A distributed power dispatching communication topology optimization method and related device

Abstract

The application discloses a distributed power dispatching communication topology optimization method and a related device, relates to the technical field of power dispatching communication optimization, and introduces the concept of a characteristic ratio to represent the ratio of algebraic connectivity and spectral radius. The algebraic connectivity is a quantitative index of convergence rate, and the spectral radius is a quantitative index of trigger interval. Therefore, the performance indexes of the power system in terms of communication resources and convergence rate of the consensus algorithm are taken into account. On this basis, an edge is selected from the set of edges of the complementary graph of the communication topology graph in a way of adding edges, the selected edge is added to the communication topology graph to update the Laplace matrix corresponding to the communication topology graph, the result of each iteration can make the current characteristic ratio of the communication topology graph reach the maximum, and therefore the performance of the power system in terms of the convergence rate of the consensus algorithm and communication resources is improved, and the stable operation of the power system is ensured.

Description

Technical Field

[0001] This application relates to the field of power dispatch communication optimization technology, and in particular to a distributed power dispatch communication topology optimization method and related apparatus. Background Technology

[0002] Currently, distributed economic dispatch strategies are widely used in power system economic dispatch due to their ability to efficiently and flexibly optimize resource allocation and reduce operating costs. Distributed economic dispatch strategies utilize multi-agent consensus algorithms to treat each generating unit as an agent, coordinating information and decisions among these units to achieve a globally optimal generation strategy. In distributed economic dispatch of power systems based on multi-agent consensus algorithms, information transmission between generating units relies on the communication topology network; therefore, optimizing the communication topology plays a crucial role in improving the performance of the power system in distributed economic dispatch. In traditional consensus algorithms, each generating unit requires continuous information transmission through the communication topology, which wastes the power system's communication resources, reduces the system response rate, and results in a low convergence rate for the consensus algorithm. Therefore, how to improve the performance of the power system in terms of communication resources and consensus algorithm convergence rate through topology optimization, while ensuring the stable operation of the power system, has become a pressing technical problem to be solved in this field. Summary of the Invention

[0003] The purpose of this application is to provide a distributed power dispatch communication topology optimization method and related apparatus, which can improve the performance of the power system in terms of both consensus algorithm convergence rate and communication resources, and ensure the stable operation of the power system.

[0004] To achieve the above objectives, this application provides the following solution:

[0005] Firstly, this application provides a method for optimizing the topology of distributed power dispatch communication, which includes the following steps:

[0006] Step S1: Using the basic knowledge of graph theory, model the distributed economic dispatch communication topology network of the power system to obtain the distributed economic dispatch communication topology network model of the power system; each node in the distributed economic dispatch communication topology network model of the power system represents a power generation unit, and each edge represents a communication link between two power generation units;

[0007] Step S2: Determine the characteristic ratio based on the power system distributed economic dispatch communication topology network model; the characteristic ratio is the ratio of the algebraic connectivity to the spectral radius of the Laplace matrix corresponding to the power system distributed economic dispatch communication topology network model.

[0008] Step S3: Based on the power system distributed economic dispatch communication topology network model and the characteristic ratio, establish a topology optimization model;

[0009] Step S4: Based on the topology optimization model, select an edge from the edge set of the complement graph of the communication topology graph that maximizes the current feature ratio increment, add the edge to the communication topology graph, update the Laplacian matrix corresponding to the communication topology graph, and obtain the updated Laplacian matrix.

[0010] Step S5: Detect whether the number of communication links to be added has been reached. If so, determine the optimized communication topology based on the updated Laplace matrix. Otherwise, return to step S4 and continue searching for communication links that meet the requirements from the remaining edge set of the communication topology until the number of communication links to be added is reached, and determine the optimized communication topology based on the updated Laplace matrix.

[0011] In a second aspect, this application provides a computer device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the distributed power dispatch communication topology optimization method described in the first aspect.

[0012] Thirdly, this application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the distributed power dispatch communication topology optimization method described in the first aspect.

[0013] Fourthly, this application provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the distributed power dispatch communication topology optimization method described in the first aspect.

[0014] According to the specific embodiments provided in this application, the following technical effects are disclosed:

[0015] This application provides a method and related apparatus for optimizing the communication topology of distributed power dispatching. By introducing the concept of eigenratio, which represents the ratio of algebraic connectivity to spectral radius, the method balances the performance of the power system in terms of both communication resources and consensus algorithm convergence rate. Furthermore, by adding an edge, an edge is selected from the edge set of the complement graph of the communication topology graph that maximizes the current eigenratio increment. This edge is then added to the communication topology graph to update the corresponding Laplacian matrix. This ensures that each iteration maximizes the current eigenratio of the communication topology graph, thereby improving the performance of the power system in terms of both consensus algorithm convergence rate and communication resources, and guaranteeing the stable operation of the power system. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is a flowchart illustrating a distributed power dispatch communication topology optimization method provided in an embodiment of this application.

[0018] Figure 2 This is a schematic diagram of the MGEA algorithm provided in an embodiment of this application.

[0019] Figure 3 This is a schematic diagram of the structure of an IEEE 10-machine 39-node system provided in an embodiment of this application.

[0020] Figure 4 This is an original communication topology diagram provided for one embodiment of this application.

[0021] Figure 5 This is a graph showing the trend of topological feature ratio as edge addition, provided in an embodiment of this application.

[0022] Figure 6 A communication topology graph based on algebraic connectivity optimization is provided as an embodiment of this application.

[0023] Figure 7 This is a communication topology diagram based on spectral radius optimization provided in one embodiment of this application.

[0024] Figure 8 This is a communication topology diagram based on eigenratio optimization provided in one embodiment of this application.

[0025] Figure 9 This is a simulation diagram of distributed economic scheduling based on an event-triggered consensus algorithm under the original topology, provided as an embodiment of this application.

[0026] Figure 10 This is a simulation diagram of a distributed economic scheduling algorithm based on the original topology and the event-triggered consensus algorithm under an embodiment of this application.

[0027] Figure 11 A simulation diagram of distributed economic scheduling based on an event-triggered consensus algorithm under an algebraic connectivity optimization topology with consistency variables, provided as an embodiment of this application.

[0028] Figure 12 A simulation diagram of distributed economic scheduling of event-triggered consensus algorithm under algebraic connectivity optimization topology based on event triggering time provided in an embodiment of this application.

[0029] Figure 13 A simulation diagram of distributed economic scheduling of event-triggered consensus algorithm under spectral radius optimization topology based on consistency variables, provided as an embodiment of this application.

[0030] Figure 14 A simulation diagram of distributed economic scheduling of event-triggered consensus algorithm under spectral radius optimization topology based on event triggering time provided in an embodiment of this application.

[0031] Figure 15 A simulation diagram of distributed economic scheduling of event-triggered consensus algorithm under eigenratio optimization topology based on consistency variables, provided as an embodiment of this application.

[0032] Figure 16 A simulation diagram of distributed economic scheduling of event-triggered consensus algorithm under eigenratio optimization topology based on event triggering time provided in an embodiment of this application.

[0033] Figure 17 This is a system simulation diagram based on load demand changes using a consistency variable, provided as an embodiment of this application.

[0034] Figure 18 This is a system simulation diagram based on load demand changes at the event triggering time, provided as an embodiment of this application.

[0035] Figure 19 This is a system simulation diagram based on load demand changes in output power, provided as an embodiment of this application.

[0036] Figure 20 This is a system simulation diagram under load demand changes based on system supply and demand balance, provided as an embodiment of this application.

[0037] Figure 21This is a system simulation diagram of a power component "plug and play" based on a consistency variable, provided as an embodiment of this application.

[0038] Figure 22 This is a system simulation diagram of a power component under "plug and play" based on the event triggering time, provided as an embodiment of this application.

[0039] Figure 23 A system simulation diagram of a power element based on output power in a "plug and play" configuration provided in an embodiment of this application.

[0040] Figure 24 A system simulation diagram of a power component under "plug and play" based on system supply and demand balance provided in an embodiment of this application.

[0041] Figure 25 A system simulation diagram under communication link failure based on consistency variables is provided for an embodiment of this application.

[0042] Figure 26 This is a system simulation diagram under a communication link failure based on output power, provided as an embodiment of this application. Detailed Implementation

[0043] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0044] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0045] like Figure 1 As shown in the figure, this embodiment proposes a distributed power dispatch communication topology optimization method, which specifically includes the following steps:

[0046] Step S1: Using the basic knowledge of graph theory, model the distributed economic dispatch communication topology network of the power system to obtain the distributed economic dispatch communication topology network model of the power system; each node in the distributed economic dispatch communication topology network model of the power system represents a power generation unit, and each edge represents a communication link between two power generation units.

[0047] In this embodiment, when establishing the distributed economic dispatch communication topology network model of the power system in step S1, based on the fundamental knowledge of graph theory, let graph G represent the communication topology of the power system, denoted as G = (V, E), where V = {v1, v2, ..., v...} n} represents the set of nodes, indicating the generating units in the power system; E = {e ij =(v i ,v j Let} be the set of edges, representing the communication links between power generation units. In an undirected connected graph, if node v i With v j If there is an edge between two nodes, then those two nodes are each other's neighbor nodes, denoted as a. ij =1 and a ji =1, otherwise a ij =0 and a ji =0; the adjacency matrix consists of element a ij Composition, representing the link relationship between nodes, denoted as A = [a ij ]. Node v i The number of edges a node has is called its degree, denoted as d(v). i The degree matrix is ​​composed of elements d(v) i A diagonal matrix composed of d(v1), d(v2), ..., d(v...) is denoted as D = diag{d(v1), d(v2), ..., d(v...)}. n )}.

[0048] In an undirected connected graph, the Laplacian matrix is ​​represented as L = DA. Its eigenvalues ​​satisfy 0 = λ₁(L) ≤ λ₂(L) ≤ … ≤ λ n (L)=λ max The second smallest eigenvalue λ2(L) is the Fiedler eigenvalue, also known as the algebraic connectivity of an undirected connected graph. Its corresponding normalized eigenvector is the Fiedler vector. Algebraic connectivity reflects the strength of graph connectivity in an undirected connected graph; the largest eigenvalue λ max The spectral radius of the Laplace matrix L reflects the stability of the topological graph.

[0049] This embodiment focuses on constructing a power system economic dispatch model within the consensus algorithm. The economic dispatch problem aims to find the optimal power generation capacity of each generating unit under system constraints, thereby minimizing the total power generation cost of the system. The cost function of the i-th generating unit is:

[0050] F i (P Gi ) = a i +b i P Gi +c i P Gi 2 (1).

[0051] In equation (1), P Gi Let a be the output power of the i-th power generation unit. i b i c i This is the cost coefficient for the power generation unit.

[0052] This embodiment takes into account that the power system needs to meet supply and demand balance and the output power limit of each generation unit during the dispatching process. Therefore, the economic dispatching problem can be described as the following optimization problem:

[0053]

[0054] In equation (2), The total cost of power generation is given by n, which represents the total number of power generation units. P represents the total power generation of the system. D To meet system load requirements; and These are the minimum and maximum output power of the i-th power generation unit, respectively.

[0055] The optimal solution to the above optimization problem can be found using the Lagrange multiplier method. That is:

[0056]

[0057] In equation (3), λ is a Lagrange multiplier. For the system power difference, the function L with respect to P Gi Taking the partial derivatives of λ and setting them equal to zero, we get:

[0058]

[0059] achievable That is, the power system's generation cost is minimized when the incremental cost rate of all generation units converges to λ.

[0060] This embodiment addresses the cost increment λ of each power generation unit using an event-triggered consensus algorithm. i As a consistency variable, combined with information from neighboring power generation units, the consistency variables of each power generation unit converge to a consensus through multiple iterations. Simultaneously, to alleviate the communication burden on the power system during the algorithm's computation, this embodiment introduces an event-triggered mechanism. Therefore, the event-triggered consensus algorithm considering system supply and demand balance can be expressed as:

[0061]

[0062] In equation (5), ε is the system power difference adjustment coefficient. and Let represent the consistency variables of the i-th and j-th power generation units at the time of event triggering, respectively. The event triggering function can be expressed as:

[0063]

[0064] In equation (6), Let the state error function be... This indicates the trigger time of the k-th event in the i-th power generation unit. For threshold function, satisfy when When the event triggering condition is met, the power generation unit communicates with neighboring nodes and updates the consistency variables, while simultaneously clearing the state error function to zero.

[0065] Step S2: Determine the characteristic ratio based on the power system distributed economic dispatch communication topology network model; the characteristic ratio is the ratio of the algebraic connectivity to the spectral radius of the Laplace matrix corresponding to the power system distributed economic dispatch communication topology network model.

[0066] In this embodiment, step S2 specifically includes the following steps:

[0067] Step S21: Determine the Laplace matrix based on the distributed economic dispatch communication topology network model of the power system.

[0068] Step S22: Determine the algebraic connectivity and the spectral radius based on the Laplace matrix.

[0069] Step S23: Calculate the eigenratio based on the algebraic connectivity and the spectral radius.

[0070] This embodiment designs a communication topology optimization method based on a greedy algorithm. First, it analyzes the quantitative indicators affecting system performance and then optimizes the power system communication topology by adding edges. Adding edges improves the connectivity of the communication topology, which is represented by an increase in the algebraic connectivity λ²(L) in the Laplace matrix of graph G. A larger algebraic connectivity λ²(L) results in a faster convergence rate for the consensus algorithm; therefore, adding edges can improve the convergence rate of consensus variables in distributed economic dispatch of the power system.

[0071] In event triggering, for an undirected connected graph, if the event triggering condition for each agent in the undirected connected graph is... Then the interval between any two adjacent event trigger times in the system satisfies Therefore, this multi-agent system does not exhibit Zeno-like behavior. Hence, it can be deduced that... The larger the value, the longer the trigger interval, and the fewer communications between power generation units in the power system. This is due to the threshold function... satisfy Right now The upper bound is determined by the topological graph radius, therefore the problem of trigger interval size can be transformed into the problem of topological graph radius size, i.e., the spectral radius λ. max The smaller the value, the larger the trigger interval.

[0072] This embodiment will consider both the algebraic connectivity λ2(L) of the quantization index for convergence rate and the spectral radius λ of the quantization index for trigger interval. max Combining the two, a new quantitative indicator, the characteristic ratio ξ, is introduced:

[0073]

[0074] In equation (7), ξ represents the characteristic ratio, λ²(L) is the algebraic connectivity, and λ max denoted as spectral radius.

[0075] Clearly, the larger the eigenvalue ratio, the more significant the performance improvement in saving communication resources and the convergence rate of the consensus algorithm. Therefore, the idea of ​​the topology optimization algorithm in this embodiment is: based on the same communication resource consumption for each edge addition, select the optimal edge to maximize the eigenvalue ratio of the current topology graph after adding the edge. Thus, the communication topology optimization problem of distributed economic dispatch in power systems can be transformed into the problem of maximizing the eigenvalue ratio of the current topology graph after each edge addition.

[0076] Step S3: Establish a topology optimization model based on the power system distributed economic dispatch communication topology network model and the characteristic ratio.

[0077] In this embodiment, when establishing the topology optimization model, an undirected connected graph G(V0,E0) is given to represent the initial communication topology, i.e., the original communication topology graph, where k is the number of communication links to be added, and d... i Given the maximum degree constraint of the i-th generating unit in the power system, the problem of maximizing the eigenvalue of the current topology graph after each edge addition can be described by the following topology optimization model for optimization:

[0078]

[0079] In equation (8), L represents the Laplace matrix, E0 is the existing edge set in the original communication topology graph, ΔE is the edge set to be added to the original communication topology graph, and E c Let A be the edge set of the complement graph, i.e., the edge set in the complement graph of graph G, containing the candidate edges to be added. Let A represent the adjacency matrix, b be an n-dimensional column vector of all ones, and d be the edge set of graph G. i The resulting n-dimensional column vector.

[0080] In graph G, let h be the edge vector connecting the i-th power generation unit (i.e., node i) and the j-th power generation unit (i.e., node j).e , and h e To form an n-dimensional column vector where the i-th component is 1, the j-th component is -1, and all other components are 0, the Laplacian matrix of graph G can be represented as the sum of the dot products of the edge vectors. Assuming the topological graph has m edges, the Laplacian matrix can be expressed as:

[0081]

[0082] Combining equations (8) and (9), the above topology optimization model can be rewritten as:

[0083]

[0084] In equation (10), L0 is the Laplace matrix of the original communication topology, and ΔL is the change in the Laplace matrix after adding ΔE. e is E c The index of the candidate edge, x e As a Boolean variable, when x e When x = 1, edge e in the representation is selected into ΔE, when x e When = 0, it means E c Edge e was not selected in ΔE. x is a variable of length |E. c | a column vector, where the elements are determined by x e Composition: 1 represents an all-one vector, and T represents the transpose operation.

[0085] Step S4: Based on the topology optimization model, select an edge from the edge set of the complement graph of the communication topology graph that maximizes the current feature ratio increment, add the edge to the communication topology graph, update the Laplacian matrix corresponding to the communication topology graph, and obtain the updated Laplacian matrix.

[0086] This embodiment focuses on topology optimization algorithm design. Based on the aforementioned algorithmic ideas and model, an MGEA (Modified Greedy Edge Addition) algorithm is designed. Essentially, it's a topology optimization strategy based on a greedy algorithm for adding edges. The specific design process of the MGEA algorithm is as follows:

[0087] Algebraic connectivity can be expressed as:

[0088]

[0089] In equation (11), y is an n-dimensional non-zero column vector orthogonal to the all-1 vector 1. Substituting the normalized vector γ = y / ||y|| into equation (11), we get:

[0090] λ²(L(x))=min{γ T L(x)γ|||γ‖=1,1 Τγ=0} (12).

[0091] When the normalized vector γ in equation (12) is a Fiedler vector, the following equation can be obtained from matrix theory:

[0092] L(x)γ=λ2(L(x))γ (13).

[0093] Multiply both sides of the expression by γ T We can obtain:

[0094] γ T L(x)γ=γ T λ2(L(x))γ (14).

[0095] Since γ is a normalized vector, and combining it with equation (14), we can deduce that:

[0096] γ T λ²(L(x))γ=λ²(L(x))(γ T γ)=λ2(L(x))=γ T L(x)γ (15).

[0097] Therefore, when γ is a Fiedler vector, the minimum value of equation (12) can be obtained as:

[0098] λ²(L(x))=γ T L(x)γ (16).

[0099] This indicates that the Fiedler vector γ can relate algebraic connectivity to the Laplace matrix.

[0100] From equation (10), we can see that the Laplace matrix after adding the edges is:

[0101]

[0102] At this point, L(x) is x e The function is λ2(L(x)), therefore λ2(L(x)) is a function of x. e The first-order partial derivative can be expressed as:

[0103]

[0104] Combining equations (17) and (18), since L0 is not x before adding the edge, e The function is given by , therefore we can obtain:

[0105]

[0106] In equation (19), γ i and γ jLet represent the i-th and j-th terms of the Fiedler vector γ of the current Laplacian matrix after adding the edge. Clearly, (γ... i -γ j ) 2 The condition ≥ 0 always holds true, meaning that the algebraic connectivity exhibits a monotonically increasing trend with the increase of communication links, and (γ) i -γ j ) 2 The value of γ is strongly positively correlated with the increase in connectivity of the descendant number of added edges, i.e., (γ) i -γ j ) 2 The larger the value of , the greater the increase in the algebraic connectivity of the Laplace matrix after adding an edge. Therefore, finding the maximum value of the eigenratio after each edge addition in equation (7) is equivalent to finding the value of (γ) after each edge addition. i -γ j ) 2 The maximum value of the ratio to the spectral radius, i.e., the basic idea of ​​the MGEA algorithm, is: in each iteration, the edge set E of the complement graph of graph G must be used. c From the remaining candidate edges, select one (γ) i -γ j ) 2 The edge with the largest ratio to the spectral radius maximizes the eigenvalue increment.

[0107] Step S5: Detect whether the number of communication links to be added has been reached. If so, determine the optimized communication topology based on the updated Laplace matrix. Otherwise, return to step S4 and continue searching for communication links that meet the requirements from the remaining edge set of the communication topology until the number of communication links to be added is reached, and determine the optimized communication topology based on the updated Laplace matrix.

[0108] In this embodiment, the MGEA algorithm flow is as follows: Figure 2 As shown, firstly, the Laplace matrix L0 of the original communication topology graph and the edge set E of the complement graph are given. c Then, set the number of communication links to be added k in the communication topology graph, and then retrieve a (γ) link from the remaining edge set. i -γ j ) 2 Find the edge with the largest ratio to the spectral radius and add that edge. Then update the Laplacian matrix L = L0 + ΔL. Next, check if the number of communication links to be added, k, has been reached. If so, obtain the optimized communication topology from the updated Laplacian matrix; otherwise, return to "retrieve an edge (γ) from the remaining complement edge set". i -γ j ) 2The process of "the edge with the largest ratio to the spectral radius" continues until the number of communication links to be increased, k, is reached, and the optimized communication topology is obtained from the updated Laplace matrix.

[0109] In this embodiment, to verify the performance of the MGEA algorithm's communication topology optimization strategy during simulation and performance analysis, the power system uses an IEEE 10-machine 39-bus system in the following four scenarios for simulation analysis, the structure of which is as follows: Figure 3 As shown, the corresponding original communication topology is as follows: Figure 4 As shown in Table 1, the 10 communication nodes G1-G10 represent 10 controllable distributed generation units, and the operating parameters of each generation unit are shown in Table 1. This embodiment uses an undirected connected graph for study, so these 10 nodes can transmit information bidirectionally through the communication link. After optimization using the MGEA algorithm, the trend of the eigenvalue ratio of the communication topology graph with increasing edge number is as follows: Figure 5 As shown.

[0110] analyze Figure 5 It can be seen that the eigenvalue ratio generally increases with the increase of the number of connected edges, but it shows a flat or slightly decreasing trend with the increase of a few communication links. Based on the... Figure 5 Based on the analysis and consideration of actual construction costs, this embodiment will increase the number of communication links k to 5 and limit the maximum degree of nodes to d. i Based on ≤5, the MGEA algorithm is used to optimize the original communication topology.

[0111] Table 1 Operating parameters of each power generation unit

[0112]

[0113]

[0114] Scenario 1: Validation of the effectiveness of the topology optimization strategy based on the MGEA algorithm.

[0115] To better highlight the MGEA algorithm's quantization of convergence rate λ2(L) and trigger interval λ max To balance these considerations, this scenario introduces an optimized topology that considers only the algebraic connectivity λ²(L) of the original topology and an optimized topology that considers only the spectral radius λ. max The optimized topologies are as follows: Figure 6 , Figure 7 As shown. The communication topology based on eigenratio optimization using the MGEA algorithm is as follows. Figure 8 As shown.

[0116] In this embodiment, the system load demand P is set. DThe system power differential adjustment coefficient ε is -0.0002, and the iteration interval of the event-triggered consensus algorithm is 0.01s. Simulation results of distributed economic scheduling based on the event-triggered consensus algorithm using four communication topologies are as follows: Figures 9 to 16 As shown.

[0117] Figure 9 , Figure 10 The convergence rate and the number of events triggered for the power system consistency variables under the original topology are displayed respectively. Figure 11 , Figure 12 , Figure 13 , Figure 14 The figures show the simulation results of the power system operation after topology optimization based on algebraic connectivity and spectral radius. Figure 15 and Figure 16 The simulation results show the operation of the power system after topology optimization using the MGEA algorithm.

[0118] Compare the distributed economic scheduling results of the event-triggered consensus algorithm under four different topologies. Figure 9 , Figure 11 , Figure 13 , Figure 15 The analysis reflects the convergence process of the system consistency variable with increasing iteration number under different topologies. It shows that the increase in communication links does not affect the final convergence value of the system consistency variable; that is, the system consistency variable λ based on the four topologies... i Ultimately, they all converged to the same value of $4.571 / MW; in comparison Figure 9 , Figure 10 and Figure 15 , Figure 16 As can be seen, compared with the original topology, the topology optimized by the eigenvalue ratio has a significant improvement in the convergence rate of the consistency variables and a significant reduction in the number of event triggers. Therefore, it can be seen that the MGEA algorithm is effective in improving the convergence speed of the system's consistency variables and saving communication resources.

[0119] contrast Figure 11 , Figure 12 , Figure 13 , Figure 14 and Figure 15 , Figure 16It can be seen that the algebraic connectivity optimization topology achieves the fastest convergence speed for the consistency variables, reaching consensus after 191 iterations, but has the highest number of event triggers (230). The spectral radius optimization topology has the slowest convergence speed, requiring 258 iterations, but has the fewest event triggers (199). These two optimization topologies only consider one of the two metrics: the convergence rate of the consistency variables or the number of event triggers. While the eigenratio-based optimization topology does not achieve the best convergence rate or the number of event triggers, it balances both metrics, achieving consensus after 232 iterations with 221 event triggers. Therefore, the eigenratio-based optimization topology considers system performance metrics more comprehensively, not only accelerating the convergence speed of the power system's consistency variables but also effectively conserving the power system's communication resources.

[0120] Scenario 2: Simulation analysis under changing load demand.

[0121] To verify the adaptability of the MGEA algorithm to the power system under changes in actual load demand after communication topology optimization, a system load demand P is set in this scenario. D The MW capacity decreased from 1600MW to 1400MW after 200 iterations, and then recovered to 1600MW after 350 iterations. The simulation results are as follows. Figures 17 to 20 As shown.

[0122] analyze Figure 17 , Figure 19 , Figure 20 It can be seen that when the system load demand drops to 1400MW, the power system no longer meets the supply and demand balance constraint. After each generation unit breaks the original state, it readjusts its own generation power. After a certain number of iterations, the consistency variable of each generation unit converges to a new value of 4.347$ / MW. When the system load demand recovers to 1600MW, each generation unit makes corresponding adjustments again, and the consistency variable converges to the initial value of 4.571$ / MW, thus satisfying the supply and demand balance of the power system again. Figure 18 This reflects that when load demand changes, the event-triggered mechanism increases the number of event triggers, thereby increasing the number of information exchanges between generation units and enabling the consistency variables to converge quickly. Therefore, it can be concluded that after optimizing the communication topology, the MGEA algorithm enables the power system to operate stably under load demand changes, demonstrating strong adaptability.

[0123] Scenario 3: Simulation analysis of "plug-and-play" power components.

[0124] Suppose that when the iteration number is 200, generator unit G1 exits the power system due to a fault, and then when the iteration number is 350, generator unit G1 is reconnected to the power system. The simulation results are as follows. Figures 21 to 24 As shown, when generator unit G1 exits, the remaining nine generator units readjust their power output, and the consistency variable converges to a new value of $4.848 / MW. When generator unit G1 rejoins the power system, the consistency variable converges back to its initial value of $4.571 / MW. Therefore, this verifies that after optimizing the communication topology, the MGEA algorithm demonstrates strong robustness of the power system under "plug-and-play" power component conditions.

[0125] It should be noted that, Figure 10 , Figure 12 , Figure 14 , Figure 16 , Figure 18 as well as Figure 22 The black rectangles in the upper right corner are for illustration purposes. The circles in the illustrations represent the event trigger times. i This represents the i-th power generation unit. The legend uses a small circle to represent the event trigger time for each power generation unit.

[0126] Scenario 4: Simulation analysis under communication link failure.

[0127] Suppose that when the number of iterations is 50, the communication links between power generation units G1 and G2, and G8 and G9 are disconnected due to interference factors such as noise and weather, and then return to normal when the number of iterations is 200. The system simulation results are as follows. Figure 25 and Figure 26 As shown. Observation Figure 25 and Figure 26 Analysis reveals that when the number of system iterations is between 50 and 200, the algebraic connectivity decreases due to communication link failures, leading to poor connectivity in the topology graph. Consequently, the consistency variable curves (cost incremental rate curves) of each generator converge slowly. However, when the number of iterations exceeds 200, the communication link recovers, the connectivity of the topology graph increases, and the consistency variable curves of each generator in the power system sharply converge towards the center. The output power of the generating units experiences slight fluctuations, and the convergence speed accelerates. At the 307th iteration, convergence is achieved, with a value identical to the convergence value of the four different topologies in Scenario 1, which is 4.571 $ / MW. Therefore, this verifies that after optimizing the communication topology, the MGEA algorithm still exhibits strong robustness in the face of communication link failures in the power system.

[0128] In one exemplary embodiment, a computer device is provided, which may be a server or a terminal. The computer device includes a processor, memory, input / output interfaces (I / O), and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is connected to the system bus via the I / O interfaces. The processor of the computer device provides computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The database of the computer device stores relevant data for distributed power dispatch communication topology optimization in a power system. The I / O interfaces of the computer device are used for exchanging information between the processor and external devices. The communication interface of the computer device is used for communication with external terminals via a network connection. When the computer program is executed by the processor, it implements the distributed power dispatch communication topology optimization method.

[0129] In one exemplary embodiment, a computer device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the distributed power dispatch communication topology optimization method described above.

[0130] In one exemplary embodiment, a computer-readable storage medium is provided storing a computer program that, when executed by a processor, implements the distributed power dispatch communication topology optimization method.

[0131] In one exemplary embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the distributed power dispatch communication topology optimization method.

[0132] This embodiment applies a multi-agent consensus algorithm to distributed economic dispatch in power systems, introducing a quantitative indicator feature ratio that balances convergence rate and communication resources. Based on this, a topology optimization algorithm is designed using an edge-adding approach to solve the topology optimization problem in distributed economic dispatch based on event-triggered consensus algorithms. Simulation analysis shows that the topology optimization algorithm designed in this embodiment can significantly improve the convergence rate of system consensus variables and save communication resources. Furthermore, in three scenarios—load demand changes, "plug-and-play" power components, and communication link failures—the power system with optimized topology exhibits strong adaptability and robustness. Therefore, the edge-adding topology optimization strategy based on a greedy algorithm proposed in this embodiment can effectively improve the performance of distributed economic dispatch systems and ensure the stability of power systems under distributed operation, making it suitable for future power system development.

[0133] With the increasing penetration of distributed energy resources, the economic dispatch of power systems is becoming increasingly complex. Multi-agent consensus algorithms, due to their superior distributed decision-making and coordination capabilities, have been widely applied in distributed economic dispatch scenarios. To alleviate the communication burden on the power system and improve the convergence rate of the consensus algorithm during distributed economic dispatch, this embodiment introduces an event-triggered mechanism. Theoretical analysis yields quantitative indicators affecting the triggering interval and convergence rate of the event-triggered consensus algorithm. A greedy algorithm is designed using an edge-adding topology optimization approach to improve these two indicators. Simulation results in various scenarios demonstrate that the power system with topology optimized by the greedy algorithm not only alleviates the communication burden but also achieves rapid convergence of the event-triggered consensus algorithm.

[0134] This embodiment utilizes a multi-agent consensus algorithm to design a distributed economic dispatch system for power systems with an undirected connected graph communication topology. By introducing the concept of eigenvalue ratio, it takes into account both the performance indicators of the power system in terms of communication resources and the convergence rate of the consensus algorithm. A topology optimization model is established and a topology optimization algorithm is designed by adding edges. The result of each iteration of this algorithm maximizes the current eigenvalue ratio of the topology graph, thereby improving the performance of the power system in terms of both the convergence rate of the consensus algorithm and communication resources, and ensuring the stable operation of the power system.

[0135] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0136] This embodiment uses specific examples to illustrate the principles and implementation methods of this application. The description of the above embodiments is only for the purpose of helping to understand the method and core ideas of this application; at the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this application. In summary, the content of this specification should not be construed as a limitation of this application.

Claims

1. A method for optimizing the communication topology of distributed power dispatching, characterized in that, The distributed power dispatch communication topology optimization method includes: Step S1: Using the basic knowledge of graph theory, model the distributed economic dispatch communication topology network of the power system to obtain the distributed economic dispatch communication topology network model of the power system; each node in the distributed economic dispatch communication topology network model of the power system represents a power generation unit, and each edge represents a communication link between two power generation units; Step S2: Determine the characteristic ratio based on the power system distributed economic dispatch communication topology network model; the characteristic ratio is the ratio of the algebraic connectivity to the spectral radius of the Laplace matrix corresponding to the power system distributed economic dispatch communication topology network model. Step S3: Based on the power system distributed economic dispatch communication topology network model and the characteristic ratio, establish a topology optimization model; Step S4: Based on the topology optimization model, select an edge from the edge set of the complement graph of the communication topology graph that maximizes the current feature ratio increment, add the edge to the communication topology graph, update the Laplacian matrix corresponding to the communication topology graph, and obtain the updated Laplacian matrix. Step S5: Detect whether the number of communication links to be added has been reached. If so, determine the optimized communication topology based on the updated Laplace matrix. Otherwise, return to step S4 and continue searching for communication links that meet the requirements from the remaining edge set of the communication topology until the number of communication links to be added is reached, and determine the optimized communication topology based on the updated Laplace matrix.

2. The distributed power dispatch communication topology optimization method according to claim 1, characterized in that, Step S1 specifically includes: Based on the fundamentals of graph theory, a distributed economic dispatch communication topology network for the power system is modeled. A graph G is defined to represent the communication topology of the power system, denoted as G = (V, E), where V = {v1, v2, ..., v...}. n } represents the set of nodes, indicating the power generation unit in the power system; E = {e ij =(v i ,v j Let )} be the set of edges, representing the communication links between the power generation units.

3. The distributed power dispatch communication topology optimization method according to claim 1, characterized in that, Step S2 specifically includes: The Laplace matrix is ​​determined based on the power system distributed economic dispatch communication topology network model. The algebraic connectivity and the spectral radius are determined based on the Laplace matrix. The eigenratio is calculated based on the algebraic connectivity and the spectral radius.

4. The distributed power dispatch communication topology optimization method according to claim 1, characterized in that, The Laplace matrix is ​​expressed by the following formula: Where L represents the Laplacian matrix, m is the total number of edges in the communication topology graph, and e is the edge set E of the complement graph. c The index of the candidate edge, h e Let T represent the edge vector connecting the i-th power generation unit and the j-th power generation unit, and let T represent the transpose operation.

5. The distributed power dispatch communication topology optimization method according to claim 1, characterized in that, The characteristic ratio is calculated using the following formula: Where ξ represents the eigenratio, λ²(L) is the algebraic connectivity, and λ max Let L be the spectral radius and L be the Laplace matrix.

6. The distributed power dispatch communication topology optimization method according to claim 1, characterized in that, The expression for the topology optimization model is: Where ξ represents the eigenratio, L0 is the Laplace matrix of the original communication topology, ΔL is the change in the Laplace matrix after adding ΔE, ΔE is the set of edges to be added to the original communication topology, and E c Let e ​​be the edge set of the complement graph, and E be the edge set of the complement graph. c The index of the candidate edge, k is the number of communication links to be added, A represents the adjacency matrix, b is an n-dimensional column vector of all ones, and d is the index of the candidate edge. i The resulting n-dimensional column vector, d i x is the maximum kilowatt-hour limit for the i-th generating unit in the power system. e h is a Boolean variable. e Let x represent the edge vector connecting the i-th power generation unit and the j-th power generation unit, where x is a vector of length |E. c | is a column vector, 1 is a vector of all 1s, and T represents the transpose operation.

7. The distributed power dispatch communication topology optimization method according to claim 1, characterized in that, The updated Laplace matrix is ​​expressed by the following formula: Where L(x) represents the updated Laplace matrix, L0 is the Laplace matrix of the original communication topology, and E c Let e ​​be the edge set of the complement graph, and E be the edge set of the complement graph. c The index of the candidate edge, x e h is a Boolean variable. e Let T represent the edge vector connecting the i-th power generation unit and the j-th power generation unit, and let T represent the transpose operation.

8. A computer device, comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that the processor executes the computer program to implement the distributed power dispatch communication topology optimization method according to any one of claims 1-7.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the distributed power dispatch communication topology optimization method as described in any one of claims 1-7.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the distributed power dispatch communication topology optimization method as described in any one of claims 1-7.

Images