Distributed intelligent power grid post-disaster flexible recovery regulation and control method and system

Abstract

The invention discloses a post-disaster flexible recovery regulation and control method and system for a distributed smart power grid, and the method comprises the following steps: building an emergency power supply region division model in consideration of the damaged state and load demands of the post-disaster distributed smart power grid; starting a post-disaster emergency power supply area division model by using an improved sparrow search algorithm for a post-disaster recovery scene of the distributed smart grid; constructing a flexible resource participated distributed smart grid post-disaster recovery model; based on a distributed smart grid post-disaster recovery model, a network tuning model oriented to a feasible topology set is constructed, and a necessary basis is established for subsequently screening out an optimal topology meeting a post-disaster recovery target; the method comprises the following steps: grouping pheromone groups in a network tuning model by adopting a synchronization driving principle, introducing a dynamic perception self-tuning mechanism, and obtaining an optimal topology result according to an iteration progress and quality self-adaptive adjustment of a solution of an ant colony optimal objective function, thereby providing a strategy for a dispatcher to implement post-disaster recovery regulation.

Description

Technical Field

[0001] This invention relates to the field of distributed smart grid post-disaster technology, and in particular to a method and system for flexible recovery and control of distributed smart grids after a disaster. Background Technology

[0002] In recent years, the trend of global climate change has become increasingly pronounced, with the frequency and intensity of extreme weather events continuing to rise. As the "last link" connecting end users in the power supply network, the post-disaster functional recovery capability of distributed smart grids is directly related to socio-economic development and people's livelihoods. Extreme disasters not only directly cause the failure of power facilities but may also lead to a complete power outage in a region, affecting the reliability of power supply and the operation of industry and commerce. Against this backdrop, establishing the core capability of distribution systems to resist interference and restore pre-disaster functions has become one of the core objectives of the construction of new power systems.

[0003] An existing method for islanded fault recovery and reconfiguration in distribution networks (CN202411690083.1) has shortcomings in the utilization of distributed resources. While it divides islanded power supply areas and utilizes distributed power sources, and maintains stable island operation through a simple genetic algorithm, it fails to explore the flexible adjustment potential of distributed resources. It treats distributed power sources as fixed power sources and does not consider load demand changes during fault recovery or the progress of network reconnection in designing dynamic output adjustment strategies for distributed power sources. Furthermore, it does not dynamically adjust power supply priorities according to load importance or consider the flexible response characteristics of controllable loads. Therefore, the key issues to address are: how to explore the flexible adjustment potential of distributed resources, design dynamic output strategies based on fault recovery conditions, supplement compensation mechanisms for different power source fluctuations, and establish a load flexible adjustment mechanism; how to adjust power supply priorities according to load importance and incorporate controllable load response characteristics to achieve source-load coordinated adaptation; and how to improve the efficiency and stability of distributed smart grid fault recovery in complex fault scenarios. Summary of the Invention

[0004] To address the aforementioned technical problems, the present invention aims to provide a distributed smart grid post-disaster flexible recovery and control method that comprehensively considers the flexible recovery and control capabilities of distributed resources and loads, accelerates the load recovery process during the post-disaster load emergency recovery phase and the resource flexible recovery and control phase, reduces the overall economic cost, and improves voltage quality.

[0005] The objective of this invention is achieved by at least one of the following technical solutions.

[0006] A method for flexible recovery and control of distributed smart grids after disasters includes the following steps: S1. Considering the post-disaster damage status and load demand of the distributed smart grid, establish an emergency power supply area division model; S2. For the disaster recovery scenario of distributed smart grids, the sparrow search algorithm is improved to provide more efficient optimization calculation support for disaster emergency power supply strategies; S3. Use the improved sparrow search algorithm to start the post-disaster emergency power supply area division model to prioritize power supply to key loads in the power protection area; S4. Construct a distributed smart grid post-disaster recovery model with flexible resource participation to optimize load power supply recovery; S5. Based on the distributed smart grid post-disaster recovery model, construct a network optimization model for feasible topology sets, which lays the necessary foundation for the subsequent selection of the optimal topology that meets the post-disaster recovery objectives. S6. The pheromone population in the network optimization model is grouped using the synchronization and parallel driving principle to improve the optimization ability. S7. A dynamic sensing and self-adjustment mechanism is introduced, which adaptively adjusts the topology based on the iteration progress and the quality of the solution to the ant colony's optimal objective function to obtain the optimal topology result, providing a strategy for schedulers to implement post-disaster recovery and control.

[0007] Furthermore, in step S1, considering the post-disaster damage status and load demand of the distributed smart grid, an emergency power supply area division model is established, specifically including: S1.1 Design Individual Coding Rules: Map the emergency power supply area division scheme into discrete individuals, including the status of multiple switches and the affiliation of multiple nodes; S1.2 Build a system with the goal of maximizing the total recoverable load value in the emergency power supply area, including the importance of nodes, the number of load nodes in the emergency power supply area, and the duration of continuous power supply from the emergency power supply area to the nodes. S1.3, the constraints corresponding to the post-disaster emergency power supply area division model are: I. Power Balance Constraints For each emergency power supply area, the sum of the active power of all distributed power sources in the area minus the sum of the active load of all load nodes in the area, and then minus the power transmission line losses in the area, the result must be greater than the safety margin required to maintain the stability of the area. II. Node voltage constraints: The node voltage must not exceed the specified upper and lower limits; III. Connectivity Constraints Depth-first search (DFS) was used to verify that all nodes within the emergency power supply area form a single connected graph through closed switches, meaning that starting from any DG node, all nodes can be traversed and belong to the nodes included in the island.

[0008] Furthermore, in step S2, the improved sparrow search algorithm includes improving the SSA initialization strategy using the theory of optimal point sets: Using optimal point set theory, a more uniform coverage than pseudo-random numbers can be achieved within a continuous unit hypercube. When mapping continuous points that meet the conditions to a discrete permutation space, the exploration of neighboring solutions is added; In the original follower update formula, when the number of sparrows is greater than half the size of the sparrow population, updates are made only based on the worst position.

[0009] Furthermore, in step S3, the improved sparrow search algorithm is used to initiate the division of post-disaster emergency power supply areas, prioritizing power supply to key loads within the power protection area, specifically including: S3.1 Initialize the parameters of the improved Sparrow Search Algorithm SSA; S3.2, Generate the initial discrete population; S3.3. Use the improved SSA algorithm to partition islands.

[0010] Furthermore, in step S4, the construction of the distributed smart grid post-disaster recovery model includes: first obtaining all feasible network topology sets based on random generation trees, and then solving each feasible topology set using the Lorentz cone optimization method.

[0011] Furthermore, in step S5, the network optimization model for feasible topology sets incorporates load recovery and voltage quality improvement as indicators into the objective function, with the same constraints as those in the distributed smart grid power flow calculation of the distributed smart grid network flexible regulation model.

[0012] Furthermore, in step S6, the pheromone population in the network tuning model is grouped using the synchronization and parallel driving principle to improve the optimization speed of the network tuning model, including: S6.1 Each group follows a single-group optimization strategy, so that individuals visit the next node in the current node according to the pseudo-random principle. At this time, the variable of path transition probability is determined by taking the maximum value of the product of the power of the individual's path pheromone and the power of the heuristic function. S6.2. After the group completes the optimal path search, update the pheromones; S6.3 As the number of iterations increases, the optimal group becomes the elite group, and the pheromones of the elite group are updated into the path pheromones of the other groups.

[0013] Furthermore, step S7 specifically includes: S7.1 Introducing a dynamic sensing self-adjustment mechanism, which adaptively adjusts the algorithm based on the iteration progress and the quality of the solution to the ant colony's optimal objective function; S7.2. In parallel groups, inter-group pheromone exchange rules are introduced, and the pheromone sharing concentration is dynamically adjusted according to the degree of difference between groups.

[0014] The system for implementing the aforementioned distributed smart grid post-disaster flexible recovery and control method includes: The emergency power supply area modeling module is used to analyze the damage status and load demand of the distributed smart grid after a disaster using the emergency power supply area division model. The emergency power supply module is used to initiate the division of post-disaster emergency power supply areas using an improved sparrow search algorithm, and to prioritize power supply to key loads within the power protection area. The flexible resource adjustment module is used to obtain the set of network topologies with the least power outage load based on the random spanning tree as the feasible solution set, and solves each feasible topology set by the Lorentz cone optimization method; Network structure adjustment module: used to build a network optimization model for feasible topology sets, laying the necessary foundation for subsequent selection of the optimal topology that meets the post-disaster recovery goals; Group-driven optimization module: This module is used to group the pheromone population in the network optimization model according to the principle of synchronization and parallel driving, thereby improving the optimization capability. Dynamic optimization module: This module introduces a dynamic sensing and self-adjustment mechanism, which adaptively adjusts the topology based on the iteration progress and solution quality to obtain the optimal topology result, providing strategies for schedulers to implement post-disaster recovery and control.

[0015] The present invention provides a computer device and its software medium, comprising: a memory and a processor, and a computer program stored in the memory, wherein when the computer program is executed on the processor, the method is implemented.

[0016] Compared with the prior art, the beneficial effects of the present invention are as follows: This invention proposes a flexible recovery and control method for distributed smart grids after disasters. Considering the power supply support capacity of distributed power sources, it constructs emergency power supply zones prioritizing primary loads to achieve rapid power restoration for local critical loads. By combining the flexibility of the distributed smart grid infrastructure, distributed resources, and the flexible adjustment potential of the user side, it coordinates the output of distributed power sources with load demand, balancing grid-side voltage quality improvement with user-side power outage losses. This invention provides a technical path for the post-disaster recovery of distributed smart grids that balances recovery efficiency, operational stability, and socio-economic benefits. It fully mobilizes the collaborative capabilities of both power sources and loads, which is conducive to improving the resilience of distributed smart grids under extreme disasters, while simultaneously ensuring grid security and user interests. Attached Figure Description

[0017] Figure 1 This is a schematic diagram illustrating the steps of a distributed smart grid post-disaster flexible recovery and control method in an embodiment of the present invention.

[0018] Figure 2 This is a diagram illustrating the emergency power supply area division scheme in an embodiment of the present invention.

[0019] Figure 3 This is a diagram of a radial scheme that maximizes power restoration after flexible adjustment of the network structure, as shown in this embodiment of the invention. Detailed Implementation

[0020] The specific embodiments of the present invention will be further described below with reference to the accompanying drawings and examples. It is obvious that the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments of the present invention obtained by those skilled in the art based on the embodiments of the present invention without inventive effort, and all other embodiments obtained by those skilled in the art without inventive effort, are within the scope of protection of the present invention.

[0021] Example 1 A method for flexible recovery and control of distributed smart grids after disasters, such as Figure 1 As shown, it includes the following steps: S1. Establish an emergency power supply area division model to analyze the post-disaster damage status and load demand of the distributed smart grid, such as... Figure 2 As shown, it specifically includes: S1.1 Design individual coding rules: Map the island partitioning scheme to discrete individuals =[x i1 ,x i2 , ...,x ik ,...,x iK ,y i1 ,y i2 ,..., y in ,...,y iN ], where x ik ∈{0,1} represents the first... k The state of a switch (1 = closed, 0 = open). y in ∈{0,1} represents the first... n The affiliation of each node (1 = included in the island, 0 = excluded, and DG nodes default to y). id =1); K For the number of switches, N This represents the number of nodes.

[0022] S1.2 The objective function formula is: Maximizing the total recoverable load value in the emergency power supply area.

[0023] in: μi For nodes i Whether it is defined as a node in the emergency power supply area; ω i For nodes i The degree of importance; N is The number of load nodes within the emergency power supply area; T i,sup For emergency power supply areas to nodes i The duration of continuous power supply; F ( X i () represents the total load restored within the emergency power supply area.

[0024] S1.3, the constraints corresponding to the post-disaster emergency power supply area division model are: I. Power Balance Constraints

[0025] P DGi Represents the regional nodes for emergency power supply in distributed smart grids i The active power of the connected distributed generation (DG); P Li Indicates emergency power supply area ν internal nodes i Active load; in emergency power supply area ν Internally, line losses during power transmission are used P ν loss express; ε Emergency power supply area ν The safety margin required to maintain stability; G ν Emergency power supply area ν The sum of internal DG nodes; J ν It is an emergency power supply area ν The set of internal load nodes; the total number of independent emergency power supply areas in the entire system is n ν .

[0026] II. Node Voltage Constraints The node voltage does not exceed the specified upper and lower voltage limits; III. Connectivity Constraints Depth-First Search (DFS) was used to verify that all nodes within the emergency power supply area form a single connected graph through closed switches, meaning that all nodes can be traversed starting from any DG node.y in Nodes with a value of 1; S2. For disaster recovery scenarios of distributed smart grids, an improved sparrow search algorithm is used to provide more efficient optimization computational support for disaster emergency power supply strategies, specifically including: S2.1 To avoid generating many invalid solutions at the initial stage of the algorithm in a distributed smart grid, which would reduce the subsequent search speed, the optimal point set theory is used to improve the SSA initialization strategy, including: Using optimal point set theory, a more uniform coverage than pseudo-random numbers can be achieved within a continuous unit hypercube:

[0027] When the deviation of the point set φ ( e )satisfy φ ( e )≤ C ( r, ) e -1+ hour, P e ( k g It is called GPS. r For best results, e Indicates the number of points. The dimension is s and points e The function composed of; { } represents taking the decimal part; C ( r, ) is a constant; Let represent any positive number. When mapping these continuous points to a discrete permutation space, incorporating the exploration of neighboring solutions can prevent the process from losing some uniformity.

[0028] S2.2, In the original follower update formula, when i sp > When the value is 2, updating only based on the worst-case position can incorporate consideration of the globally optimal position, improving the ability to cluster towards high-quality regions. The improved formula combines the worst-case and globally optimal positions. .

[0029]

[0030] in: Indicates the size of a sparrow population; For the first i sp Only sparrowst The current position is the worst possible position. This is the current optimal position; The globally optimal position; A Let be a random matrix with elements in {-1, 1}. A + = A T ( AA T ) -1 ; L mat It is a matrix whose values ​​are all 1; H These are random numbers that follow a normal distribution. β ∈(0,1] is the weighting coefficient, used to balance the influence of the worst position and the global best position.

[0031] S3. Utilize the improved sparrow search algorithm to delineate post-disaster emergency power supply areas, prioritizing power supply to key loads within the power protection areas. Specifically, this includes: S3.1 Initialize the parameters of the improved sparrow search algorithm SSA. As an example, set the population size. M =50, Maximum number of iterations T =100, step size factor in the range (0,1] (take 0.5), safety threshold ST ∈[0.5,1] (take 0.8), precision threshold χ =10 -5 Random numbers r 1, r 2, r 3, r 4, r 5∈[0,1].

[0032] S3.2, Generate the initial discrete population, including: Randomly assign values ​​to each individual X i Switch status x ik Node affiliation y in And DG node y id =1; Verify the topological connectivity of individuals using DFS. If they are not connected, adjust the state of the disconnected critical switch to 1 until they are connected. Calculate the power balance of individual nodes. If the node power and line network loss power exceed the power of the DG, then proceed according to... ω s Descending order will prioritize low-importance load nodes. y in Set to 0 until the power balance constraint is met; Repeat the above steps to generate the output. M The initial population consists of 10 feasible discrete individuals. P (0), and calculate the objective function value for each individual. P ( X i ).

[0033] S3.3. Applying the improved SSA algorithm for island partitioning includes: Set the iteration counter t =0, recording the best individual in the initial population. With the worst individual ; Select individual roles in the population proportionally: select 10%-20% of individuals as discoverers, 70%-80% as followers, and randomly select 10%-20% as vigilants; Perform a location update operation using the following formula: I. Discoverer Update:

[0034] in: For the first i sp Only sparrows t The position of the generation; γ ∈(0,1] is a random number; T max The maximum value representing the number of iterations; ∈[0,1] represents the warning value; ST sp Indicates the safety threshold; H Represents a random number that follows a normal distribution; L mat It is a matrix whose values ​​are all 1; II. Follower Update: As shown in step S2.2; III. Vigilant Update:

[0035] in: This is currently the best location; These are random numbers that follow a normal distribution. M ran A random number in the range [-1, 1]. f isp For the first i The fitness value of a sparrow ； It is a minimal constant that guarantees the denominator is not zero; This is currently the worst position; This represents the worst fitness value globally; This represents the global optimal fitness value.

[0036] Update the optimal individual in the population With the worst individual ,make t = t +1; like t ≥ T Or for 10 consecutive generations F ( If there is no improvement, terminate the iteration; otherwise, return to the update operation. After the iteration terminates, output the final optimal individual. Corresponding switch state x ik Node Attribution y in and the objective function value F ( ), Total load restored ∑ P s,rec Continuous power supply time T s,sup Important load recovery rate.

[0037] S4. Construct a distributed smart grid post-disaster recovery model for optimizing load power supply restoration under the electricity market environment, including the following steps: S4.1 Introduce the spanning tree model and use the radial constraint mechanism to generate the radial topology of the distributed smart grid and obtain all feasible network topology sets; S4.2 Solve for each feasible topology set using the Lorentz cone optimization method, including the following steps: S4.21. Utilizing the flexible regulation capabilities of distributed generation (DG) and adjustable loads, an objective function is established with the goals of minimizing various costs and maximizing voltage stability of the distributed smart grid, namely:

[0038]

[0039]

[0040] in: For the network loss of each branch road, L For branch set; To reduce the cost of purchasing electricity from higher authorities, For electricity sales revenue, N m A set of nodes; V i Represents a node i voltage, V refThe node reference voltage, As an economic weight, As the voltage deviation weight, f ( x Various costs associated with distributed smart grids g ( x () represents voltage stability; S4.22, Propose constraints for power flow calculation in distributed smart grids:

[0041]

[0042]

[0043]

[0044] In the formula: B ij For nodes i , j susceptance; G ij For nodes i and j The electrical conductivity; For nodes i Active power; For nodes i Reactive power; δ ij For nodes i , j Voltage phase angle difference; node i During the period t The voltage value is used U i express; and They are nodes i Maximum and minimum voltage values; N z This represents the total number of nodes in the distributed smart grid. U i Represents a node i The voltage; and Representing branch roads l The active power and the maximum value of the active power; g To restructure the power distribution network; G To account for the weighted sum of power loss load and node voltage; S4.23. Resource flexibility adjustment and optimization through the Lorentz cone, including energy storage resource adjustment, distributed resource adjustment, and demand-side response: I. Distributed resource constraints

[0045] in: B DG For DG node set, For DG at the node j Maximum output DG node j of effort.

[0046] II. Energy Storage Resource Constraints Energy storage resources must meet the following requirements: the sum of charging and discharging state variables in the same period does not exceed 1; the discharging power is between the upper and lower limits of the corresponding state, and the charging power is the same; the energy of the next period is calculated by combining the energy of the current period with the charging and discharging power and efficiency coefficient; the energy is within the upper and lower limits considering factors such as lifespan, and the constraints apply to all nodes containing energy storage. III. Demand-side response constraints

[0047] Where: Δ L t This is the load adjustment amount; This is the price elasticity of demand coefficient; L base Basic load; p t and p base These are the real-time electricity price and the benchmark electricity price, respectively. S4.24. Based on the above objective function and constraints, a distributed smart grid disaster recovery model based on Lorentz cone relaxation can be obtained to optimize the disaster recovery situation.

[0048] S5. To enhance the post-disaster self-healing capabilities of distributed smart grids, and considering adjustable resources in the electricity market environment, a network optimization model for feasible topology sets is constructed. This lays the necessary foundation for subsequently selecting the optimal topology that meets the post-disaster recovery objectives. Specifically, this includes: S5.1 To avoid excessively low node voltage due to line overload during power transfer, load restoration, switching quantities, and network losses are incorporated as indicators into the objective function, which can be expressed as:

[0049] in, G To account for the weighted sum of power loss load and node voltage; y i To restore the binary variable to the load, set it to 1 or 0; P M,i For nodes i Active power at the point; Δ t Indicates cumulative time; and These are the weighting coefficients; m , n Indicates the route; t Indicates time; Ω The set of all branches of a distributed smart grid; decision variables for t Time-of-day routes m - n Switch status, for t- 1-hour route m - n Switch status; 0 indicates circuit m - n When the switch is open, take 1 to represent the circuit. m - n The switch is closed; Indicates time t branch road m - n Current value; branch road m - n The resistance value; S5.2 The constraints of the network optimization model are the same as those in step S4.22 for distributed smart grid power flow calculation constraints.

[0050] S6. To improve the optimization speed of the network optimization model, the pheromone population of the network optimization model is grouped using the synchronization and parallel driving principle to enhance the optimization capability. Specifically, this includes: S6.1, Each group follows a single-group optimization strategy to ensure that the first group... Individuals follow pseudo-random principles at nodes i Visit the next node j The formula is as follows:

[0051] in: It is the first w Path pheromones of individuals within a group; It is a heuristic function. Q It is a variable representing the path transition probability; α For pheromone factors; β Heuristic function factor; S6.2 When the group completes the optimal path search, the pheromone rules are updated as follows:

[0052]

[0053] in: Pheromones loss rate; It is a very small positive number; S For pheromone weights; L w For the first wThe optimal path length for the group; R wopt For the first w The optimal path for the group; S6.3 As the number of iterations increases, the optimal group, as the elite group, updates its pheromones to the path pheromones of the other groups, according to the following update rules:

[0054]

[0055] in: To communicate pheromone loss rate; L wopt It is the optimal path length among all groups; Indicates the first g Group of ants on the path ( i , j The increase in pheromones on the surface.

[0056] S7. To optimize adaptability, a dynamic sensing and self-adjusting mechanism is introduced. This mechanism adaptively adjusts based on the iteration progress and the quality of the solution to the ant colony's optimal objective function, yielding the optimal topology result. This provides a strategy for schedulers to implement post-disaster recovery and control measures, specifically including: S7.1 Introduces a dynamic sensing self-adjustment mechanism, which adaptively adjusts the algorithm based on the iteration progress and the quality of the solution to the ant colony's optimal objective function:

[0057]

[0058] in: S For pheromone weights, For pheromone loss rate, The attenuation coefficient; For iteration progress; This is the incentive coefficient; It is a local minimum; S init The initial pheromone weights; F f It is the fitness of the current solution; F l It is the fitness of the previous solution; k it This represents the current iteration number; K it This represents the maximum number of iterations. S7.2. In a co-current group, inter-group pheromone exchange rules are introduced, which can dynamically adjust the pheromone sharing concentration based on the degree of difference between groups:

[0059] in: It refers to the concentration of pheromones exchanged within the same group. The pheromone diversity within the same group represents the standard deviation of the optimal value for each group. S7.3, Select the radial network with the largest power restoration amount, such as [[ID= thirty]]Figure 3 As shown.

[0060] Example 2 The system for implementing the aforementioned distributed smart grid post-disaster flexible recovery and control method includes: Emergency Power Supply Area Modeling Module: This module includes an emergency power supply area division model, used to analyze the damage status and load demand of the distributed smart grid after a disaster.

[0061] Algorithm improvement module: Includes an improved sparrow search algorithm for disaster recovery scenarios of distributed smart grids; Emergency power supply module: Utilizes an improved sparrow search algorithm to initiate the division of post-disaster emergency power supply areas, prioritizing power supply to key loads within the power protection zone; Flexible resource regulation module: Constructs a disaster recovery model for distributed smart grids with flexible resource participation to optimize load power supply recovery under the power market environment. It obtains all feasible network topologies based on random generation tree and solves each feasible topology using the Lorentz cone optimization method. Network structure adjustment module: To enhance the self-healing capability of distributed smart grids after disasters, and considering the adjustable resources in the power market environment, a network optimization model for feasible topology sets is constructed, laying the necessary foundation for subsequent selection of the optimal topology that meets the post-disaster recovery goals; Group co-drive optimization module: To improve the optimization rate of the algorithm in the network optimization model, the pheromone group in the algorithm is grouped according to the synchronization and co-drive principle, thereby improving the optimization capability. Dynamic optimization module: Introduces a dynamic sensing and self-adjustment mechanism, which adaptively adjusts the topology based on the iteration progress and the quality of the solution to the ant colony's optimal objective function to obtain the optimal topology result, providing strategies for schedulers to implement post-disaster recovery and control.

[0062] Example 3 A computer device and its software medium include a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps of a distributed smart grid post-disaster flexible recovery and control method according to an embodiment.

[0063] The above embodiments are preferred embodiments of the present invention, but the embodiments of the present invention are not limited to the above embodiments. Any modifications, alterations, substitutions, combinations, or simplifications made without departing from the spirit and principle of the present invention shall be considered equivalent substitutions and shall be included within the protection scope of the present invention.

Claims

1. A method for flexible recovery and control of distributed smart grids after disasters, characterized in that, Includes the following steps: S1. Considering the post-disaster damage status and load demand of the distributed smart grid, establish an emergency power supply area division model; S2. For the disaster recovery scenario of distributed smart grids, the sparrow search algorithm is improved to provide more efficient optimization calculation support for disaster emergency power supply strategies; S3. Use the improved sparrow search algorithm to start the post-disaster emergency power supply area division model to prioritize power supply to key loads in the power protection area; S4. Construct a distributed smart grid post-disaster recovery model with flexible resource participation to optimize load power supply recovery; S5. Based on the distributed smart grid post-disaster recovery model, construct a network optimization model for feasible topology sets, which lays the necessary foundation for the subsequent selection of the optimal topology that meets the post-disaster recovery objectives. S6. The pheromone population in the network optimization model is grouped using the synchronization and parallel driving principle to improve the optimization ability. S7. A dynamic sensing and self-adjustment mechanism is introduced, which adaptively adjusts the topology based on the iteration progress and the quality of the solution to the ant colony's optimal objective function to obtain the optimal topology result, providing a strategy for schedulers to implement post-disaster recovery and control.

2. The distributed smart grid post-disaster flexible recovery and control method according to claim 1, characterized in that, In step S1, considering the post-disaster damage status and load demand of the distributed smart grid, an emergency power supply area division model is established, specifically including: S1.1 Design Individual Coding Rules: Map the emergency power supply area division scheme into discrete individuals, including the status of multiple switches and the affiliation of multiple nodes; S1.2 Build a system with the goal of maximizing the total recoverable load value in the emergency power supply area, including the importance of nodes, the number of load nodes in the emergency power supply area, and the duration of continuous power supply from the emergency power supply area to the nodes. S1.3, the constraints corresponding to the post-disaster emergency power supply area division model are: I. Power Balance Constraints For each emergency power supply area, the sum of the active power of all distributed power sources in the area minus the sum of the active load of all load nodes in the area, and then minus the power transmission line losses in the area, the result must be greater than the safety margin required to maintain the stability of the area. II. Node voltage constraints: The node voltage must not exceed the specified upper and lower limits; III. Connectivity Constraints Depth-first search (DFS) was used to verify that all nodes within the emergency power supply area form a single connected graph through closed switches, meaning that starting from any DG node, all nodes can be traversed and belong to the nodes included in the island.

3. The distributed smart grid post-disaster flexible recovery and control method according to claim 1, characterized in that, In step S2, the improved sparrow search algorithm includes improving the SSA initialization strategy by applying the theory of optimal point sets: Using optimal point set theory, a more uniform coverage than pseudo-random numbers can be achieved within a continuous unit hypercube. When mapping continuous points that meet the conditions to a discrete permutation space, the exploration of neighboring solutions is added; In the original follower update formula, when the number of sparrows is greater than half the size of the sparrow population, updates are made only based on the worst position.

4. The distributed smart grid post-disaster flexible recovery and control method according to claim 1, characterized in that, In step S3, the improved sparrow search algorithm is used to initiate the division of post-disaster emergency power supply areas, prioritizing power supply to key loads within the power protection area. Specifically, this includes: S3.1 Initialize the parameters of the improved Sparrow Search Algorithm SSA; S3.2, Generate the initial discrete population; S3.

3. Use the improved SSA algorithm to partition islands.

5. The distributed smart grid post-disaster flexible recovery and control method according to claim 1, characterized in that, In step S4, the construction of the distributed smart grid post-disaster recovery model includes: firstly, obtaining all feasible network topology sets based on random generation trees, and then solving each feasible topology set using the Lorentz cone optimization method.

6. The distributed smart grid post-disaster flexible recovery and control method according to claim 1, characterized in that, In step S5, the network optimization model for feasible topology sets incorporates load recovery and voltage quality improvement as indicators into the objective function, with the same constraints as those in the distributed smart grid power flow calculation of the distributed smart grid network flexible regulation model.

7. The distributed smart grid post-disaster flexible recovery and control method according to claim 1, characterized in that, In step S6, the pheromone population in the network tuning model is grouped using the synchronization and parallel driving principle to improve the optimization speed of the network tuning model, including: S6.1 Each group follows a single-group optimization strategy, which allows individuals to visit the next node in the current node according to the pseudo-random principle. At this time, the variable of path transition probability is determined by taking the maximum value of the product of the power of the individual's path pheromone and the power of the heuristic function. S6.

2. After the group completes the optimal path search, update the pheromones; S6.3 As the number of iterations increases, the optimal group becomes the elite group, and the pheromones of the elite group are updated into the path pheromones of the other groups.

8. The distributed smart grid post-disaster flexible recovery and control method according to claim 1, characterized in that, Step S7 specifically includes: S7.1 Introducing a dynamic sensing self-adjustment mechanism, which adaptively adjusts the algorithm based on the iteration progress and the quality of the solution to the ant colony's optimal objective function; S7.

2. In parallel groups, inter-group pheromone exchange rules are introduced, and the pheromone sharing concentration is dynamically adjusted according to the degree of difference between groups.

9. A system for implementing the distributed smart grid post-disaster flexible recovery and control method as described in claim 1, characterized in that, include: The emergency power supply area modeling module is used to analyze the damage status and load demand of the distributed smart grid after a disaster using the emergency power supply area division model. The emergency power supply module is used to initiate the division of post-disaster emergency power supply areas using an improved sparrow search algorithm, and to prioritize power supply to key loads within the power protection area. The flexible resource adjustment module is used to obtain the set of network topologies with the least power outage load based on the random spanning tree as the feasible solution set, and solves each feasible topology set by the Lorentz cone optimization method; Network structure adjustment module: used to build a network optimization model for feasible topology sets, laying the necessary foundation for subsequent selection of the optimal topology that meets the post-disaster recovery goals; Group-driven optimization module: This module is used to group the pheromone population in the network optimization model according to the principle of synchronization and parallel driving, thereby improving the optimization capability. Dynamic optimization module: It is used to introduce a dynamic sensing and self-adjustment mechanism, which adaptively adjusts according to the iteration progress and solution quality to obtain the optimal topology result, providing strategies for schedulers to implement post-disaster recovery and control.

10. A computer device and its software medium, characterized in that, include: The memory and processor, and the computer program stored in the memory, when the computer program is executed on the processor, implement the distributed smart grid disaster recovery and control method as described in any one of claims 1-7.

Images