A Distributed Economic Dispatch Method and System for Microgrids Based on Dynamic Feedback Broadcast Rumor Algorithm

Abstract

This invention belongs to the field of microgrid economic dispatch technology and discloses a microgrid distributed economic dispatch method and system based on a dynamic feedback broadcast rumor algorithm. This invention solves the economic dispatch problem of microgrids by designing an asynchronous distributed rumor algorithm with dynamic feedback gain. Through a dynamic adjustment mechanism, its convergence speed is significantly better than that of fixed-gain algorithms, thereby greatly shortening the time required for the system to reach its optimal operating state. It also avoids the contradiction between convergence speed and stability common in traditional methods. The method of this invention also employs rigorous theoretical analysis to establish sufficient conditions for algorithm convergence. By theoretically deriving the upper bound of the dynamic feedback gain, it ensures the convergence of the algorithm and the robustness of the system, significantly improving system operating efficiency and avoiding the blindness of repeated debugging in traditional trial-and-error methods. This greatly shortens the system debugging cycle, making the microgrid distributed economic dispatch method of this invention have superior convergence performance compared to traditional methods.

Description

Technical Field

[0001] This invention belongs to the field of microgrid economic dispatch technology, specifically relating to a microgrid distributed economic dispatch method and system based on dynamic feedback broadcast rumor algorithm. Background Technology

[0002] In recent years, microgrids have developed rapidly due to their flexibility, environmental friendliness, and high efficiency, making economic dispatch (ED) a key issue in the energy sector. With the increasing penetration of renewable energy and the growing complexity of load characteristics, microgrid economic dispatch faces higher requirements in terms of power supply reliability and economy, particularly in dynamic response and algorithm robustness. Existing microgrid economic dispatch methods still have significant shortcomings in adaptability and convergence under communication-constrained conditions, necessitating the development of more advanced economic dispatch technologies.

[0003] Microgrid economic dispatch methods can be divided into two categories: centralized and distributed. Centralized methods, such as Genetic Algorithm (GA) and Particle Swarm Optimization (PSO), rely on a central control unit and suffer from high cost, low reliability, and poor topology adaptability. Distributed methods, on the other hand, significantly improve system reliability and autonomy due to their advantages of strong scalability, high flexibility, and low communication costs. Fully distributed algorithms based on consensus achieve decentralized dispatch, but most of them rely on ideal communication assumptions and struggle to effectively address communication problems in real-world systems, such as data loss, latency, and random topology changes.

[0004] To address the challenges of communication randomness, gossip algorithms, due to their asynchronous communication characteristics, have been introduced into microgrid scheduling. Traditional gossip algorithms suffer from packet collisions, while broadcast gossip algorithms, based on the broadcast characteristics of wireless sensor networks, offer a more promising solution by reducing the number of transmissions and accelerating consensus convergence. However, the convergence performance of broadcast gossip algorithms is significantly affected by the dynamic characteristics of the network, often resulting in slow convergence. Slow convergence leads to several problems, such as decreased system dynamic response capability, difficulty in handling load fluctuations and fault scenarios; increased consumption of communication resources, creating a performance degradation cycle; impact on economic operation, with scheduling delays leading to increased costs and missed market opportunities; and even threatening power supply security in extreme cases. This severely limits the practical application of broadcast gossip algorithms.

[0005] Feedback gain parameters are crucial for balancing convergence speed and system robustness. High gain accelerates convergence but easily induces oscillations, while low gain promotes stability but slows convergence. Traditional fixed-gain strategies are ill-suited to dynamic scenarios, and obtaining the optimal gain requires extensive simulations, which is costly. Although dynamic gain adjustment has shown potential in distributed optimization, its application in the broadcast gossip algorithm still requires further exploration. Therefore, overcoming convergence performance bottlenecks and developing intelligent gain control methods have become key research directions for improving the economic dispatch performance of microgrids. Summary of the Invention

[0006] The purpose of this invention is to propose a distributed economic dispatch method for microgrids based on a dynamic feedback broadcast rumor algorithm. This method aims to solve the problems of slow convergence speed, poor dynamic adaptability, and complex parameter tuning of traditional broadcast rumor algorithms in distributed economic dispatch of microgrids. Through a dynamic feedback gain adjustment mechanism, it automatically optimizes the algorithm parameters according to the real-time operating status of the microgrid, which can significantly improve the optimization efficiency while ensuring convergence.

[0007] To achieve the above objectives, the present invention adopts the following technical solution:

[0008] A distributed economic dispatch method for microgrids based on a dynamic feedback broadcast rumor algorithm includes the following steps:

[0009] Step 1. Establish a microgrid economic dispatch problem model, which includes the generation cost function of the intelligent agent (generator) in the microgrid, the objective function of the economic dispatch problem, the supply and demand balance constraints of the generator output power, and the limiting constraints of the generator output power.

[0010] Step 2. Establish the relationship between cost increment, generator output power, and generation cost function; set initial values ​​for the output power of each generator, and calculate the initial value of cost increment using the initial values ​​of generator output power;

[0011] Step 3. Construct the Lagrange function using the Lagrange multiplier method for the microgrid economic dispatch problem model, solve for the optimal cost increment that minimizes the generation cost function, and then obtain the optimal output power of each generator;

[0012] Step 4. Select the generator cost increment and mismatch as consistency variables, and establish the consistency variable update relationship;

[0013] Step 5. Based on the theory of multi-agent consensus and stochastic approximation, the fixed feedback gain in the consistency variable update relation is improved, and an asynchronous distributed rumor algorithm with dynamic feedback gain is designed. The upper bound of the feedback gain that guarantees the convergence of the algorithm is derived.

[0014] Step 6. Design the consistency variable update iteration relationship of agents in the microgrid based on the asynchronous distributed rumor algorithm with dynamic feedback gain, so as to minimize the power generation cost function while satisfying the system supply and demand balance.

[0015] Furthermore, based on the microgrid distributed economic dispatch method based on the dynamic feedback broadcast rumor algorithm, this invention also proposes a microgrid distributed economic dispatch system based on the dynamic feedback broadcast rumor algorithm, the technical solution of which is as follows:

[0016] A distributed economic dispatch system for microgrids based on a dynamic feedback broadcast rumor algorithm includes:

[0017] The scheduling model establishment module is used to establish a microgrid economic scheduling problem model, which includes the power generation cost function of the intelligent agent (i.e., the generator) in the microgrid, the objective function of the economic scheduling problem, the supply and demand balance constraints of the generator output power, and the limiting constraints of the generator output power.

[0018] The cost increment establishment module is used to establish the relationship between cost increment, generator output power, and generation cost function; set the initial value of each generator output power, and calculate the initial value of cost increment based on the initial value of generator output power;

[0019] The optimal increment solution module is used to construct the Lagrange function for the microgrid economic dispatch problem model using the Lagrange multiplier method, solve for the optimal cost increment that minimizes the generation cost function, and then obtain the optimal output power of each generator.

[0020] The update relationship establishment module is used to select the generator cost increment and mismatch as consistency variables and establish the consistency variable update relationship;

[0021] The feedback gain improvement module is used to improve the fixed feedback gain in the consistency variable update relation based on the theory of multi-agent consensus and stochastic approximation. It designs an asynchronous distributed rumor algorithm with dynamic feedback gain and derives the upper bound of the feedback gain that guarantees the convergence of the algorithm.

[0022] And a variable update and iteration module, which is used to design the consistency variable update and iteration relationship of agents in the microgrid based on the asynchronous distributed rumor algorithm with dynamic feedback gain, so as to minimize the power generation cost function while satisfying the system supply and demand balance.

[0023] Furthermore, based on the above-mentioned microgrid distributed economic dispatch method based on dynamic feedback broadcast rumor algorithm, this invention also proposes a computer device, which includes a memory and one or more processors.

[0024] The memory stores executable code, and when the processor executes the executable code, it implements the steps of the microgrid distributed economic dispatch method based on the dynamic feedback broadcast rumor algorithm mentioned above.

[0025] Furthermore, based on the aforementioned microgrid distributed economic dispatch method based on the dynamic feedback broadcast rumor algorithm, this invention also proposes a computer-readable storage medium storing a program thereon; when executed by a processor, this program is used to implement the steps of the microgrid distributed economic dispatch method based on the dynamic feedback broadcast rumor algorithm described above.

[0026] The present invention has the following advantages:

[0027] As described above, this invention discloses a distributed economic dispatch method for microgrids based on a dynamic feedback broadcast gossip algorithm. It addresses the economic dispatch problem of microgrids by designing an asynchronous distributed gossip algorithm with dynamic feedback gain. The proposed dynamic adjustment mechanism significantly outperforms the fixed-gain algorithm in terms of convergence speed. This dynamic adjustment mechanism can drastically shorten the time required for the system to reach its optimal operating state, while avoiding the contradiction between convergence speed and stability common in traditional methods. Especially when renewable energy output changes rapidly, this invention can more quickly coordinate the various power generation units to reach their optimal operating points. Furthermore, this invention employs an asynchronous communication design, which significantly reduces the system's requirements for communication synchronization, making distributed optimization easier to implement in microgrids containing heterogeneous equipment. In addition, this invention establishes sufficient conditions for algorithm convergence through rigorous theoretical analysis. The upper bound of the dynamic feedback gain, determined through rigorous theoretical derivation, can significantly improve system operating efficiency, avoid the blindness of repeated debugging in traditional trial-and-error methods, greatly shorten the system debugging cycle, and ensure algorithm convergence and system robustness. Therefore, this invention's method has superior convergence performance compared to traditional methods. Especially for large-scale microgrid systems, the method of this invention can effectively solve the problem of increased algorithm coordination complexity caused by the expansion of system scale, and ensure that the distributed algorithm can still maintain stable convergence performance in complex network environments. It is particularly suitable for application in modern microgrids with a high proportion of renewable energy, diverse equipment types, and complex communication conditions, and can provide a more effective solution to the distributed economic dispatch problem of microgrids. Attached Figure Description

[0028] Figure 1 This is a flowchart of the microgrid distributed economic dispatch method based on the dynamic feedback broadcast rumor algorithm of the present invention.

[0029] Figure 2 This is a topology diagram of the microgrid in an embodiment of the present invention.

[0030] Figure 3 This is a schematic diagram illustrating the convergence result of the cost increment in an embodiment of the present invention.

[0031] Figure 4 This is a schematic diagram of the convergence result of the output power in an embodiment of the present invention.

[0032] Figure 5 This is a schematic diagram illustrating the convergence result of power mismatch, i.e., the amount of mismatch, in an embodiment of the present invention.

[0033] Figure 6 This is a schematic diagram of the convergence result of supply and demand balance in an embodiment of the present invention.

[0034] Figure 7 This diagram illustrates the convergence time of distributed economic dispatching of microgrids using the traditional broadcast rumor algorithm.

[0035] Figure 8 This diagram illustrates the convergence time of distributed economic dispatching of microgrids using the method of this invention. Detailed Implementation

[0036] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments:

[0037] Example 1

[0038] This invention aims to address key issues in traditional broadcast rumor algorithms for distributed economic dispatching of microgrids, such as slow convergence speed, poor dynamic adaptability, and complex parameter tuning. By innovatively proposing a dynamic feedback gain adjustment mechanism, the algorithm parameters can be automatically optimized based on the real-time operating status of the microgrid, significantly improving optimization efficiency while ensuring convergence.

[0039] Specifically, this invention first establishes a theoretical upper bound for the convergence of the feedback gain based on the system matrix spectral radius analysis, providing a mathematical basis for parameter selection and avoiding the enormous workload associated with traditional trial-and-error methods. Secondly, it designs a strategy for dynamically adjusting the feedback gain based on local information, using the difference between the node's output power and the demand as the feedback quantity. By employing local information, the algorithm can adapt to dynamic scenarios such as network topology changes and power fluctuations, accelerating convergence and enhancing system robustness. Furthermore, this invention optimizes the broadcast communication mechanism, reducing redundant information transmission and thus lowering communication resource consumption.

[0040] Compared with existing technologies, this invention can significantly improve the real-time performance, stability and economy of microgrid economic dispatch, while avoiding the risk of frequency overruns or power imbalances caused by convergence delays, providing an efficient and reliable distributed optimization solution for microgrids with a high proportion of renewable energy access.

[0041] like Figure 1 As shown, the distributed economic dispatch method for microgrids based on the dynamic feedback broadcast rumor algorithm includes the following steps:

[0042] Step 1. Establish a microgrid economic dispatch problem model, which includes the power generation cost function of the intelligent agent (i.e., the generator) in the microgrid, the objective function of the economic dispatch problem, the supply and demand balance constraints of the generator output power, and the limiting constraints of the generator output power.

[0043] The generation cost function of generators in a microgrid is as follows:

[0044] .

[0045] Where i = 1, 2, ..., n, and n is the number of generator nodes in the microgrid. i (P i Let P be the power generation cost function of the i-th generator. i Let a be the output power of the i-th generator. i b i c i Let a be the power generation cost coefficient of the i-th generator. i >0, b i >0, c i >0.

[0046] The objective function of the economic dispatch problem, the supply and demand balance constraints of the generator output power, and the limiting constraints of the generator output power are as follows:

[0047] .

[0048] in, Let be the objective function of the economic scheduling problem. For the supply and demand balance constraint of the generator output power, P D This represents the total power requirement. These are the limiting constraints on the generator's output power. This is the minimum output power limit for the i-th generator. This is the maximum output power limit for the i-th generator.

[0049] Each generator node i satisfies its own maximum output power limit. and minimum output power limit This simultaneously satisfies the supply and demand balance on both the generation and demand sides, meaning the total system output power equals the load power, thus minimizing the total cost.

[0050] Step 2. Establish the relationship between cost increment, generator output power, and the generation cost function. Set initial values ​​for the output power of each generator, and calculate the initial value for cost increment using these initial generator output power values.

[0051] Cost increment refers to the change in total system cost when increasing power generation per unit during the optimized scheduling process. The objective function in step 1, i.e., the power generation cost function C... i (P i The minimization problem uses cost increment λ. i To describe it.

[0052] The cost increment is represented by the power generation cost function in step 1. For generator output power P i Find the derivative to obtain the generator output power P. i With cost increment λ i The relation is: .

[0053] Set the initial values ​​for the required parameters, with the initial output power of the i-th generator being P. i (0), let .

[0054] The initial value of generator output power P i (0) Calculate the initial value of the generator cost increment. The initial cost increment of the i-th generator is _____. .

[0055] Step 3. Construct the Lagrange function for the microgrid economic dispatch problem model using the Lagrange multiplier method, solve for the optimal cost increment that minimizes the generation cost function, and then obtain the optimal output power of each generator.

[0056] The Lagrange multiplier method is applied to the microgrid economic dispatch problem model to obtain the Lagrange function. for:

[0057] .

[0058] Where λ is the Lagrange multiplier, which is the cost increment in the microgrid economic dispatch problem.

[0059] Let the Lagrange function be applied to P i Taking the partial derivatives of λ and λ respectively, the optimal cost increment is obtained using the Lagrange multiplier method. for:

[0060] .

[0061] Based on the generator output power P i With cost increment λ i Relationship To obtain the optimal output power for:

[0062] .

[0063] Step 4. Select the generator cost increment and mismatch as consistency variables and establish the consistency variable update relationship.

[0064] Selecting the generator cost increment and mismatch (i.e., local power mismatch) as consistency variables, the consistency variable update relationship is established as follows:

[0065] .

[0066] .

[0067] Where j = 1, 2, ..., n. λ j (k) represents the cost increment of the j-th generator at time k, λ i (k) represents the cost increment of the i-th generator at time k, λ i (k+1) represents the cost increment of the i-th generator at time k+1. j (k) represents the mismatch of the j-th generator at time k, y i (k) represents the mismatch of the i-th generator at time k, y i (k+1) represents the mismatch of the i-th generator at time k+1. For fixed feedback gain. i (k) represents the generator output power of the i-th generator at time k, p i (k+1) represents the generator output power of the i-th generator at time k+1. , , This represents the adjacency weight between the i-th generator and the j-th generator.

[0068] , , Each of the following conditions must be met:

[0069] .

[0070] .

[0071] .

[0072] in, Let represent the edge set. Define the i-th generator as node i and the j-th generator as node j. This indicates that there is a direct connection between node i and node j, meaning that node i and node j are neighboring nodes. This indicates that there is no direct connection between node i and node j.

[0073] Treating the i-th generator as its own node and the j-th generator as its neighbor node, the cost increment update is achieved by weighted averaging of the cost increments of the own node and its neighbor nodes, in order to drive the cost increment λ of all generator nodes. i Convergence.

[0074] By incorporating the local power mismatch, or misalignment, as a feedback value into the cost increment update formula, the accumulated local power imbalance, such as the difference between load and generation, plays a regulatory role in the cost increment, thereby driving the cost increment λ. i Adjustment.

[0075] Step 5. Based on the theory of multi-agent consensus and stochastic approximation, the fixed feedback gain in the consistency variable update relation is improved, and an asynchronous distributed rumor algorithm with dynamic feedback gain, namely the dynamic feedback consensus economic scheduling algorithm, is designed. The upper bound of the feedback gain that guarantees the convergence of the algorithm is derived.

[0076] Traditional distributed methods typically employ fixed feedback gain, which, while guaranteeing convergence under static, ideal network conditions, struggles to handle dynamically changing scenarios. Furthermore, the optimal feedback gain for ensuring algorithm convergence is often difficult to calculate precisely, requiring extensive simulation data experiments to obtain an approximate solution. This leads to a significant increase in scheduling workload and resource consumption.

[0077] In step 5 of this embodiment, based on multi-agent consensus and stochastic approximation theory, the cost increment update relation obtained in step 4 is updated. Fixed feedback gain Improvements were made to design an asynchronous distributed rumor algorithm with dynamic feedback gain.

[0078] Applying dynamic feedback gain function to the microgrid economic dispatch problem model for:

[0079] .

[0080] in, This indicates the initial gain, which is set according to the system's maximum allowable adjustment range. The decay rate is represented by parameters that need to be tuned through simulation or Lyapunov analysis to balance convergence speed and stability.

[0081] This represents the initial value of the feedback gain; a larger initial value indicates a larger initial value. It can enhance the mismatch amount y i The feedback effect of (k) forces the cost increment λ i It rapidly approximates a uniform value; as k increases, The correction will gradually decrease to avoid oscillations caused by excessive corrections that make the algorithm difficult to converge.

[0082] The initial value of the feedback gain should be selected within a reasonable range for the dynamic feedback consensus economic scheduling algorithm to achieve asymptotic convergence. An excessively large initial gain can cause system oscillations, making rapid convergence difficult or even leading to divergence. Therefore, the next step is to define the upper bound of the feedback gain, which guarantees the convergence of the asynchronous distributed rumor algorithm with dynamic feedback gain. Perform a systematic solution:

[0083] Based on the system matrix spectral radius analysis, and according to the eigenvalue perturbation theory and the system matrix spectral distance constraint, we obtain:

[0084] .

[0085] in, This represents the system matrix before the disturbance. This represents the system matrix after the disturbance. , Here is the perturbation matrix. Representation matrix The spectrum is the set of eigenvalues ​​of the matrix before the perturbation. Representation matrix The spectrum is the set of eigenvalues ​​of the perturbated matrix. The Hausdorff distance represents the set of eigenvalues ​​of the matrix before and after the perturbation. Representation matrix The infinite norm, Representation matrix The infinite norm of .

[0086] The initial gain is derived. The maximum allowable adjustment range of the system, i.e., the upper bound of the feedback gain that guarantees algorithm convergence. for:

[0087] .

[0088] in, , to These represent the cost coefficients of the 1st to the nth generator nodes, respectively. For matrix The third largest eigenvalue in the modulus sense.

[0089] Step 6. Design the consistency variable update iteration relationship of agents in the microgrid based on the asynchronous distributed rumor algorithm with dynamic feedback gain, so as to minimize the power generation cost function while satisfying the system supply and demand balance.

[0090] The broadcast rumor algorithm leverages the inherent broadcasting advantage of wireless sensors in smart grids, enabling all neighboring nodes of a node to update information in each iteration.

[0091] At time k+1, node j is activated as the sending node. Based on the asynchronous distributed rumor algorithm with dynamic feedback gain, the consistency variable update iteration relationship of its neighbor nodes is designed, that is, the state update expression of node i is expressed as:

[0092] .

[0093] in, and For cost parameters, , .

[0094] Set the initial value before the update, using the initial value P set in step 2. i (0) and as well as ,in Let represent the initial value of the mismatch of the i-th generator. The feedback gain is selected based on the upper bound of the gain obtained in step 5, and then the above formula is iteratively updated.

[0095] Based on the broadcast characteristics, at time k, node j is randomly activated as a sending node set j, and its self-update formula is:

[0096] .

[0097] Where, λ j (k+1) represents the cost increment of the j-th generator at time k+1, y j (k+1) represents the mismatch of the j-th generator at time k+1.

[0098] All other nodes At time k+1, the cost increment and mismatch amount from the previous time step remain unchanged:

[0099] .

[0100] in, This represents the set of nodes that have a direct connection to node j.

[0101] Enter the corresponding cost parameters and The algorithm iteratively updates the cost increment and mismatch, and compares the cost increment generated at each time step with the optimal cost increment calculated in step 3. Compare them.

[0102] If the two are the same, that is, the cost increment generated at the current moment is the same as the optimal cost increment. If they are equal, the consistency variable is determined to have converged to the optimal solution, at which point the generator cost function value in the microgrid is minimized. If they are different, that is, the cost increment generated at the current moment differs from the optimal cost increment... If they are not equal, the cost increment and mismatch amount will continue to be iteratively updated until the consistency variables converge to the optimal solution.

[0103] Compared to traditional broadcast rumor algorithms, the broadcast rumor-based distributed microgrid economic dispatch method proposed in this invention has better convergence performance and can enable the system state to reach consensus more quickly.

[0104] This embodiment uses, as follows: Figure 2 The microgrid topology connection method shown is used to verify the method of the present invention through simulation experiments using the data shown in Table 1.

[0105] Figure 2 In the five-bus system shown, DG1, DG2, DG3, DG4, and DG5 represent the first generator node, the second generator node, the third generator node, the fourth generator node, and the fifth generator node, respectively; Bus1, Bus2, Bus3, Bus4, and Bus5 represent the first bus, the second bus, the third bus, the fourth bus, and the fifth bus, respectively; and Load1, Load2, Load3, Load4, and Load5 represent the first load, the second load, the third load, the fourth load, and the fifth load, respectively.

[0106] Table 1. Parameter Selection Diagram for a Five-Bus System in a Microgrid

[0107]

[0108] Figure 3 This represents the convergence result of the cost increment at each generator node, based on... Figure 3 As can be seen, the experimental results are the same as the optimal cost increment calculated in step 3, and the method of the present invention can achieve accurate convergence. Figure 4 This indicates the convergence status of the output power of each generator node, based on... Figure 4 It can be seen that the method of the present invention can enable each generator node to accurately converge to the optimal output power. Figure 5 This indicates the convergence result of the mismatch, based on... Figure 5 It can be seen that the result obtained by using the method of the present invention converges to 0, that is, the output power of each node is equal to the load demand. Figure 6 This represents the result of the total output power and load demand, based on... Figure 6 It can be seen that the method of the present invention can make the two the same, that is, the system achieves supply and demand balance.

[0109] Figure 7 and Figure 8These figures illustrate the convergence speeds of distributed economic dispatching of microgrids using the traditional broadcast rumor algorithm and the method proposed in this invention, respectively. In this embodiment, distributed economic dispatching of microgrids is performed using both the traditional broadcast rumor algorithm and the method proposed in this invention, and the first iteration satisfying the termination condition is recorded. The convergence time is no more than 0.001. That is, the process is repeated 1000 times, and 1000 data points that meet the termination condition are obtained in each of the two algorithms. Figure 7 and Figure 8 Each point in the graph represents the number of convergence iterations in one run, primarily illustrating the distribution of convergence speed. This numerical result is calculated based on the specific values ​​from 1000 runs. Based on the experimental data, the following results can be obtained: the average number of convergence iterations for the method of this invention is 112, while the average number of convergence iterations for the traditional broadcast rumor algorithm is 151. In terms of average convergence iterations, the method of this invention is also superior to the traditional broadcast rumor algorithm. The probability of obtaining a poor convergence speed is lower in the method of this invention than in the traditional broadcast rumor algorithm, and the convergence times of the method of this invention are more concentrated. Therefore, the results show that the method of this invention can effectively improve the convergence speed.

[0110] The microgrid distributed economic dispatch method based on the dynamic feedback broadcast rumor algorithm proposed in this invention has achieved several key technological breakthroughs in the field of microgrid economic dispatch. This invention's method significantly improves convergence speed through an innovative dynamic feedback gain mechanism. Its adaptive adjustment characteristic can dynamically adjust parameters according to the real-time operating status of the system, allowing the algorithm to use a larger gain value in the initial stage to accelerate the convergence process, and automatically reduce the gain to ensure stability as it approaches the equilibrium point. The dynamic feedback gain mechanism greatly shortens the time required for the system to reach its optimal operating state, while avoiding the contradiction between convergence speed and stability common in traditional methods.

[0111] In terms of operational efficiency, the method of this invention reduces unnecessary computation and communication overhead by optimizing the information exchange mechanism. Each distributed unit only needs to perform necessary information exchanges with neighboring nodes to complete global optimization, significantly improving the utilization rate of computing resources. Particularly in terms of robustness, the asynchronous communication architecture adopted in this invention performs economic scheduling through an asynchronous broadcast rumor algorithm. This fully considers the performance differences of equipment in actual microgrids. By eliminating the dependence on the global clock, the system can tolerate communication delays or interruptions of some nodes, ensuring that core scheduling functions can still be maintained under various abnormal operating conditions. This innovative design, which integrates fast convergence, efficient computation, precise control, and strong robustness, provides a high-performance economic scheduling solution for microgrids with a high proportion of renewable energy, and effectively overcomes the technical limitations of traditional methods in multi-objective optimization.

[0112] Furthermore, the dynamic feedback gain upper bound determined by rigorous theoretical derivation in this invention provides key technical support for the engineering application of distributed economic dispatch algorithms for microgrids, which can significantly improve system operating efficiency. The gain parameter range precisely defined by mathematical analysis provides clear guidance for parameter setting, avoiding the blindness of repeated debugging in traditional trial and error methods and greatly shortening the system debugging cycle.

[0113] In terms of energy consumption optimization, the rigorously proven upper bound of the gain ensures the stable convergence of the algorithm under various operating conditions, effectively preventing system oscillations or divergences caused by parameter mismatch, thereby significantly reducing unnecessary energy consumption and equipment wear.

[0114] Meanwhile, the theoretically guided parameter tuning method significantly reduces the debugging and testing steps required by traditional methods, making the system deployment process more streamlined and efficient. Especially for large-scale microgrid systems, this invention effectively solves the problem of increased algorithm coordination complexity caused by system scaling, ensuring that distributed algorithms maintain stable convergence performance in complex network environments, and providing a unified and reliable technical solution for the economic dispatch of microgrids of different sizes.

[0115] Example 2

[0116] This embodiment 2 describes a microgrid distributed economic dispatch system based on the dynamic feedback broadcast rumor algorithm. This system is based on the same inventive concept as the microgrid distributed economic dispatch method based on the dynamic feedback broadcast rumor algorithm in embodiment 1.

[0117] Specifically, the microgrid distributed economic dispatch system based on the dynamic feedback broadcast rumor algorithm includes the following modules:

[0118] The scheduling model establishment module is used to establish a model of the economic scheduling problem of microgrids. It includes the power generation cost function of the intelligent agent (i.e., the generator) in the microgrid, the objective function of the economic scheduling problem, the supply and demand balance constraints of the generator output power, and the limiting constraints of the generator output power.

[0119] The cost increment establishment module is used to establish the relationship between cost increment, generator output power, and the generation cost function. It sets initial values ​​for the output power of each generator and calculates initial values ​​for cost increment based on these initial values.

[0120] The optimal increment solution module is used to construct the Lagrange function for the microgrid economic dispatch problem model using the Lagrange multiplier method, solve for the optimal cost increment that minimizes the generation cost function, and then obtain the optimal output power of each generator.

[0121] The update relationship establishment module is used to select the generator cost increment and mismatch as consistency variables and establish the consistency variable update relationship.

[0122] The feedback gain improvement module is used to improve the fixed feedback gain in the consistency variable update relation based on multi-agent consensus and stochastic approximation theory, design an asynchronous distributed rumor algorithm with dynamic feedback gain, and derive the upper bound of the feedback gain that guarantees the convergence of the algorithm.

[0123] And a variable update and iteration module, which is used to design the consistency variable update and iteration relationship of agents in the microgrid based on the asynchronous distributed rumor algorithm with dynamic feedback gain, so as to minimize the power generation cost function while satisfying the system supply and demand balance.

[0124] It should be noted that in the microgrid distributed economic dispatch system based on the dynamic feedback broadcast rumor algorithm, the implementation process of the functions and roles of each functional module is detailed in the corresponding steps of the method in Example 1, and will not be repeated here.

[0125] Example 3

[0126] This embodiment 3 describes a computer device that includes a memory and one or more processors.

[0127] The memory stores executable code, which, when executed by the processor, is used to implement the steps of the microgrid distributed economic dispatch method based on the dynamic feedback broadcast rumor algorithm in Embodiment 1 above.

[0128] In this embodiment, the computer device can be any device or apparatus with data processing capabilities, and will not be described in detail here.

[0129] Example 4

[0130] This embodiment 4 describes a computer-readable storage medium storing a program that, when executed by a processor, implements the steps of a microgrid distributed economic dispatch method based on a dynamic feedback broadcast rumor algorithm.

[0131] The computer-readable storage medium can be an internal storage unit of any device or apparatus with data processing capabilities, such as a hard disk or memory, or an external storage device of any device with data processing capabilities, such as a plug-in hard disk, smart media card (SMC), SD card, flash card, etc.

[0132] Of course, the above description is only a preferred embodiment of the present invention. The present invention is not limited to the above-described embodiments. It should be noted that any equivalent substitutions or obvious modifications made by those skilled in the art under the guidance of this specification fall within the scope of this specification and should be protected by the present invention.

Claims

1. A distributed economic dispatch method for microgrids based on a dynamic feedback broadcast rumor algorithm, characterized in that, Includes the following steps: Step 1. Establish a microgrid economic dispatch problem model, which includes the generation cost function of the intelligent agent (generator) in the microgrid, the objective function of the economic dispatch problem, the supply and demand balance constraints of the generator output power, and the limiting constraints of the generator output power. Step 2. Establish the relationship between cost increment, generator output power, and generation cost function; set initial values ​​for the output power of each generator, and calculate the initial value of cost increment using the initial values ​​of generator output power; Step 3. Construct the Lagrange function using the Lagrange multiplier method for the microgrid economic dispatch problem model, solve for the optimal cost increment that minimizes the generation cost function, and then obtain the optimal output power of each generator; Step 4. Select the generator cost increment and mismatch as consistency variables and establish the consistency variable update relationship; Step 5. Based on the theory of multi-agent consensus and stochastic approximation, the fixed feedback gain in the consistency variable update relation is improved, and an asynchronous distributed rumor algorithm with dynamic feedback gain is designed. The upper bound of the feedback gain that guarantees the convergence of the algorithm is derived. Step 6. Design the consistency variable update iteration relationship of the agents in the microgrid based on the asynchronous distributed rumor algorithm with dynamic feedback gain, so as to minimize the power generation cost function while satisfying the system supply and demand balance. Step 4 specifically involves: Selecting the generator cost increment and mismatch (i.e., local power mismatch) as consistency variables, the consistency variable update relationship is established as follows: ； ； Where i = 1, 2, ..., n; n is the number of generator nodes in the microgrid; j = 1, 2, ..., n; λ j (k) represents the cost increment of the j-th generator at time k, λ i (k) represents the cost increment of the i-th generator at time k, λ i (k+1) represents the cost increment of the i-th generator at time k+1; y j (k) represents the mismatch of the j-th generator at time k, y i (k) represents the mismatch of the i-th generator at time k, y i (k+1) represents the mismatch of the i-th generator at time k+1; For fixed feedback gain; p i (k) represents the generator output power of the i-th generator at time k, p i (k+1) represents the generator output power of the i-th generator at time k+1; , , This represents the adjacency weight between the i-th generator and the j-th generator; , , Each of the following conditions must be met: ； ； ； in, Let represent the edge set; define the i-th generator as node i and the j-th generator as node j; This indicates that there is a direct connection between node i and node j, meaning that node i and node j are neighboring nodes. This indicates that there is no direct connection between node i and node j.

2. The microgrid distributed economic dispatch method based on dynamic feedback broadcast rumor algorithm according to claim 1, characterized in that, Step 1 specifically involves: The generation cost function of generators in a microgrid is as follows: ； Among them, C i (P i Let P be the power generation cost function of the i-th generator; i Let a be the output power of the i-th generator; i b i c i Let a be the power generation cost coefficient of the i-th generator. i >0, b i >0, c i >0; The objective function, the supply-demand balance constraint, and the limiting constraint on generator output power for the economic dispatch problem are as follows: ； in, Let the objective function be the economic scheduling problem. For the supply and demand balance constraint of the generator output power, P D Total power requirement; These are the limiting constraints on the generator's output power. This is the minimum output power limit for the i-th generator. This is the maximum output power limit for the i-th generator.

3. The microgrid distributed economic dispatch method based on dynamic feedback broadcast rumor algorithm according to claim 2, characterized in that, Step 2 specifically involves: The cost increment is a function of the power generation cost. For generator output power P i Find the derivative; Obtain the generator output power P i With cost increment λ i The relation is: ; The initial output power of the i-th generator is set to P. i (0), let ; The initial cost increment of the generator is calculated by using the initial value of the generator output power, thus obtaining the initial cost increment of the i-th generator. for: .

4. The microgrid distributed economic dispatch method based on dynamic feedback broadcast rumor algorithm according to claim 3, characterized in that, Step 3 specifically involves: The Lagrange multiplier method is applied to the microgrid economic dispatch problem model to obtain the Lagrange function. for: ； Where λ is the Lagrange multiplier, which is the cost increment in the microgrid economic dispatch problem; Let the Lagrange function be applied to P i Taking the partial derivatives of λ and λ respectively, the optimal cost increment is obtained using the Lagrange multiplier method. for: ； Based on the generator output power P i With cost increment λ i relational expression To obtain the optimal output power for: 。 5. The microgrid distributed economic dispatch method based on dynamic feedback broadcast rumor algorithm according to claim 4, characterized in that, Step 5 specifically involves: Based on multi-agent consensus and stochastic approximation theory, the cost increment update relation obtained in step 4 is... Fixed feedback gain Improvements were made to design an asynchronous distributed rumor algorithm with dynamic feedback gain; Applying dynamic feedback gain function to the microgrid economic dispatch problem model for: ； in, Indicates the initial gain; Indicates the decay rate; The upper bound of the convergence of the feedback gain is the upper bound of the feedback gain that guarantees the convergence of the asynchronous distributed rumor algorithm with dynamic feedback gain. Perform a systematic solution; Based on the eigenvalue perturbation theory and the system matrix spectral distance constraint, we obtain: ； in, This represents the system matrix before the disturbance. This represents the system matrix after the disturbance. , Here is the perturbation matrix. Representation matrix The spectrum is the set of eigenvalues ​​of the matrix before the perturbation. Representation matrix The spectrum is the set of eigenvalues ​​of the perturbated matrix. The Hausdorff distance represents the set of eigenvalues ​​of the matrix before and after the perturbation. Representation matrix The infinite norm, Representation matrix The infinite norm; The initial gain is derived. The maximum allowable adjustment range of the system, i.e., the upper bound of the feedback gain that guarantees algorithm convergence. for: ； in, , to These represent the cost coefficients of the 1st to the nth generator nodes, respectively. For matrix The third largest eigenvalue in the modulus sense.

6. The microgrid distributed economic dispatch method based on dynamic feedback broadcast rumor algorithm according to claim 5, characterized in that, Step 6 specifically involves: At time k+1, node j is activated as the sending node. Based on the asynchronous distributed rumor algorithm with dynamic feedback gain, the consistency variable update iteration relationship of its neighbor nodes is designed, that is, the state update expression of node i is expressed as: ； in, and For cost parameters, , ; Set the initial value before the update, using the initial value P set in step 2. i (0) and as well as ,in This represents the initial value of the mismatch of the i-th generator, and the feedback gain is selected based on the upper bound of the gain obtained in step 5. Based on the broadcast characteristics, at time k, node j is randomly activated as a sending node set j, and its self-update formula is: ； Where, λ j (k+1) represents the cost increment of the j-th generator at time k+1, y j (k+1) represents the mismatch of the j-th generator at time k+1; All other nodes At time k+1, the cost increment and mismatch amount from the previous time step remain unchanged: ； in, This represents the set of nodes that have a direct connection to node j. Input cost parameters and The cost increment and mismatch are iteratively updated, and the cost increment generated at each time step is compared with the optimal cost increment calculated in step 3. Compare; If the cost increment generated at the current moment is equal to the optimal cost increment, then the consistency variable is determined to have converged to the optimal solution, at which point the power generation cost function value of the generator in the microgrid is minimized; If the cost increment generated at the current moment is not equal to the optimal cost increment, then continue to iteratively update the cost increment and mismatch until the consistency variables converge to the optimal solution.

7. A microgrid distributed economic dispatch system based on the dynamic feedback broadcast rumor algorithm for implementing the microgrid distributed economic dispatch method based on the dynamic feedback broadcast rumor algorithm as described in claim 1, characterized in that, The microgrid distributed economic dispatch system based on the dynamic feedback broadcast rumor algorithm includes: The scheduling model establishment module is used to establish a microgrid economic scheduling problem model, which includes the power generation cost function of the intelligent agent (i.e., the generator) in the microgrid, the objective function of the economic scheduling problem, the supply and demand balance constraints of the generator output power, and the limiting constraints of the generator output power. The cost increment establishment module is used to establish the relationship between cost increment, generator output power, and generation cost function; set the initial value of each generator output power, and calculate the initial value of cost increment based on the initial value of generator output power; The optimal increment solution module is used to construct the Lagrange function for the microgrid economic dispatch problem model using the Lagrange multiplier method, solve for the optimal cost increment that minimizes the generation cost function, and then obtain the optimal output power of each generator. The update relationship establishment module is used to select the generator cost increment and mismatch as consistency variables and establish the consistency variable update relationship; The feedback gain improvement module is used to improve the fixed feedback gain in the consistency variable update relation based on the theory of multi-agent consensus and stochastic approximation. It designs an asynchronous distributed rumor algorithm with dynamic feedback gain and derives the upper bound of the feedback gain that guarantees the convergence of the algorithm. And a variable update and iteration module, which is used to design the consistency variable update and iteration relationship of agents in the microgrid based on the asynchronous distributed rumor algorithm with dynamic feedback gain, so as to minimize the power generation cost function while satisfying the system supply and demand balance.

8. A computer device comprising a memory and one or more processors, wherein the memory stores executable code, characterized in that, When the processor executes the executable code, it implements the steps of the microgrid distributed economic dispatch method based on the dynamic feedback broadcast rumor algorithm as described in any one of claims 1 to 6.

9. A computer-readable storage medium having a program stored thereon, characterized in that, When the program is executed by the processor, it implements the steps of the microgrid distributed economic dispatch method based on the dynamic feedback broadcast rumor algorithm as described in any one of claims 1 to 6.

Images