Cooperative guarantee power supply auxiliary decision-making method and system for smart power grid

Abstract

The invention relates to the technical field of smart power grids, in particular to a collaborative guarantee power supply auxiliary decision-making method and system for a smart power grid, and the method comprises the steps: collecting the real-time operation data of a power grid, and generating an initial population containing a plurality of scheduling schemes; calculating a topological weight and a resource margin; constructing a role potential energy index based on the objective function value, the topological weight and the resource margin to allocate role identities in a sparrow search algorithm; performing position updating operation on each scheduling scheme based on the role identity, and performing power flow verification on the scheduling scheme after position updating; and iteratively executing position updating operation until a preset termination condition is met, and taking the scheduling scheme with the minimum target function value as a decision result. Through the technical scheme provided by the invention, the safety of the power grid architecture and the anti-risk capability of coping with sudden risks are considered while the generated auxiliary decision-making scheme pursues the optimal objective function.

Description

Technical Field

[0001] This application relates to the field of smart grid technology, and in particular to a collaborative power supply support decision-making method and system for smart grids. Background Technology

[0002] With the rapid development of smart grid technology, electricity users have increasingly stringent requirements for power supply security. Power supply security work has shifted from traditional experience-based dispatching to digitalization and intelligentization. Power supply security scenarios often involve the coordinated operation of various emergency resources, such as emergency generator dispatching and repair team deployment. As a crucial link in ensuring the safe and stable operation of the power system, how to achieve rapid and accurate coordinated dispatching of various emergency resources in complex and ever-changing power grid operating environments has become an urgent problem to be solved in the power industry.

[0003] Currently, swarm intelligence algorithms such as particle swarm optimization are commonly used to solve this type of multi-constraint, multi-objective cooperative scheduling problem. The sparrow search algorithm, as a type of swarm intelligence algorithm, simulates the foraging behavior of a sparrow flock, dividing individuals into discoverers, joiners, and scouts for iterative optimization, and possesses strong global search capabilities.

[0004] However, traditional sparrow search algorithms typically judge the quality of individuals and guide the search direction solely based on their fitness values ​​when assigning roles and updating the population. They do not consider the impact of the scheduling scheme on the power grid's topology, making the algorithm highly susceptible to being misled by schemes with low objective function values ​​but poor risk resistance. The final decision scheme cannot balance the security of the power grid architecture with its ability to cope with sudden risks, making it difficult to guarantee power supply security. Summary of the Invention

[0005] To address the technical challenge of ensuring power supply security, this application provides a collaborative power supply support decision-making method and system for smart grids. This method ensures that the generated support decision-making scheme, while pursuing the optimal objective function, also considers the security of the power grid architecture and its ability to withstand sudden risks.

[0006] In a first aspect, this application provides a collaborative power supply support decision-making method for smart grids. The method includes: collecting real-time power grid operation data and generating an initial population containing multiple scheduling schemes, wherein the scheduling schemes include switch states and equipment output parameters; for any scheduling scheme, calculating topology weights representing the degree of influence on the power grid architecture and resource margins representing the ability to cope with uncertainties; constructing a role potential index based on the objective function value, topology weights, and resource margins; sorting the scheduling schemes in the population according to the role potential index to assign the roles of discoverer, joiner, and scout in a sparrow search algorithm; performing a position update operation on each scheduling scheme based on the role identity, and performing power flow verification on the scheduling schemes after the position update; iteratively executing the position update operation until a preset termination condition is met, and taking the globally optimal scheduling scheme with the smallest objective function value as the decision result.

[0007] For any scheduling scheme, the topology weight and resource margin are calculated, and then a role potential index is constructed to rank the population and assign roles in the sparrow search algorithm. Since the role potential index comprehensively considers the objective function value of the scheduling scheme, the topology weight, and the resource margin, it ensures that the scheduling scheme of the discoverer, i.e. the population leader, has sufficient resource safety margin, avoiding high-risk schemes from misleading the search direction, thereby improving the reliability and feasibility of the decision results.

[0008] Preferably, the scheduling scheme includes discrete variables representing the switching status of switching equipment, continuous variables representing equipment output commands, and spatial variables representing emergency resource access nodes; the objective function value is a weighted sum of load loss, voltage deviation, and scheduling cost; wherein, the load loss is the power shortage of key users, the voltage deviation is the degree to which the voltage of each node deviates from the rated value, and the scheduling cost includes at least the personnel dispatch distance cost.

[0009] Preferably, the calculation of the topology weights characterizing the degree of influence on the power grid architecture includes: identifying switching nodes whose states have changed in the scheduling scheme; obtaining the betweenness centrality of the switching nodes and the preset power supply protection level of the load supplied by the switching nodes, wherein the betweenness centrality is characterized as the ratio of the number of shortest paths through the switching node in the power grid topology to the total number of shortest paths in the entire network; calculating the product of the betweenness centrality and the load weight coefficient, wherein the load weight coefficient is positively correlated with the preset power supply protection level; and summing the products of all switching nodes whose states have changed to obtain the topology weights.

[0010] Betweenness centrality reflects the hub status of a node in the network, while power protection level reflects the importance of the power supply object. It realizes the precise quantification of the structural influence of the scheduling scheme, so that operations involving key hubs or important power supply objects are given higher weights, preventing the algorithm from ignoring the protection of the core of the power grid during the search process.

[0011] Preferably, the calculation of the resource margin characterizing the ability to cope with uncertainty includes: calculating the difference between the rated capacity of each emergency power source and the planned output in the scheduling scheme to obtain the power margin; calculating the difference between the maximum available time of each repair team and the expected task duration, and taking the ratio of the difference to the maximum available time as the time margin; and calculating the weighted sum of the power margin and the time margin to obtain the resource margin.

[0012] Resource adequacy directly reflects the emergency resources remaining after the current plan is implemented; it enables a quantitative assessment of the risk resistance capability of the scheduling plan, ensuring that the selected scheduling plan has sufficient room for adjustment when dealing with possible future failures or emergencies.

[0013] Preferably, the construction of the role potential index based on the objective function value, topology weights, and resource adequacy includes: calculating the reciprocal of the objective function value as a benchmark score; determining a resource adjustment coefficient, wherein the resource adjustment coefficient is positively correlated with the difference between resource adequacy and resource safety threshold; calculating the product of the topology weights and the resource adjustment coefficients, and adding the product to the benchmark score to obtain the role potential index.

[0014] The role potential index is calculated from multiple dimensions, including the objective function value of the integrated scheduling scheme, topology weight, and resource margin, to ensure that the scheduling scheme of the discoverer, i.e. the population leader, has sufficient resource safety margin and avoids high-risk schemes from misleading the search direction.

[0015] Preferably, the step of sorting the scheduling schemes in the population according to the role potential index to assign the roles of discoverer, joiner and scout in the sparrow search algorithm includes: selecting the scheduling scheme with the highest preset proportion of role potential index as the discoverer; among the remaining scheduling schemes, selecting the scheduling scheme with a topological weight higher than the topological threshold and a resource margin lower than the resource safety threshold as the scout; and using the remaining scheduling schemes as joiners.

[0016] Preferably, the location update operation for each scheduling scheme based on the assigned role identity includes performing the following operations on the discoverer: generating a random warning value and comparing it with a warning safety threshold; if the warning value is less than the warning safety threshold, then causing the discoverer to perform a location update based on exponential decay; if the warning value is greater than or equal to the warning safety threshold, then causing the discoverer to perform a random walk based on a normal distribution to narrow the search range.

[0017] Preferably, the position update operation for each scheduling scheme based on the assigned role identity includes performing the following operations on the joiner: determining whether the objective function value of the joiner is less than the median of the objective function values ​​of all scheduling schemes; if so, moving the joiner closer to the position of the current globally optimal scheduling scheme; if not, moving the joiner away from the globally worst scheduling scheme.

[0018] Preferably, the position update operation for each scheduling scheme based on the assigned role identity includes performing the following operations on the scout: determining whether the objective function value of the scout decreases in a consecutive preset number of iterations; if not, determining that the scout has fallen into a local optimum or a high-risk area, and resetting its position to a random position in the solution space.

[0019] In a second aspect, this application also provides a collaborative power supply support decision-making system for smart grids, including a processor and a memory. The memory stores computer program instructions, which, when executed by the processor, implement the collaborative power supply support decision-making method for smart grids according to the first aspect of this application.

[0020] The technical solution of this application has the following beneficial technical effects: By collecting real-time power grid operation data, a population of scheduling schemes including switch status and equipment output is constructed. The topology weight representing the impact of the power grid architecture and the resource margin representing the ability to cope with uncertainty and risk are calculated. Then, a role potential energy index is constructed to improve the role allocation mechanism of the sparrow search algorithm. The scheduling schemes in the population are divided into discoverers, joiners and scouts. The positions are updated iteratively according to the role identity and power flow verification is performed. This can achieve collaborative optimization in complex power supply scenarios, ensuring that the generated auxiliary decision-making schemes pursue the optimal objective function while taking into account the security of the power grid architecture and the ability to cope with sudden risks. Attached Figure Description

[0021] Figure 1 This is a flowchart of a collaborative power supply support decision-making method for smart grids according to an embodiment of this application.

[0022] Figure 2 This is a graph showing the change of the objective function value of the globally optimal scheduling scheme according to an embodiment of this application with the number of iterations.

[0023] Figure 3 This is a population distribution diagram when the number of iterations is 0 according to the embodiments of this application.

[0024] Figure 4 This is a population distribution diagram when the number of iterations is 30 according to the embodiments of this application.

[0025] Figure 5This is a structural block diagram of a collaborative power supply auxiliary decision-making system for smart grids according to an embodiment of this application. Detailed Implementation

[0026] According to a first aspect of this application, this application provides a collaborative power supply support decision-making method for smart grids. Figure 1 This is a flowchart of a collaborative power supply support decision-making method for smart grids according to an embodiment of this application. Figure 1 As shown, the collaborative power supply auxiliary decision-making method for smart grids includes steps S101 to S105, which are described in detail below.

[0027] S101, collect real-time power grid operation data and generate an initial population containing multiple scheduling schemes, wherein the scheduling schemes include switch status and equipment output parameters.

[0028] In one embodiment, intelligent sensing devices and terminal acquisition devices deployed at the power supply site are first used to acquire real-time voltage and current data of the power grid, as well as real-time status data of various emergency resources, including at least emergency repair teams and mobile power supply vehicles. Subsequently, the collaborative power supply problem is transformed into a search problem in a multi-dimensional solution space, with the scheduling scheme treated as an individual in a sparrow search algorithm. The scheduling scheme contains a set vector of all decision variables.

[0029] The scheduling scheme includes discrete variables representing the switching status of switching equipment, continuous variables representing equipment output commands, and spatial variables representing emergency resource access nodes; the objective function value is a weighted sum of load loss, voltage deviation, and scheduling cost.

[0030] Specifically, discrete variables correspond to the switching states of key interconnection switches and sectionalizing switches, taking values ​​of 0 or 1, and are used to determine the power supply path topology of the power grid; continuous variables correspond to the active power output command values ​​of equipment such as emergency power generation vehicles, mobile energy storage vehicles, and distributed power sources, in kilowatts; spatial variables correspond to the node numbers or geographical coordinates of the deployment nodes of emergency repair teams and mobile power vehicles, where the nodes are any grid-connected nodes in the smart grid, and power can be supplied to the smart grid through the nodes.

[0031] In the objective function value, the load loss is the power shortage of key users, the voltage deviation is the degree to which the voltage of each node deviates from the rated value, and the scheduling cost includes at least the personnel scheduling distance cost.

[0032] To quantitatively evaluate the merits of scheduling schemes, an objective function is constructed, which satisfies the following relationship: ; In the formula, The objective function value; , , These are the weighting coefficients for load loss, voltage deviation, and dispatch cost, respectively. Power shortages for critical users; The degree to which the voltage at each node deviates from the rated value; For scheduling costs, , and These are preset normalization coefficients for load loss, voltage deviation, and dispatch cost, respectively, used to eliminate the problem of inconsistent dimensions of load loss, voltage deviation, and dispatch cost.

[0033] Among them, the key users refer to core power-consuming sites such as schools and hospitals. , , The values ​​are 0.5, 0.3 and 0.2 respectively, and the scheduling cost also includes the fuel cost of the mobile power vehicle.

[0034] In this way, a multi-objective evaluation system for power supply support decision-making was established, laying a digital foundation for subsequent intelligent optimization.

[0035] S102, for any scheduling scheme, calculate the topology weights that characterize the degree of impact on the power grid architecture and the resource margins that characterize the ability to cope with uncertainties.

[0036] In one embodiment, the topology weight of any scheduling scheme is calculated. Specifically, the calculation of the topology weight, which characterizes the degree of influence on the power grid architecture, includes: identifying switching nodes whose states have changed in the scheduling scheme; obtaining the betweenness centrality of the switching node and the preset power supply guarantee level of the load supplied by the switching node, wherein the betweenness centrality is characterized as the ratio of the number of shortest paths through the switching node in the power grid topology to the total number of shortest paths in the entire network; calculating the product of the betweenness centrality and the load weight coefficient, wherein the load weight coefficient is positively correlated with the preset power supply guarantee level; and summing the products of all switching nodes whose states have changed to obtain the topology weight.

[0037] The topology weights of the i-th scheduling scheme satisfy the following relationship: ; In the formula, The topology weight represents the i-th scheduling scheme; This is the set of switch nodes whose states change in this scheme; The betweenness centrality of the switch node n; The preset power protection level for the load supplying power to switch node n; This is the load weighting coefficient; the load weighting coefficient is set to 0.5 and is used to adjust the impact of the power supply protection level on the topology weight.

[0038] Betweenness centrality is used to characterize the hub status of the corresponding switching node in the power grid energy transmission path. Switching nodes with high betweenness centrality are key positions in the power grid energy transmission process. The preset power supply level is the importance level set for the users or sites served by the switching node. For example, the importance level of schools and hospitals is greater than that of residential areas. If a dispatching scheme operates a switch that is in a hub position and involves important users, its topology weight will increase significantly.

[0039] It should be noted that after obtaining the topology weights of each scheduling scheme, in order to accurately characterize the relative magnitude of the topology weights of each scheduling scheme, it is necessary to perform a normalization operation on the topology weights, that is, to divide the topology weight of any scheduling scheme by the maximum value of the topology weights of all scheduling schemes.

[0040] The calculation of resource margin, which characterizes the ability to cope with uncertainty, includes: calculating the difference between the rated capacity of each emergency power source and the planned output in the scheduling scheme to obtain the power margin; calculating the difference between the maximum available time of each repair team and the expected task duration, and using the ratio of this difference to the maximum available time as the time margin; and calculating the weighted sum of the power margin and the time margin to obtain the resource margin.

[0041] The resource margin of the i-th scheduling scheme satisfies the following relationship: ; In the formula, Let be the resource margin of the i-th scheduling scheme; For power margin; For time margin; and These are the weighting coefficients for power margin and time margin, respectively. In this embodiment, .

[0042] Thus, topology weights characterize the impact of the scheduling scheme on the power grid architecture; the larger the topology weight, the more important the switching nodes involved in the scheduling scheme. Resource margins characterize the robustness of the scheduling scheme in dealing with future uncertainties and emergencies.

[0043] S103, construct a role potential index based on the objective function value, topological weight and resource margin, and sort the scheduling schemes in the population according to the role potential index to assign the roles of discoverer, joiner and scout in the sparrow search algorithm.

[0044] In one embodiment, in the sparrow search algorithm, all scheduling schemes can be divided into discoverers, joiners, and scouts. The discoverers are responsible for conducting a wide-area search in the solution space to find a high-quality scheduling scheme to guide the search direction of the population. The joiners are responsible for following the discoverers and making fine adjustments near the current high-quality scheduling scheme. The scouts are responsible for being wary of getting trapped in local optima and randomly jumping to a new scheduling scheme when they find that the current scheduling scheme is stagnant or the risk is too high.

[0045] The method for constructing a role potential index based on the objective function value, topology weights, and resource adequacy includes: calculating the reciprocal of the objective function value as a benchmark score; determining a resource adjustment coefficient, wherein the resource adjustment coefficient is positively correlated with the difference between resource adequacy and resource safety threshold; calculating the product of the topology weights and the resource adjustment coefficients, and adding the product to the benchmark score to obtain the role potential index.

[0046] The role potential index of the i-th scheduling scheme Satisfying the relation: ; In the formula; Let be the objective function value of the i-th scheduling scheme. As a preset positive value, in this embodiment The value is 1; Let be the topology weight of the i-th scheduling scheme. Let i be the resource margin of the i-th scheduling scheme. As a resource security threshold, and These are the topology adjustment coefficient and the resource adjustment coefficient, respectively. The resource security threshold is set to 0.2; the topology adjustment coefficient... The value is 1, which is the resource adjustment coefficient. The value is 10.

[0047] Understandably, The baseline score is used; when the resource margin is lower than the resource safety threshold, then... The value is relatively large. The value of approaches zero. Regardless of the topology weight, the role potential index of the i-th scheduling scheme will be maintained near the baseline score. This effectively prevents high-risk schemes with better objective function values ​​but weak risk resistance from becoming the leaders of the population, thus avoiding decision traps.

[0048] When resource margin is lower than the resource safety threshold When the value of approaches 1, at this point, The value of approaches In other words, under the premise that the resources are sufficient and the state is safe, the role potential index is mainly affected by the topology weight. The higher the topology weight, the higher the role potential index of the scheduling scheme, and thus it is selected as the discoverer first. This ensures that the algorithm can actively select those high-quality scheduling schemes with strong risk resistance to guide the evolution of the population, thereby accelerating convergence and obtaining a decision result that takes into account both safety and global optimality.

[0049] In this embodiment, the step of sorting the scheduling schemes in the population according to the role potential index to assign the roles of discoverer, joiner, and scout in the sparrow search algorithm includes: selecting the scheduling scheme with the highest preset proportion of role potential index as the discoverer; among the remaining scheduling schemes, selecting the scheduling scheme with a topological weight higher than the topological threshold and a resource margin lower than the resource safety threshold as the scout; and using the remaining scheduling schemes as joiners.

[0050] For example, the preset ratio can be set to 20% of the total population; for scouts, select high-risk scheduling schemes that have high structural influence but are in a state of resource depletion to ensure that they can escape high-risk areas.

[0051] In this way, the allocation mechanism ensures that the scheduling scheme that becomes the population leader has sufficient resource margin, while forcing the use of scouts to explore high-risk areas, thus preventing the algorithm from getting stuck in local optima in high-risk areas.

[0052] S104: Perform position update operations on each scheduling scheme based on role identity, and perform power flow verification on the scheduling scheme after the position update.

[0053] In one embodiment, the roles include discoverer, joiner, and scout. Different location update operations are performed on each scheduling scheme according to the role. The location update operations for discoverer, joiner, and scout are described below in turn.

[0054] Perform the following operations on the discoverer: generate a random warning value and compare it with a warning safety threshold; if the warning value is less than the warning safety threshold, cause the discoverer to perform a location update based on exponential decay; if the warning value is greater than or equal to the warning safety threshold, cause the discoverer to perform a random walk based on a normal distribution to narrow the search range.

[0055] When the warning value Less than the warning safety threshold When needed, the position is updated using the following formula: ; When the warning value Greater than or equal to the warning safety threshold When needed, the position is updated using the following formula: ; In the formula, and This represents the position of the i-th scheduling scheme in the t-th iteration and the (t+1)-th iteration. The ranking of the i-th scheduling scheme in the population, arranged from largest to smallest according to the role potential index; This is the random step size control parameter, with a value of 1; This represents the maximum number of iterations. These are random numbers that follow a normal distribution. It is a unit vector whose values ​​are all 1.

[0056] For the joiner, the position update operation for each scheduling scheme based on the assigned role identity includes performing the following operations on the joiner: determining whether the objective function value of the joiner is less than the median of the objective function values ​​of all scheduling schemes; if so, moving the joiner closer to the position of the current globally optimal scheduling scheme; if not, moving the joiner away from the globally worst scheduling scheme.

[0057] Specifically, if the entrant is in an advantageous position, i.e., its objective function value is less than the median of the objective function values ​​of all scheduling schemes, the update formula for the entrant is: ; In the formula, It is the globally optimal position in the (t+1)th iteration, that is, the scheduling scheme with the minimum objective function value; This is a local fine-tuning coefficient used to control the tightness of the following; in this embodiment, the local fine-tuning coefficient is set to 0.1. A matrix randomly assigned a value of 1 or negative 1, used to control the direction of the search.

[0058] If the new participant is in a disadvantaged position, meaning its objective function value is not less than the median of the objective function values ​​of all scheduling schemes, the update formula for the new participant is: In the formula, For random numbers that follow a normal distribution, Let be the worst-case scheduling scheme in the t-th iteration. Let be the ranking of the i-th scheduling scheme in the population, arranged from largest to smallest according to the role potential index.

[0059] For the scout, the position update operation based on the assigned role identity for each scheduling scheme includes performing the following operations on the scout: determining whether the objective function value of the scout decreases in a preset number of iterations; if not, determining that the scout has entered a local optimum or a high-risk region, and resetting its position to a random position within the solution space. The preset number of iterations is set to 3.

[0060] Specifically, the reset operation uses the following formula: In the formula, and These are the lower and upper bounds of the solution space, respectively; It is a random vector.

[0061] After the location update is completed, the Newton-Raphson method or DC power flow model is used to perform power flow verification on all new scheduling schemes. If the verification results show that the node voltage exceeds the limit or the line is overloaded, a penalty is imposed on the objective function value of the scheduling scheme, that is, the objective function value of the scheduling scheme is increased, so that it is at a disadvantage in the next round of competition.

[0062] Thus, by employing a differentiated position update strategy, we can ensure rapid convergence to the global optimum while maintaining population diversity and the ability to escape local optima.

[0063] S105, iteratively execute the position update operation until the preset termination condition is met, and take the globally optimal scheduling scheme with the minimum objective function value as the decision result.

[0064] In one embodiment, steps S102 to S104 are repeated. In each iteration, as the virtual operating state of the power grid changes, the topology weights and resource margins of each scheduling scheme also change dynamically, leading to a redistribution of roles. A preset termination condition is reached when the number of iterations reaches a preset maximum, or when the objective function value of the globally optimal scheduling scheme changes by less than a change threshold over several consecutive generations. The maximum number of iterations is 100, and the change threshold is 0.01.

[0065] Please see Figure 2 The graph shows the change in the objective function value of the globally optimal scheduling scheme according to an embodiment of this application as a function of the number of iterations. The graph shows that the objective function value of the globally optimal scheduling scheme gradually decreases with increasing iteration count, and the change becomes gradual at iteration count 30. Please also refer to... Figure 3 as well as Figure 4 , Figure 3 This is a population distribution diagram based on an embodiment of this application when the iteration count is 0. In the two-dimensional space composed of decision variable projection dimension 1 and decision variable projection dimension 2, the distribution of scheduling schemes in the population is relatively chaotic; while Figure 4 The population distribution diagram shown in the embodiment of this application with 30 iterations shows that, in the two-dimensional space composed of decision variable projection dimension 1 and decision variable projection dimension 2, the scheduling schemes in the population with 30 iterations are concentrated near the globally optimal scheduling scheme.

[0066] The globally optimal scheduling scheme with the minimum objective function value is determined and decoded into specific control instructions, including action instructions for switches, output instructions for generators, and scheduling path instructions for repair teams.

[0067] Thus, in complex scenarios involving multi-resource collaborative scheduling, the system accurately outputs optimal auxiliary decision-making suggestions that balance security and robustness for schedulers to reference.

[0068] According to a second aspect of this application, this application also provides a collaborative power supply support decision-making system for smart grids. Figure 5 This is a structural block diagram of a collaborative power supply support decision-making system for smart grids, according to an embodiment of this application. Figure 5 As shown, the system 50 includes a processor and a memory. The memory stores computer program instructions, which, when executed by the processor, implement a collaborative power supply auxiliary decision-making method for smart grids according to the first aspect of this application. The system also includes other components well-known to those skilled in the art, such as a communication bus and a communication interface. Their configuration and functions are known in the art and will not be described further here.

[0069] It should be noted that, for those skilled in the art, several modifications and improvements can be made without departing from the concept of this application, and these all fall within the scope of protection of this application.

Claims

1. A collaborative power supply guaranteeing auxiliary decision method for a smart grid, characterized in that, The decision method comprises: collecting real-time operation data of a power grid and generating an initial population comprising a plurality of scheduling schemes, the scheduling schemes comprising switch states and device output parameters; For any scheduling scheme, a topological weight representing the degree of influence on the power grid architecture and a resource margin representing the ability to cope with uncertainty are calculated; A role potential index is constructed based on the target function value, the topological weight and the resource margin, and the scheduling schemes in the population are sorted according to the role potential index to assign the roles of discoverer, joiner and scout in the sparrow search algorithm; Based on the role identity, a position updating operation is performed on each scheduling scheme, and after the position updating, a power flow check is performed on the scheduling scheme; The position updating operation is iteratively performed until a preset termination condition is met, and a global optimal scheduling scheme with the smallest target function value is taken as the decision result.

2. The method of claim 1, wherein, The scheduling scheme comprises discrete variables representing the switch states of switch devices, continuous variables representing device output instructions, and spatial variables representing emergency resource access nodes; the target function value is the weighted sum of load loss, voltage deviation and scheduling cost; The load loss is the power shortage of key users, the voltage deviation is the degree of deviation of the voltage of each node from the rated value, and the scheduling cost at least includes personnel scheduling distance cost. 3.The method of claim 1, wherein, The calculation of the topological weight representing the degree of influence on the power grid architecture comprises: Identify the switch nodes in the scheduling scheme whose states change; Obtain the betweenness centrality of the switch node and the preset power protection level of the power load supplied by the switch node, and the betweenness centrality represents the ratio of the number of shortest paths passing through the switch node in the power grid topology to the total number of shortest paths in the whole network; The product of the betweenness centrality and the load weight coefficient is calculated, and the load weight coefficient is positively correlated with the preset power protection level; The products of all switch nodes whose states change are added to obtain the topological weight.

4. The method of claim 1, wherein, The calculation of the resource margin representing the ability to cope with uncertainty comprises: The difference between the rated capacity of each emergency power supply and the planned output in the scheduling scheme is calculated to obtain the power margin; the difference between the maximum available duration of each repair team and the expected task duration is calculated, and the ratio of the difference to the maximum available duration is taken as the time margin; The weighted sum of the power margin and the time margin is calculated to obtain the resource margin.

5. The method of claim 1, wherein, The construction of the role potential index based on the target function value, the topological weight and the resource margin comprises: calculating the reciprocal of the target function value as a benchmark score; A resource adjustment coefficient is determined, which is positively correlated with the difference between the resource margin and the resource safety threshold; The product of the topological weight and the resource adjustment coefficient is calculated, and the product and the benchmark score are added to obtain the role potential index.

6. The method of claim 5, wherein the method further comprises: The sorting of the scheduling schemes in the population according to the role potential index to assign the roles of discoverer, joiner and scout in the sparrow search algorithm comprises: A preset proportion of scheduling schemes with the highest role potential index are selected as discoverers; Among the remaining scheduling schemes, the scheduling schemes with a topological weight higher than a topological threshold and a resource margin lower than a resource safety threshold are selected as scouts; The remaining scheduling schemes are taken as joiners.

7. The method of claim 1, wherein, The position updating operation based on the assigned role identity for each scheduling scheme includes the following operations for the discoverer: generating a random warning value and comparing it with a warning safety threshold; if the warning value is less than the warning safety threshold, the discoverer performs position updating based on exponential decay; if the warning value is greater than or equal to the warning safety threshold, the discoverer performs random walk based on normal distribution to shrink the search range. 8.The method of claim 1, wherein, The position updating operation based on the assigned role identity for each scheduling scheme includes the following operations for the joiner: judging whether the target function value of the joiner is less than the median of all scheduling scheme target function values; if yes, the joiner moves towards the position of the current global optimal scheduling scheme; if no, the joiner moves away from the global worst scheduling scheme.

9. The method of claim 1, wherein, The position updating operation based on the assigned role identity for each scheduling scheme includes the following operations for the scout: judging whether the target function value of the scout is reduced in continuous preset number of iterations; if no, it is determined that the scout is trapped in local optimum or high risk area, and its position is reset to a random position in the solution space.

10. A collaborative power supply security auxiliary decision system for a smart grid, characterized in that, The device comprises a processor and a memory, and the memory stores computer program instructions which, when executed by the processor, implement the method for intelligent power grid cooperative power supply auxiliary decision according to any one of claims 1 to 9.

Images