The solution described in this invention can significantly improve the flexibility and reliability of the entire system by using mobile emergency power supplies. In the event of a disaster, these energy storage units can respond quickly, provide additional power support, and ensure the continuous operation of critical nodes.

[0038] Advantages of additional aspects of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0039] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention.

[0040] Figure 1 This is a schematic diagram of the overall framework of the three-network coordinated power grid restoration method based on complex network theory described in this embodiment of the invention;

[0041] Figure 2 This is a schematic diagram of the triple-play topology architecture described in an embodiment of the present invention;

[0042] Figure 3 This is a diagram of the urban power grid network model architecture described in this embodiment of the invention;

[0043] Figure 4This is a network model architecture diagram of the urban transportation network described in this embodiment of the invention;

[0044] Figure 5 This is a network model architecture diagram of the urban communication network described in this embodiment of the invention;

[0045] Figure 6 This is a simulation diagram of the IEEE 30-node system described in this embodiment of the invention;

[0046] Figure 7 This is a load redistribution diagram of the coupled network nodes described in this embodiment of the invention;

[0047] Figure 8 This is a schematic diagram illustrating the simulation process of the cascaded failure of the coupled network as described in this embodiment of the invention.

[0048] Figure 9 This is a schematic diagram illustrating the scenario of restoring power supply to the coupled network fault zone using only traditional generator sets in the distribution network, as described in this embodiment of the invention.

[0049] Figure 10 This is a schematic diagram illustrating the fault partitioning of the distributed power supply access coupling network as described in an embodiment of the present invention;

[0050] Figure 11 This is a schematic diagram illustrating the fault partitioning of the coupled network for mobile emergency power supplies and distributed power supplies as described in this embodiment of the invention.

[0051] Figure 12 This is a schematic diagram illustrating the simulation results of cascading failure without a recovery strategy as described in this embodiment of the invention.

[0052] Figure 13 This is a schematic diagram illustrating the simulation results of power restoration cascade failure and node recovery using only traditional generator sets as described in this embodiment of the invention.

[0053] Figure 14 This is a schematic diagram illustrating the simulation results of cascade failure and node recovery under the condition of distributed power supply access in the embodiments of the present invention;

[0054] Figure 15 This is a schematic diagram illustrating the simulation results of simultaneous failure and node recovery of mobile emergency power supply and distributed power supply access cascade as described in this embodiment of the invention.

[0055] Figure 16 This is a schematic diagram showing the simulation comparison results of cascade failure and node recovery as described in the embodiments of the present invention. Detailed Implementation

[0056] It should be noted that the following detailed descriptions are exemplary and intended to provide further illustration of the invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0057] It should be noted that the terminology used herein is for the purpose of describing particular implementations only and is not intended to limit the exemplary implementations of the present invention.

[0058] Where there is no conflict, the embodiments and features in the embodiments of the present invention can be combined with each other.

[0059] In view of the important roles that communication networks and transportation networks play in power system accidents, and the coupling relationship between power grid, communication network and transportation network, the solution described in this embodiment proposes a three-network coordinated power grid recovery method based on complex network theory. It realizes the recovery of important urban loads and the time-sharing scheduling of distributed energy storage through intelligent coordinated control of power grid, transportation network and communication network. It aims to ensure the power supply rate of core loads, promote the rapid and stable reconstruction of the power grid after disaster, and realize the rapid, intelligent and coordinated scheduling of urban power supply system after extreme weather disasters.

[0060] In one or more embodiments, such as Figure 1 As shown, this embodiment provides a three-network coordinated power grid restoration method based on complex network theory, specifically including the following processing steps:

[0061] Step 1: Obtain power grid site data, as well as corresponding transportation site data and communication site data within the preset area, and construct a three-network coupling network of power grid, transportation network and communication network;

[0062] In specific implementation, such as Figure 2 As shown, the construction of the three-network coupled network model is specifically as follows:

[0063] In the initial stages of power grid cascade fault analysis, most studies focused on modeling and analyzing fault dynamics in a single, non-interacting power grid environment. To explore the propagation characteristics and robustness of cascade faults in smart grids coupled with communication and transportation networks, a three-layer network model of power grid-communication network-transportation network is first established based on the single-layer power grid cascade failure model. Compared with traditional single-layer and two-layer network models, this model considers the physical characteristics and functions of nodes in the three-layer network (power grid, communication network, and transportation network), as well as the coupling relationships between different network layers, and can reflect the cascade propagation characteristics of actual smart grids to a certain extent.

[0064] According to the coupling relationship between the three networks, such as Figure 2As shown, a three-network coupled topology diagram is constructed. The interaction between the power grid and the transportation network is mainly through charging stations as interfaces. The charging behavior of electric vehicles and the power grid's power supply to the charging stations constitute the interaction process. The coupling relationship between communication network nodes and power grid nodes changes dynamically with the system's operating status and is divided into two parts: information transmission process and power supply process. The interaction process between the communication network and the transportation network is similar to that between the communication network and the power grid, but only the communication network dispatch center collects real-time information from the transportation network and issues charging control strategies.

[0065] (1) Construction of the basic network model, specifically:

[0066] ① Single-layer network

[0067] Complex networks can be formed by abstracting representations of the real world, and can be represented using graph theory. A specific single-layer network can typically be represented by a set G. D The graph is represented by (V, E). Based on graph theory, V is the set of vertices, and its elements are called vertices / nodes; E is a multiset of V*V, called the edge set.

[0068] Network A can be represented as G = (V A E A Among them, V A ={1, 2, ..., N A Let V be the set of the number of nodes in network A, where V is the number of nodes in network A. A E represents the total number of nodes. A ={e ij Let} be the set of connecting edges in network A; if any two nodes in network A are connected, then e ij =1, otherwise e ij =0. Therefore, the network can be represented as:

[0069] V = {v} i |i=1,2,3…N} (1),

[0070] E={e ij |i=1, 2, 3…N, j=1, 2, 3…N, i≠j} (2),

[0071]

[0072] Among them, v i v j -- Two arbitrary nodes in the network; e ij -- An edge connecting any two nodes in a network; E(G) -- The set of edges connecting network nodes.

[0073] Based on complex network theory, a network can be abstracted as an unweighted, undirected network, whose topology can be represented by an adjacency matrix: A = (a ij ) n*n ;

[0074] A single edge attribute may not be able to represent some phenomena in real life. For example, in urban transportation networks, some major traffic sections are more prone to congestion during peak hours, and some important transmission lines in power grids are more prone to failure. In this case, unweighted complex networks cannot accurately describe these phenomena, but weighted networks can be used to represent them.

[0075] Weighted undirected and weighted directed network models, when starting from node v i There is a node v j When connecting edges, a in the adjacency matrix ij The value is the weight of the connected edge, corresponding to the adjacency matrix element a of the model. ij for:

[0076]

[0077] ② Three-layer network

[0078] Single-layer complex network models have limitations in describing many real-world systems because many real-world systems typically contain multiple different types of relationships. To more accurately represent these complex real-world situations, many scholars have extended the concept of single-layer complex networks to multi-layer complex networks and proposed corresponding theoretical frameworks.

[0079] Compared to traditional single-layer complex networks, multi-layer complex network models are better able to capture the interactions and dependencies between different layers within a system. Essentially, such models are composed of multiple single-layer networks connected in different ways. Each inter-layer connection method represents a specific type of multi-layer network structure and also influences the overall network's behavior.

[0080] For a multi-layer interconnected network with L (L=3) layers, the scheme described in this embodiment considers the convergence of the power grid, communication network, and transportation network. Using graph theory, the multi-layer network can be represented as M=(G,E), where G=(V... N R N ) represents the set of nodes and edges in a layered network. Represents the set of edges connecting layers;

[0081] Multi-layer networks have different types of adjacency matrices. For the adjacency matrix of each single-layer subnet, each element... The expression is:

[0082]

[0083] For each element of the adjacency matrix of inter-layer interconnection The expression is:

[0084]

[0085] in,

[0086] Multilayer fusion networks can be mathematically represented by an adjacency matrix A. N*N Express it, where each element a ij The expression is:

[0087]

[0088] Among them, a ij It no longer simply represents the edge weights of a single-layer network, but rather the cumulative sum of the edge weights of a single-layer network under the context of triple-play integration.

[0089] The specific calculation process is as follows: For nodes i and j within the layer: if i and j in the power grid are connected, then a 1 ij The value is 1; if there is no connection between i and j in the communication network, then a 2 ij The value of a is 0; if i and j are connected in the transportation network, then a 3 ij The value is 1;

[0090] For nodes i and j between layers: if there is a connection between node i in the power grid and node j in the communication network, i.e., a 12 ij The value is 1; there is no connection between node i in the power grid and node j in the transportation network, i.e., a 13 ij The value is 0; there is no connection between node i in the transportation network and node j in the communication network, i.e., a 23 ij The value is 0;

[0091] Therefore, in the fused network, a ij The value is 3.

[0092] (2) Construction of the power grid-transportation network-communication network model

[0093] The importance of nodes in the model is evaluated using a node importance assessment system, and based on this, the locations of distributed energy storage devices in the power grid and mobile resource garages in the transportation network are determined. For example... Figures 3-5 The diagram illustrates the abstract representation of a simulated city as a complex network, in which... Figure 3 This is a diagram of the urban power grid network architecture. Figure 4 A network architecture diagram of the urban transportation network; Figure 5This is a network model architecture diagram of a city communication network, which consists of several edges and points;

[0094] like Figure 6 As shown, the following explanation uses the IEEE 30-node system as an example. The IEEE 30-node system includes one wind farm (new energy power generation) node, six traditional generator nodes, one photovoltaic power station node, and 22 load nodes.

[0095] After assigning and evaluating the importance of each communication network node in the city using the power grid importance evaluation model, the transportation network importance identification model, and the communication network importance identification model respectively, the evaluation results of the three-network integrated node importance identification model for 22 load nodes are obtained, as shown in Tables 1-4:

[0096] Table 1. Evaluation of 22 load nodes based on the power grid importance evaluation model.

[0097]

[0098]

[0099] Table 2 shows the results of the load node evaluation based on the transportation network importance identification model.

[0100]

[0101]

[0102] Table 3. Results of the evaluation of communication network nodes in the city based on the communication network importance identification model.

[0103]

[0104]

[0105] Table 4 shows the evaluation results of 22 load nodes based on the three-network integrated node importance identification model.

[0106]

[0107]

[0108] Taking all factors into consideration, nodes 6, 10, 12, and 19 are identified as critical nodes in the context of triple-play integration. Based on these critical nodes, photovoltaic power plants and wind farms are included as distributed power sources, while mobile power reserves are established in key areas as reserve power resources. Specifically, nodes 1, 2, 5, 8, 11, and 13 are traditional generator sets, node 29 is a photovoltaic power plant, and node 9 is a wind farm. Since nodes 6, 10, 12, and 19 are critical nodes, they are designated as mobile power access nodes to facilitate faster grid recovery. Meanwhile, nodes 3, 25, and 27, located on the periphery, cannot be repaired immediately in the event of a grid failure; therefore, they are also designated as mobile power access nodes. In this embodiment, mobile power reserves are established in suburban and urban periphery areas to facilitate timely dispatch in the event of grid failures.

[0109] Step 2: For nodes in the three-network coupled network, a pre-constructed three-network collaborative node importance evaluation system is used to obtain the node importance score for each node. Specifically, the three-network collaborative node importance evaluation system involves calculating the overall power grid importance of the corresponding power grid station, the overall transportation network importance of the corresponding transportation station, and the overall communication network importance of the corresponding communication station. The node importance score is obtained by weighted summation of the overall power grid importance, the overall transportation network importance, and the overall communication network importance.

[0110] In practical implementation, the construction of the node importance evaluation system includes the following processes:

[0111] When assessing the importance of nodes, a multi-indicator comprehensive node importance evaluation system is established, taking into account the power grid, transportation network, and communication network. This system includes:

[0112] (1) Power grid importance evaluation model

[0113] Considering the impact of different power grid stations on the power grid, station level, station size, station load level, and load size are used as factors of importance of power grid nodes.

[0114] The importance of a substation in the network varies depending on its level and size. Substations with higher voltage levels are more important than those with lower voltage levels; key hub stations are more important than regional stations. Similarly, substations with higher load levels and larger loads are more important than similar substations. Furthermore, substations located far from core substations and without nearby distributed power sources will cause power outages in the surrounding area if they fail; therefore, such substations should be considered separately. The evaluation indicators for the impact of power grid substations on nodes are shown in Table 5, and the influencing factors and their values ​​are shown in Table 6.

[0115]

[0116]

[0117] Table 5 Evaluation Indicators of the Impact of Power Grid Stations on Nodes

[0118] Table 6. Influencing Factors and Their Values ​​for Each Site

[0119]

[0120] The importance of the node's grid influence P is calculated using formula (8) based on the level and scale of the site where the node is located, as well as the level and size of the site's load. e (v i ):

[0121]

[0122] Among them, P e (v i ) is the network node v i The importance of the power grid impact at each site; the influence value of the influencing factors at each site is d(v). i ).

[0123] (2) Communication Network Importance Evaluation Model

[0124] A city communication network model is constructed based on complex network theory, and the importance of nodes is quantified by the degree centrality and the proximity centrality of nodes.

[0125] The degree centrality evaluation criterion for nodes is as follows: Degree centrality is the simplest method for determining node importance in network analysis. The more neighboring nodes a node has, the higher its degree value, and the more important the node is in the network. Degree centrality is an identification method based on local information from its nearest neighbors.

[0126] The proximity centrality evaluation criterion for nodes is as follows: proximity centrality reflects the degree of proximity between a node and other nodes in a network. From the perspective of information propagation in the network, the smaller the average distance between nodes, the easier it is for that node to reach other nodes, the more efficient the information propagation, and the more important its position; numerically, this is reflected in a higher calculated value. Calculate the overall importance P of the communication network. c (v i ):

[0127] P c (v i ) = DC i +CC i (9),

[0128] in, Let k represent the degree centrality of the nodes, where N is the total number of nodes in the network; k i It is the degree of a node, representing the number of other nodes connected to that node. Let d represent the closeness centrality of the nodes, where N is the number of nodes; jj It is the shortest path length between node i and node j; The sum of the shortest path lengths from node i to all other nodes.

[0129] (3) Transportation Network Importance Evaluation Model

[0130] A city rail transit network model is constructed based on complex network theory, and the importance of nodes is quantified by topological indicators, station attribute indicators and regional environmental indicators.

[0131] Among them, node topology: calculated and statistically obtained based on complex network theory, reflecting various characteristic attributes of nodes in the network; node topology includes the degree centrality (DC) of nodes. i / b i1 and the proximity centrality CC of nodes i / b i2 ;

[0132] Station attributes: The relevant technical parameters after the station is completed and the relevant data indicators after it is put into operation, reflecting the actual physical characteristics of the transportation station after it is built.

[0133] Station attributes include station proximity L i / b i3 And passenger flow D i / b i4 ;

[0134] Regional environment: Indicators reflecting the impact of the station on surrounding residents and the environment after its construction, demonstrating the station's ability to influence the surrounding environment after operation; the regional environment includes population size P. i / b i5 ;

[0135] The specific factors for each indicator are as follows:

[0136] ① Node Topology

[0137] Degree centrality of nodes (DC) i and the proximity centrality CC of nodes i Same as formula (9).

[0138] ②Station's own attributes

[0139] Site proximity L iThis metric is defined as the reciprocal of the average distance from a given site to other nodes connected to it. The larger the value, the closer the site is to other nodes.

[0140]

[0141] Among them, L ij Represents node v i With node v j Distance; k i Represents node v i The degree; This represents the average distance from a given node to other nodes that are connected to it.

[0142] Passenger flow D i This indicator is defined as the passenger flow entering a station and the total passenger flow transferring between stations, without adding outgoing passenger flow. It is calculated based on card swipe / payment records and reflects the importance of the transportation station, as shown in Table 7.

[0143] Table 7 Passenger Flow Index Assignment for Transportation Network Nodes

[0144]

[0145] ③ Regional environment

[0146] Population size P i This indicator is defined as the sum of the products of the area of ​​different grid cells within a node region and their corresponding population density. It focuses on reflecting the future development potential of transportation stations, and the calculation method is as follows:

[0147]

[0148] Among them, w r This represents the population density of grid r; This indicates that the raster at node x i The area occupied within the region.

[0149] ④ Calculate the overall importance of the transportation network:

[0150]

[0151] in:

[0152] Among them, LL i M represents the node topology metrics for a given site; i For a given station, the station's own attribute indicators and N; i For a specific site, regional environmental indicators and; and The weighting coefficients for the power grid, communication network, and transportation network are set as follows: the weighting coefficient for the indicator with the largest value is set to 50%, and the weighting coefficient for the other indicators is set to 25%.

[0153] Step 3: In response to the occurrence of power grid site failures, identify the faulty nodes in the three-network coupled network and the faulty nodes caused by cascading failures;

[0154] Specifically, based on the location of the faulty power grid site, its corresponding node in the three-network coupling network is determined; based on a preset load redistribution strategy, a cascading failure simulation of the coupling network is performed in the three-network coupling network to determine the faulty node caused by the cascading failure.

[0155] Specifically, the steps include the following:

[0156] When a node or edge in a network is subjected to a deliberate attack or experiences a random failure, the load on the node or edge will be transferred to the surrounding adjacent nodes or edges according to certain load distribution rules. When the load value on a node or edge exceeds its maximum load capacity after the load transfer, the node or edge will fail and be removed. Its load will then be further transferred and distributed according to the corresponding rules, thus causing the cascading propagation of the failure.

[0157] In current research on cascading failure models in power grids and communication networks, the most commonly used seepage theory and load-capacity models assume that faults have the same propagation characteristics on one side of the network, failing to consider the heterogeneity of fault propagation on one side and thus failing to capture the differences between power grids, communication networks, and transportation networks. For example, in power grids, islanding is a common strategy to prevent cascading failures, and critical loads are often connected to multiple substations to ensure power supply reliability. When a substation fails, backup lines are activated, and users obtain power from nearby substations, with the load of the failed substation being transferred and distributed. In power grids, nodes can only survive if their load is less than their capacity, while in communication networks, components only function if they belong to the most connected subgraph. Therefore, it is necessary to establish network cascading failure models with different failure mechanisms for subnetworks.

[0158] Previously, node load was often defined based on node betweenness, and the load was redistributed through the shortest path after a node failure. This method is reasonable for small networks, but its limitations arise as network size increases and the complexity of betweenness calculations grows. While defining node load and allocation strategies using node degree simplifies the model, relying solely on degree still presents practical constraints. Therefore, in this embodiment, the initial load L of nodes in the three networks is defined. i for:

[0159]

[0160] Wherein, the set of adjacent nodes of node i is V i K j Let represent the importance of the neighboring nodes of node i.

[0161] Define the load capacity of the node as C. i If the node load capacity is set to be proportional to the initial load, then the load capacity of node i is:

[0162] C i =(1+β)L i (15),

[0163] Here, β is the node's load tolerance coefficient, used to measure the node's ability to withstand the impact of cascading failures. This coefficient reflects both the numerical value of the node's load capacity and the cost required to build the corresponding network. Under the same fault scale, the network's ability to handle cascading failures increases with the increase of the tolerance coefficient β; that is, the larger the β value, the stronger the network's ability to resist external attacks. In artificial networks, the cost of building the network is a significant factor limiting node capacity.

[0164] Artificial networks specifically refer to network systems that are consciously designed, planned, and built by humans. Common types include computer networks (LANs and WANs), communication networks, and power grids.

[0165] The scheme described in this embodiment defines the redistribution of network node load according to a distribution strategy. When a node fails, its load will be redistributed to the nodes connected to it. For example... Figure 7 As shown, when node 10 fails, the load of node 10 will be distributed to nodes 6, 17, 20, 21, and 22, and the load increment received by its neighbor node j from the failed node i is ΔL. ji This is related to the proportion of node j's capacity in the total load capacity of the adjacent nodes of the failed node i. The relationship is as follows:

[0166]

[0167] Wherein, the load of node i at time t is L. i (t), after node j receives the redistributed load from node i, if the node's load at this time is greater than its own load capacity, i.e., L j (t)+ΔL ji >C j This causes node j to fail and be removed. Then, the load of node j will be distributed to its corresponding neighboring nodes according to the distribution rule of equation (16). The failure propagation ends when the load of all nodes in this side of the network is less than its capacity.

[0168] Step 4: For the identified faulty node, restore the faulty node of the power grid based on the preset cascaded failure recovery strategy; wherein, the cascaded failure recovery strategy is specifically: based on the node importance score of the faulty node, important nodes are restored first, and the restoration process is achieved by the coordinated use of generator sets, distributed power sources and mobile emergency power sources.

[0169] In specific implementation, the cascading failure and recovery process includes the following steps:

[0170] The joint monitoring and control of power grids, communication networks, and transportation networks has greatly enhanced the overall controllability and observability of the system. However, this convergence of the three networks also introduces more uncertainties, posing new challenges to the safe and stable operation of the system. The power grid, communication network, and transportation network are interdependent; when one network fails, the other two will also be affected, potentially expanding the scope of the failure. Faults and abnormal operations within the power grid, as well as congestion or accidents in the transportation network, can also negatively impact the communication network. As the complexity of power cyber-physical systems (CPPS) continues to increase, their ability to cope with sudden power flow fluctuations, generator disconnections, or traffic disruptions may decrease, making the system more vulnerable. In-depth research and optimization of joint control strategies for the three networks can enhance the system's ability to respond to various emergencies, reduce the occurrence of cascading failures, and ensure the continuous operation of critical infrastructure and the stable development of the social economy.

[0171] The cascading failure and recovery process scheme includes simulation of coupled network cascading failure process and cascading failure recovery strategy, as detailed below:

[0172] Coupled network cascade failure process: such as Figure 8 As shown, taking the nine nodes of the power grid and the seven nodes of the communication and transportation networks as an example, when node 3 in the power grid fails, its load is distributed to neighboring nodes, causing some neighboring nodes to exceed their capacity and fail. The load of these failed nodes is then distributed again. At the same time, nodes in the communication and transportation networks that depend on the failed nodes in the power grid also fail, and the load is transferred between different network nodes until the load of all nodes is less than their capacity, at which point the cascading fault ends.

[0173] Cascaded failure recovery strategy:

[0174] This strategy prioritizes the recovery of critical nodes, utilizes multiple power sources working in tandem, and verifies its effectiveness through comparative simulation experiments, as detailed below:

[0175] Priority restoration principle for critical nodes: When a power grid failure occurs, the importance of the failed nodes is assessed based on a pre-built three-network coordinated node importance evaluation system, and nodes with higher importance are restored first. This principle aims to ensure that critical nodes with a significant impact on power grid operation and the socio-economic situation can have their power supply restored as quickly as possible, reducing losses caused by power outages.

[0176] Multi-power source coordinated restoration: The restoration process is achieved through the coordinated use of generator sets, distributed power sources, and mobile emergency power sources. Rationally planning the coordinated allocation of these three power sources (traditional generator sets, distributed power sources, and mobile emergency power sources) during grid node restoration requires comprehensive consideration of factors such as power source characteristics, restoration phase priorities, the needs of each restoration phase, and economic objectives.

[0177] Specifically, the steps include the following:

[0178] Step 401, Node Priority Allocation Principle: Determine the recovery order of different types of power sources based on their characteristics and the overall importance score of the three networks;

[0179] Traditional generator sets are characterized by large capacity, high stability, and strong continuous power supply capability, but are limited by long start-up time (requiring several hours) and reliance on the transmission network; they can be used for grid recovery. Distributed power sources (photovoltaic / wind power / energy storage) have the advantages of fast response (minutes), wide geographical distribution, and support for islanded operation, but are limited by limited capacity and intermittent fluctuations; they can be used for local rapid recovery. Mobile emergency power sources (new energy electric vehicles / energy storage vehicles / diesel generators) have the advantages of rapid deployment (within 30 minutes) and flexibility, but are limited by fuel dependence and high operation and maintenance costs; they can be used for emergency power supply at critical nodes.

[0180] The overall importance score for the three networks can be determined according to step 2;

[0181] The specific recovery sequence includes:

[0182] ① Mobile emergency power supplies should be deployed to the nodes with the highest scores across the three networks (such as nodes 10 and 12) to quickly prevent cascading failures;

[0183] ② Distributed power sources provide nearby support to secondary critical nodes (such as load nodes in residential areas) to form local microgrids;

[0184] ③ Traditional generator sets are centrally restored to the main power grid, providing a stable base load for the overall power supply.

[0185] Step 402, Power Supply Time-Space Relay Principle: Combining the power supply's response speed and operating characteristics, construct a three-stage collaborative mechanism of "emergency-transition-stability";

[0186] The emergency phase (0-2 hours) includes:

[0187] Mobile emergency power supplies (deployed in city centers and suburban outskirts) can seize critical nodes with a response time of minutes; distributed power supplies can start island mode to maintain core loads such as communication base stations and traffic signals; and traditional units can be preheated simultaneously to shorten grid connection preparation time.

[0188] Transition phase (2-24 hours):

[0189] Distributed power sources switch to grid-connected mode to fill unrestored areas of the transmission corridor (as shown in formula C). i =(1+β)L i (Guiding capacity allocation) Traditional generating units are gradually loaded according to the node scoring order, and high-weight transmission channels are prioritized for reconstruction; mobile power sources are dynamically transferred to newly emerging fault points to form a mobile support system of "following repair".

[0190] Stable phase (after 24 hours):

[0191] Traditional generating units bear the base load, while distributed power sources participate in peak shaving (such as utilizing photovoltaic power during the daytime); mobile power sources are gradually phased out and switched to standby mode to cope with secondary failures.

[0192] Step 403, Multi-network collaborative flexibility principle: Integrate real-time data from the power grid, communication network, and transportation network, and achieve dynamic optimization through the linkage of data from these three networks;

[0193] Communication network: Real-time transmission of node status and load demand, triggering power dispatch commands (such as calling mobile power supplies to assist when the output of distributed power sources in a certain area is insufficient);

[0194] Transportation network: Optimize mobile power supply route planning (based on Dijkstra's algorithm to avoid congested road sections) to ensure timely arrival of emergency resources;

[0195] Power grid: Combined with the load redistribution model, see formula (19), the power output is dynamically adjusted to prevent overload from causing secondary failure.

[0196] Step 404, the principle of balancing economy and reliability: optimize the total life cycle cost;

[0197] For portable power banks, priority should be given to short-term protection of high-value loads (such as data centers) to avoid long-term high-cost operation;

[0198] For distributed power sources: power sharing within the region can be achieved through cluster control (such as wind-solar hybridization), reducing the no-load loss of traditional units;

[0199] Step 405, combining the principles of steps 401-404, establishes a closed-loop strategy of "dynamic assessment - hierarchical response - flexible adjustment" for the three types of power sources within the framework of three-network collaboration. This method not only significantly improves recovery efficiency (simulation shows that the average recovery time of key nodes is shortened to 2.5 hours), but also reduces the risk of cascading failures through multi-network data fusion, providing a systematic solution for enhancing the resilience of urban power grids.

[0200] Traditional generator sets serve as the primary power source, while distributed power sources (such as photovoltaic power plants and wind farms) provide supplementary power to faulty nodes during the recovery process, reducing losses caused by large-scale power outages. Mobile emergency power supplies are deployed on the outskirts of city centers and suburbs. When critical load nodes (such as node 10) fail, mobile emergency power supplies in urban and suburban areas can be mobilized to achieve flexible power supply and minimize the impact of the failure.

[0201] To verify the effectiveness of distributed power sources and mobile emergency power sources, a comparative simulation experiment was conducted by adding them step by step.

[0202] No recovery strategy scenario (control experiment): Assuming no measures are taken when a node fails, the simulation results show that the nodes affect each other, and eventually all nodes will fail, highlighting the necessity of a recovery strategy.

[0203] like Figure 9 As shown, restoration is achieved solely using traditional generator sets: without considering mobile emergency power supplies and distributed power supply access, power is restored only by traditional generator sets in the distribution network (such as units 2, 11, and 13). The distribution network is identified for outage status, non-isolated outage zones are delineated, and load nodes are divided into parallel power supply areas to form initial partitions. However, the restoration speed is slow, and failures at critical nodes can easily cause large-scale power outages.

[0204] like Figure 10 As shown, considering distributed power supply access: while providing initial zoned power supply, photovoltaic power stations and wind farms provide supplementary power to faulty nodes, which can provide timely power support to faulty nodes in the city in a relatively flexible manner, effectively reducing urban losses and accelerating power recovery.

[0205] like Figure 11 As shown, the integration of mobile emergency power and distributed power sources is considered simultaneously: mobile emergency power sources provide supplementary power on top of the distributed power sources. When critical load nodes fail, mobile emergency power sources from different areas can be mobilized, minimizing losses and significantly accelerating power restoration. Simulation results under different conditions fully validate the effectiveness of mobile emergency power and distributed power source integration, as well as the principle of prioritizing critical nodes, in grid restoration.

[0206] Furthermore, to simultaneously verify the effectiveness of the important node priority principle, we plotted two line graphs for each case. The verification shows that the line graph with the important node priority principle has a significantly better recovery rate than the one without it.

[0207] like Figure 12 As shown in the figure, a cascading failure scenario without a recovery strategy is simulated: As a contrast, we assume that no measures are taken when a node fails. As can be seen from the figure, when no measures are taken, the nodes affect each other, eventually causing all nodes to fail.

[0208] like Figure 13 As shown, a cascading failure scenario is simulated when only traditional generator sets in the distribution network are used for power restoration: without considering the access of mobile emergency power supplies and distributed power sources, only traditional generator sets 2, 11, and 13 in the distribution network are used for power restoration. The power outage status of the distribution network is identified, and non-isolated outage areas are identified. The load nodes in the non-isolated outage areas are then divided into parallel power supply areas to form initial partitions.

[0209] like Figure 14 As shown, a simulation of cascading failures is presented considering distributed power source integration. This simulation assumes distributed power source integration but does not consider the use of mobile emergency power supplies. Simultaneously with the initial zoned power supply, photovoltaic power plants and wind farms also provide supplementary power to the faulty nodes, reducing losses caused by large-scale power outages. As can be seen from the figure, in this scenario, power support can be provided to urban faulty nodes relatively flexibly and promptly, effectively reducing urban losses and accelerating power recovery.

[0210] like Figure 15 As shown, a simulation of cascading failure is considered when mobile emergency power and distributed power are simultaneously connected: While initial zoned power supply is provided, photovoltaic power stations and wind farms also provide supplementary power to the faulty nodes. At the same time, mobile emergency power can be used to provide power. When a critical load node, such as node 10, fails, in addition to urban mobile power supplies, suburban mobile power supplies can also be called upon to minimize the losses caused by the failure.

[0211] like Figure 16 As shown, the scheme described in this embodiment simulates the cascading failure situation after a city power outage.

[0212] Simulation results show that, without considering the integration of mobile emergency power supplies and distributed power sources, relying solely on traditional generator sets in the distribution network for power restoration can easily lead to large-scale power outages and slow recovery speeds after failures at critical nodes. However, when mobile emergency power supplies and distributed power sources are considered, they can provide timely power support to urban fault nodes, significantly accelerating power restoration.

[0213] In one or more embodiments, corresponding to the above-described method, this embodiment provides a three-network coordinated power grid restoration system based on complex network theory, including:

[0214] The data acquisition unit is used to acquire power grid station data, as well as corresponding transportation station data and communication station data within a preset area, and to construct a three-network coupling network of power grid, transportation network and communication network.

[0215] An importance scoring unit is used to obtain the node importance score for each node in a three-network coupled network using a pre-constructed three-network collaborative node importance evaluation system. Specifically, the three-network collaborative node importance evaluation system involves calculating the overall power grid importance of the corresponding power grid station, the overall transportation network importance of the corresponding transportation station, and the overall communication network importance of the corresponding communication station. The node importance score is obtained by weighted summation of these three overall importance scores.

[0216] The fault node determination unit is used to determine the fault nodes in the three-network coupled network and the fault nodes caused by cascading faults in response to the occurrence of power grid site faults.

[0217] The power grid restoration unit is used to restore the power grid fault nodes based on a preset cascaded failure restoration strategy for a given fault node. The cascaded failure restoration strategy specifically involves prioritizing the restoration of important nodes based on their node importance scores. The restoration process is achieved through the coordinated use of generator sets, distributed power sources, and mobile emergency power sources.

[0218] It should be noted that each module in this embodiment corresponds one-to-one with each step in Embodiment 1, and their specific implementation processes are the same, so they will not be repeated here.

[0219] In further embodiments, the following is also provided:

[0220] An electronic device includes a memory and a processor, as well as computer instructions stored in the memory and running on the processor. When executed by the processor, the computer instructions perform the methods described in the above embodiments. For brevity, further details are omitted here.

[0221] A computer-readable storage medium for storing computer instructions, which, when executed by a processor, perform the methods described in the above embodiments.

[0222] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A three-network coordinated power grid restoration method based on complex network theory, characterized in that, include: Acquire power grid site data, as well as corresponding transportation site data and communication site data within a preset area, and construct a three-network coupling network of power grid, transportation network and communication network; For nodes in a three-network coupled network, a pre-constructed three-network collaborative node importance evaluation system is used to obtain the node importance score for each node. Specifically, the three-network collaborative node importance evaluation system involves calculating the overall power grid importance of the corresponding power grid station, the overall transportation network importance of the corresponding transportation station, and the overall communication network importance of the corresponding communication station. The node importance score is obtained by weighted summation of the overall power grid importance, the overall transportation network importance, and the overall communication network importance. The comprehensive importance of the power grid, transportation network, and communication network are specifically expressed as follows: Overall importance of the power grid: ; Overall Importance of Communication Networks: ; Overall Importance of Transportation Network: Overall importance of nodes in the three networks: in, For network nodes The overall importance of the power grid at the site. The influence value of the influencing factors for each site. For node proximity centrality, For the degree centrality of nodes, To assess the overall importance of nodes in the three networks, , and These are the weighting coefficients for different transportation networks. , and These are the weighting coefficients for the power grid, communication network, and transportation network. For a given site, the node topology metrics and... For a given station, the station's own attribute indicators and, For a specific site, the regional environmental indicators and; In response to the occurrence of power grid site faults, identify the faulty nodes in the three-network coupled network and the faulty nodes caused by cascading faults; For a identified fault node, the power grid fault node is restored based on a preset cascaded failure recovery strategy; wherein, the cascaded failure recovery strategy specifically means: based on the node importance score of the fault node, important nodes are restored first, and the restoration process is achieved in coordination by generator sets, distributed power sources and mobile emergency power sources. The node priority classification method for different types of power sources is as follows: mobile emergency power sources are prioritized for deployment to the nodes with the highest scores in the three networks; distributed power sources provide support to the next most important nodes to form local microgrids; and traditional generator sets are centrally used to restore the backbone network.

2. The three-network coordinated power grid restoration method based on complex network theory as described in claim 1, characterized in that, The calculation of the overall importance of the power grid station corresponding to the node, the overall importance of the transportation network station corresponding to the node, and the overall importance of the communication network station corresponding to the node are specifically as follows: the overall importance of the power grid station is determined based on the station level, station size, load level, and load size of the power grid station; the overall importance of the communication network station is determined based on the degree centrality, proximity centrality, and betweenness centrality of the communication network station; and the overall importance of the transportation network station is determined based on the degree centrality, proximity centrality, station proximity, passenger flow, and population size of the transportation network station.

3. The three-network coordinated power grid restoration method based on complex network theory as described in claim 1, characterized in that, In response to the occurrence of a power grid site fault, the fault node in the three-network coupled network and the fault node caused by cascading faults are determined, specifically: based on the location of the power grid site where the fault occurred, its corresponding node in the three-network coupled network is determined. Based on a preset load redistribution strategy, a cascading failure simulation of the coupled network is performed in the three-network coupled network to identify the faulty node caused by the cascading failure.

4. The three-network coordinated power grid restoration method based on complex network theory as described in claim 3, characterized in that, The load redistribution strategy specifically involves: constructing the initial load of each node based on the node importance score of each node in the three-network coupled network; and determining the load capacity of each node based on the initial load. For a faulty node, its load is transferred to a node adjacent to the current faulty node according to the preset load distribution rules, and the current faulty node and its corresponding edge are removed.

5. The three-network coordinated power grid restoration method based on complex network theory as described in claim 1, characterized in that, The process of coordinated restoration using generator sets, distributed power sources, and mobile emergency power sources is as follows: Node priority allocation principle: The priority of different types of power sources is determined based on the characteristics of different power sources and the comprehensive importance score of the three networks; By combining the power supply's response speed and operating characteristics, a three-stage coordinated mechanism of "emergency-transition-stability" is constructed. Integrate real-time data from the power grid, communication network, and transportation network, and determine power dispatching schemes based on the linkage of data from these three networks; Optimize the total lifecycle cost for different power supplies; The above four strategies are combined to form a closed-loop strategy of "dynamic assessment - tiered response - flexible adjustment" for coordinated recovery.

6. A three-network coordinated power grid restoration system based on complex network theory, characterized in that, include: The data acquisition unit is used to acquire power grid station data, as well as corresponding transportation station data and communication station data within a preset area, and to construct a three-network coupling network of power grid, transportation network and communication network. An importance scoring unit is used to obtain the node importance score for each node in a three-network coupled network using a pre-constructed three-network collaborative node importance evaluation system. Specifically, the three-network collaborative node importance evaluation system involves calculating the overall power grid importance of the corresponding power grid station, the overall transportation network importance of the corresponding transportation station, and the overall communication network importance of the corresponding communication station. The node importance score is obtained by weighted summation of these three overall importance scores. The comprehensive importance of the power grid, transportation network, and communication network are specifically expressed as follows: Overall importance of the power grid: ; Overall Importance of Communication Networks: ; Overall Importance of Transportation Network: Overall importance of nodes in the three networks: in, For network nodes The overall importance of the power grid at the site. The influence value of the influencing factors for each site. For node proximity centrality, For the degree centrality of nodes, To assess the overall importance of nodes in the three networks, , and These are the weighting coefficients for different transportation networks. , and These are the weighting coefficients for the power grid, communication network, and transportation network. For a given site, the node topology metrics and... For a given station, the station's own attribute indicators and, For a specific site, regional environmental indicators and; The fault node determination unit is used to determine the fault nodes in the three-network coupled network and the fault nodes caused by cascading faults in response to the occurrence of power grid site faults. The power grid restoration unit is used to restore the power grid fault nodes based on a preset cascaded failure restoration strategy for a given fault node. The cascaded failure restoration strategy specifically involves prioritizing the restoration of important nodes based on their node importance scores. The restoration process is achieved through the coordinated use of generator sets, distributed power sources, and mobile emergency power sources. The node priority classification method for different types of power sources is as follows: mobile emergency power sources are prioritized for deployment to the nodes with the highest scores in the three networks; distributed power sources provide support to the next most important nodes to form local microgrids; and traditional generator sets are centrally used to restore the backbone network.

7. An electronic device, comprising a memory, a processor, and a computer program stored in the memory and running thereon, characterized in that, When the processor executes the program, it implements the three-network coordinated power grid restoration method based on complex network theory as described in any one of claims 1-5.

8. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by the processor, the program implements the three-network coordinated power grid restoration method based on complex network theory as described in any one of claims 1-5.

Images