Virtual power plant scheduling method and system based on cloud side-end architecture, medium and equipment

Abstract

The invention discloses a virtual power plant scheduling method, system, medium and equipment based on a cloud side-end architecture, and relates to the technical field of virtual power plant scheduling, the method comprises the following steps: establishing a virtual power plant information-physical system model, a physical layer being equipment of a virtual power plant, each equipment corresponding to an intelligent agent, a communication network between the intelligent agents forming an information layer, constructing a multi-agent system; the virtual power plant is divided into a plurality of areas according to a multi-agent topological graph, each area is controlled by an edge server, all the edge servers are connected to a cloud server, and a cloud edge-end collaborative architecture of the multi-agent system is constructed; according to the equipment model of the virtual power plant, setting to meet the supply and demand balance and the power limitation of each equipment, and minimizing the total expected operation cost as an optimization target; and according to the optimization target, performing virtual power plant scheduling by using a fixed time consistent optimization algorithm based on the cloud edge-end collaborative architecture of the multi-agent system. The scheme is high in reliability, and the cost can be effectively reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of virtual power plant scheduling, and in particular to a virtual power plant scheduling method, system, medium, and device based on a cloud-edge-end architecture. Background Art

[0002] With the increase in the proportion of renewable energy, the volatility and randomness problems of high-proportion renewable energy output pose key challenges to the stable operation of the power system. At the same time, energy storage devices, flexible loads, and distributed energy power plants at the distribution network level exacerbate the uncertainty of power consumption on the demand side. Power scheduling is an important tool for improving the integration of renewable energy into the grid and the stable operation of the power system. To more effectively adapt to renewable energy, the concept of a virtual power plant has been proposed, whose main function is to participate in grid scheduling as a whole.

[0003] Traditional virtual power plant economic scheduling problems are usually solved using a centralized control framework. The characteristic of the centralized algorithm is that it includes a central controller that collects information from all sensors and performs calculations, and then feeds back corresponding commands to each local actuator. The centralized algorithm is easy to implement and convenient to control, but there are still some problems: (1) The control center controls the entire virtual power plant. Once a failure occurs, it will cause the control of the entire system to fail, which not only poses a major hidden danger to the stable operation of the virtual power plant but also has low system reliability. (2) As the number of distributed energy sources increases, the information that the control center needs to collect and process gradually increases, resulting in a significant reduction in the operating efficiency of the microgrid. At the same time, the gradual increase in two-way communication links leads to a high construction cost of the communication network. (3) The centralized algorithm needs to collect information from all units, resulting in less privacy for individual entities. Summary of the Invention

[0004] The purpose of the present invention is to propose a virtual power plant scheduling method based on a cloud-edge-end architecture to solve the problems of low reliability and high cost of using a centralized control framework for power plant economic scheduling, including the following steps:

[0005] S1. Establish an information-physical system model of the virtual power plant. The physical layer of the information-physical system is the equipment of the virtual power plant. Each equipment corresponds to an agent, and the communication network between agents constitutes the information layer in the information-physical system. A multi-agent system is constructed, and a topology graph of the multi-agent system is constructed;

[0006] S2. Divide the virtual power plant into multiple regions according to the topology graph. Each region is controlled by an edge server, and all edge servers are connected to a cloud server to construct a cloud-edge-end collaborative architecture of the multi-agent system;

[0007] S3. Establish an equipment model of the virtual power plant;

[0008] S4. According to the equipment model of the virtual power plant, set the optimization goal to minimize the overall expected operating cost while meeting the supply-demand balance and the power limits of each device.

[0009] S5. According to the optimization goal, based on the cloud-edge-end collaborative architecture of the multi-agent system, use the fixed-time consensus optimization algorithm for virtual power plant scheduling.

[0010] Furthermore, the topology graph of the virtual power plant is represented by a binary tuple where represents the node set, each element in the node set represents an agent, ε is the edge set, and the adjacency matrix A = [a ij . If the i-th node and the j-th node communicate with each other, then a ij = 1; otherwise, a ij = 0.

[0011] Furthermore, the equipment of the virtual power plant includes: power generation equipment, energy storage equipment, and loads.

[0012] Furthermore, the equipment model of the virtual power plant includes:

[0013] Power generation equipment cost function:

[0014]

[0015] where represents the cost function of the i-th power generation equipment, represents the power of the i-th power generation equipment, represents different cost coefficients of the i-th power generation equipment;

[0016] Energy storage equipment cost function:

[0017]

[0018] where represents the cost function of the i-th energy storage equipment, represents the power of the i-th energy storage equipment, represents different cost coefficients of the i-th energy storage equipment;

[0019] Energy storage equipment charge and discharge constraints:

[0020]

[0021] where SoC i (η) is the current power of the i-th energy storage equipment at time η, Γ i is the charging or discharging time, B i is the capacity of the energy storage equipment, is the charge and discharge efficiency, Represents the SoC i (η) minimum value, Represents the SoC i (η) maximum value;

[0022] Load cost function:

[0023]

[0024] Wherein, Represents the cost function of the i-th load, Represents the power of the i-th load, And Are both non-negative cost coefficients of the i-th load;

[0025] Active power balance constraint:

[0026]

[0027] Wherein, Represents the set of adjacent nodes of the i-th node, P ji And P ik Respectively represent the active power input and output of the i-th power supply node, B ji And B ik Respectively represent the admittances of the current transmission line ji and the current transmission line ik, θ i 、θ j And θ k Respectively represent the phase angles of the i-th, j-th and k-th nodes, P i Represents the active power generated by the i-th node;

[0028] Power balance of the virtual power plant:

[0029]

[0030] Wherein, P l Represents a constant load, and nw, ns, and nu respectively represent the numbers of power generation equipment, energy storage equipment, and loads.

[0031] Furthermore, the optimization objective function is:

[0032]

[0033] Wherein, τ represents time, with the unit of 1 hour, and T sum Represents the total time;

[0034] Should satisfy the power constraint as follows:

[0035]

[0036] Wherein, And respectively represent the minimum power of the power generation equipment, energy storage equipment, and load, and respectively represent the maximum power of the power generation equipment, energy storage equipment, and load.

[0037] Furthermore, based on the fixed-time consensus optimization algorithm, the optimization problem is described as:

[0038]

[0039] where f(x,τ) represents the total cost of all devices at time τ, x i represents the power generation power of the i-th device, N represents the number of devices, d i represents the power of the i-th load, and Ω represents the set of linear inequality constraints.

[0040] Furthermore, based on the fixed-time consensus optimization algorithm, the distributed algorithm for each agent is described as:

[0041]

[0042] where x i represents the state of the i-th agent, u i represents the parameter used for the i-th agent to obtain the optimal solution within a fixed time, represents the parameter used to drive the state x i to approach the optimal value, represents x i the convergence constraint set, k 1 , k 2 , m 1 , m 2 are positive constants, 0 < a < 1, b > 1, a + b = 2, e i represents the parameter used for the i-th agent to achieve the total load demand constraint, represents the gradient, d i represents the power of the i-th load, f j (x j ) represents the device cost of the j-th agent.

[0043] The present invention also proposes a virtual power plant scheduling system based on a cloud-edge-end architecture, including:

[0044] A multi-agent system construction module for establishing a virtual power plant cyber-physical system model. The physical layer of the cyber-physical system is the equipment of the virtual power plant. Each device corresponds to an agent, and the communication network between the agents constitutes the information layer in the cyber-physical system. A multi-agent system is constructed, and a topology graph of the multi-agent system is constructed;

[0045] A cloud-edge-terminal collaborative architecture construction module, which is used to divide a virtual power plant into multiple regions according to the topology map, each region is controlled by an edge server, and all edge servers are connected to a cloud server to construct a cloud-edge-terminal collaborative architecture of a multi-agent system;

[0046] An equipment model establishment module, which is used to establish an equipment model of the virtual power plant;

[0047] An optimization objective setting module, which is used to set, according to the equipment model of the virtual power plant, an optimization objective of minimizing the overall expected operation cost while meeting the supply-demand balance and the power limits of each device;

[0048] A virtual power plant scheduling module, which is used to perform virtual power plant scheduling according to the optimization objective, based on the cloud-edge-terminal collaborative architecture of the multi-agent system, using a fixed-time consensus optimization algorithm.

[0049] The present invention also provides a computer-readable storage medium, which stores a computer program, and when the computer program is executed by a processor, the virtual power plant scheduling method based on the cloud-edge-terminal architecture described above is implemented.

[0050] The present invention also provides an electronic device, which includes a processor and a memory, and the processor is connected to the memory. Among them, the memory is used to store a computer program, the computer program includes computer-readable instructions, and the processor is configured to call the computer-readable instructions to execute the virtual power plant scheduling method based on the cloud-edge-terminal architecture described above.

[0051] The beneficial effects brought by the technical solution provided by the present invention are:

[0052] The present invention regards the equipment of the virtual power plant as independent agents, constructs a cloud-edge-terminal collaborative architecture of a multi-agent system, and performs virtual power plant scheduling based on the cloud-edge-terminal collaborative architecture of the multi-agent system using a fixed-time consensus optimization algorithm according to the mathematical model and optimization objective of the virtual power plant equipment. The distributed framework of the present invention is a decentralized structure, and local controllers distributed on local agents cooperate with each other. Each agent establishes local communication with its neighboring agents, and can achieve the global control goal only through local control and local information interaction. It significantly reduces the communication construction cost, improves the overall reliability and flexibility of the virtual power plant, and provides an effective remedy for the problems of low reliability and high cost of the centralized algorithm. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 is a flowchart of the virtual power plant scheduling method based on the cloud-edge-terminal architecture according to an embodiment of the present invention;

[0054] Figure 2 is a model diagram of the information-physical system of the virtual power plant according to an embodiment of the present invention;

[0055] Figure 3 It is a framework diagram of the virtual power plant cloud-edge-end collaboration architecture according to an embodiment of the present invention;

[0056] Figure 4 It is a block diagram of an electronic device in an exemplary embodiment of Embodiment 1 of the present invention;

[0057] Figure 5 It is the 24-hour load prediction values of Regions A, B, and C according to an embodiment of the present invention;

[0058] Figure 6 It is the effect of the distributed algorithm of all power generation devices in Region A according to an embodiment of the present invention, where Figure 6 in (a) is the consistency analysis of the output power of all generators in Region A, Figure 6 in (b) is the consistency analysis of the state variables of all generators in Region A;

[0059] Figure 7 It is the effect of the distributed algorithm of all power generation devices in Region B according to an embodiment of the present invention, where Figure 7 in (a) is the consistency analysis of the output power of all generators in Region B, Figure 7 in (b) is the consistency analysis of the state variables of all generators in Region B;

[0060] Figure 8 It is the effect of the distributed algorithm of all power generation devices in Region C according to an embodiment of the present invention, where Figure 8 in (a) is the consistency analysis of the output power of all generators in Region C, Figure 8 in (b) is the consistency analysis of the state variables of all generators in Region C;

[0061] Figure 9 It is the effect of the algorithm without using the cloud-edge-end framework for all power generation devices in all regions (combining ABC into one region), where Figure 9 in (a) is the consistency analysis of the output power of all generators in all regions, Figure 9 in (b) is the consistency analysis of the state variables of all generators in all regions. Detailed implementation manners

[0062] To make the objectives, technical solutions, and advantages of the present invention clearer, the embodiments of the present invention will be further described below in conjunction with the accompanying drawings.

[0063] Embodiment 1: The flowchart of the virtual power plant scheduling method based on the cloud-edge-end architecture according to the embodiment of the present invention is as Figure 1 , and specifically includes the following steps:

[0064] S1. Establish a virtual power plant information - physical system model. The virtual power plant information - physical system model diagram of the embodiments of the present invention is as shown in Figure 2 . The physical layer of the information - physical system is the equipment of the virtual power plant. Each piece of equipment corresponds to an agent, and the communication network between the agents constitutes the information layer in the information - physical system. A multi - agent system is constructed, and a topology graph of the multi - agent system is constructed.

[0065] The equipment of the virtual power plant includes: power generation equipment, energy storage equipment, and loads.

[0066] The topology graph of the virtual power plant is represented by the binary tuple , where represents the node set, and each element in the node set represents an agent. ε is the edge set, and the adjacency matrix A = [a ij is defined. If the i - th node and the j - th node communicate with each other, then a ij = 1; otherwise, a ij = 0.

[0067] S2. Divide the virtual power plant into multiple regions according to the topology graph of the multi - agent system. Each region is controlled by an edge server, and all edge servers are connected to the cloud server to construct a cloud - edge - terminal collaborative architecture of the multi - agent system. Subsequently, the computing resources of the cloud server are used to calculate the power that each region should output. Finally, the cloud server sends the calculation results to each edge server, and the edge server enables each agent to reach an agreement through a distributed optimization algorithm. The framework diagram of the virtual power plant cloud - edge - terminal collaboration architecture of the embodiments of the present invention is as shown in Figure 3 . Among them, the edge - terminal server establishes a local connection with its corresponding region, establishes a communication network connection between the edge - terminal servers, and the edge - terminal server and the cloud server establish 5G communication.

[0068] S3. Establish the equipment model of the virtual power plant. The virtual power plant of the present invention mainly consists of power generation equipment, energy storage equipment, and loads. The above - mentioned equipment will be mathematically modeled below.

[0069] (1) Power generation equipment

[0070] The virtual power plant of the present invention mainly considers thermal power units. As a traditional power generation system, this type of unit converts chemical energy into electrical energy by burning fossil fuels. The new power system needs to rely on small - scale thermal power units to maintain the flexibility and stability of the system.

[0071] Power generation equipment cost function:

[0072]

[0073] Among them, represents the cost function of the i - th power generation equipment, Represents the power of the i-th power generation device, Represents the different cost coefficients of the i-th power generation device. Since the actual cost function is obtained by curve fitting based on data obtained from thermal efficiency tests or power plant design engineers, a non-quadratic function can be used to obtain a better fitting effect.

[0074] (2) Energy storage device

[0075] Battery energy storage is an important energy storage technology. It realizes the storage and output of the battery by using the conversion between electrical energy and chemical energy, and has the advantages of strong environmental adaptability, short construction period, and convenient small-scale configuration. The charge and discharge states of the battery energy storage system must be limited within a certain range to avoid overcharging or over-discharging.

[0076] Cost function of energy storage device:

[0077]

[0078] Among them, Represents the cost function of the i-th energy storage device, Represents the power of the i-th energy storage device, Represents the different cost coefficients of the i-th energy storage device.

[0079] The charge and discharge states of the energy storage device must be limited within a certain range to avoid overcharging or over-discharging. The charge and discharge of the energy storage device need to meet the following constraints:

[0080]

[0081] Among them, SoC i (η) is the current power level of the i-th energy storage device at time η, Γ i is the charging or discharging time, B i is the capacity of the energy storage device, is the charge and discharge efficiency, Represents SoC i (η) minimum value, Represents SoC i (η) maximum value.

[0082] (3) Load

[0083] Load cost function:

[0084]

[0085] Among them, Represents the cost function of the i-th load, Represents the power of the i-th load, and are both non-negative cost coefficients of the i-th load.

[0086] Active power balance constraint:

[0087]

[0088] Among them, represents the set of adjacent nodes of the i-th node, P ji and P ik respectively represent the active power input and output of the i-th power source node, B ji and B ik respectively represent the admittances of the current transmission lines ji and ik, θ i 、θ j and θ k respectively represent the phase angles of the i-th, j-th and k-th nodes, P i represents the active power generated by the i-th node;

[0089] Power balance of the virtual power plant:

[0090]

[0091] Among them, P l represents the constant load, and nw, ns, and nu respectively represent the numbers of power generation equipment, energy storage equipment, and loads.

[0092] S4. According to the equipment model of the virtual power plant, set the optimization objective to minimize the overall expected operating cost while meeting the supply-demand balance and power limits of each equipment.

[0093] The optimization objective function is:

[0094]

[0095] Among them, τ represents time, with the unit of 1 hour, and T sum represents the total time. The cost functions of the equipment in the virtual power plant: The cost function of the power generation equipment The cost function of the load The cost function of the energy storage equipment is related to the power of the equipment (the power per hour), so the cost function is also the cost per hour. When calculating the total operating cost, it is necessary to calculate the operating cost for the total time.

[0096] The following power constraints should be met:

[0097]

[0098] Among them, and respectively represent the minimum powers of the power generation equipment, energy storage equipment, and load, and respectively represent the maximum power of the power generation equipment, energy storage equipment, and load.

[0099] S5. According to the optimization objective, based on the cloud-edge-end collaborative architecture of the multi-agent system, use the fixed-time consensus optimization algorithm for virtual power plant scheduling.

[0100] Based on the fixed-time consensus optimization algorithm, the optimization problem is described as:

[0101]

[0102] where f(x,τ) represents the total cost of all devices at time τ, and x i represents the power generation of the i-th device, N represents the number of devices, and d i represents the power of the i-th load, and Ω represents the set of linear inequality constraints.

[0103] Based on the fixed-time consensus optimization algorithm, the distributed algorithm for each agent is described as:

[0104]

[0105] where x i represents the state of the i-th agent, u i represents the parameter used for the i-th agent to obtain the optimal solution within a fixed time, represents the parameter used to drive the state x i to approach the optimal value, represents x i the convergence constraint set, k 1 ,k 2 ,m 1 ,m 2 are positive constants, 0 < a < 1, b > 1, a + b = 2, and e i represents the parameter used for the i-th agent to achieve the total load demand constraint, represents the gradient, and d i represents the power of the i-th load, and f j (x j ) represents the device cost of the j-th agent.

[0106] In the cloud-edge-end collaborative control framework, the power generation in each region is first solved by the cloud. Then, the distributed consensus algorithm is applied in each region. The steps of the distributed multi-agent consensus algorithm are as follows:

[0107] 1. Input the topological structure and predicted load value, and set the iteration number k = 0;

[0108] 2. Update the consensus variables of the agents by exchanging the gradients of the agents;

[0109] 3. Calculate the updated value x of each agent i (k + 1);

[0110] 4. Calculate the power deviation

[0111] 5. If the active power deviation value meets the requirements, the iteration ends. If not, set the iteration number k = k + 1 and repeat the above steps.

[0112] Embodiment 2: The present invention also proposes a virtual power plant scheduling system based on a cloud-edge-end architecture, including:

[0113] A multi-agent system construction module for establishing a virtual power plant cyber-physical system model. The physical layer of the cyber-physical system is the equipment of the virtual power plant, each equipment corresponds to an agent, and the communication network between agents constitutes the information layer in the cyber-physical system, constructing a multi-agent system and constructing a topology map of the multi-agent system;

[0114] A cloud-edge-end collaborative architecture construction module for dividing the virtual power plant into multiple regions according to the topology map, each region is controlled by an edge server, and all edge servers are connected to the cloud server, constructing a cloud-edge-end collaborative architecture of the multi-agent system;

[0115] An equipment model establishment module for establishing an equipment model of the virtual power plant;

[0116] An optimization objective setting module for setting the optimization objective to minimize the overall expected operating cost while meeting the supply-demand balance and power limits of each equipment according to the equipment model of the virtual power plant;

[0117] A virtual power plant scheduling module for performing virtual power plant scheduling based on the cloud-edge-end collaborative architecture of the multi-agent system using a fixed-time consensus optimization algorithm according to the optimization objective.

[0118] Embodiment 3: In an exemplary embodiment, it includes a computer-readable storage medium. The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, it implements the above-mentioned virtual power plant scheduling method based on the cloud-edge-end architecture.

[0119] Embodiment 4: Please refer to Figure 4 , in an exemplary embodiment, it also includes an electronic device, including at least one processor, at least one memory, and at least one communication bus.

[0120] Among them, the memory stores a computer program, the computer program includes computer-readable instructions, and the processor calls the computer-readable instructions stored in the memory through the communication bus to execute the above-mentioned virtual power plant scheduling method based on the cloud-edge-end architecture.

[0121] The present invention uses a three - area system to illustrate the effect of the solution of the present invention. The schematic diagram of the three - area system is as shown in Figure 3 . Each area contains energy storage devices, power generation devices, and shed - dable loads, with a total of 29 agents. In addition, the parameters of the algorithm are designed as k 1 = k 2 = m 1 = m 2 = 1, a = 0.5, and b = 1.5.

[0122] For this three - area system, the platform collects information from all areas and then solves for the optimal power generation values of areas A, B, and C. Figure 5 Shows the predicted loads for each time period in the three areas A, B, and C obtained according to the 24 - hour load prediction values.

[0123] Figure 6 Is the effect of the distributed algorithm of all power generation devices in area A of the embodiment of the present invention, where Figure 6 (a) is the consistency analysis of the output power of all generators in area A, Figure 6 (b) is the consistency analysis of the state variables of all generators in area A; Figure 7 Is the effect of the distributed algorithm of all power generation devices in area B of the embodiment of the present invention, where Figure 7 (a) is the consistency analysis of the output power of all generators in area B, Figure 7 (b) is the consistency analysis of the state variables of all generators in area B; Figure 8 Is the effect of the distributed algorithm of all power generation devices in area C of the embodiment of the present invention, where Figure 8 (a) is the consistency analysis of the output power of all generators in area C, Figure 8 (b) is the consistency analysis of the state variables of all generators in area C; Figure 9 Is the effect of the algorithm without using the cloud - edge - terminal framework for all power generation devices in all areas (combining ABC into one area), where Figure 9 (a) is the consistency analysis of the output power of all generators in all areas, Figure 9 (b) is the consistency analysis of the state variables of all generators in all areas.

[0124] In Figure 6 (a), Figure 7 (a), and Figure 8 (a), it is shown that the consistency algorithm can solve the optimal power of each power generation device. In the simulation, different power generation units have different initial power generation amounts, which also indicates that their consistency variables have different initial values. Figure 6 (b), Figure 7 (b), andFigure 8 (b) It shows that the algorithm can achieve consistency through several iterations within a fixed time. To highlight the advantages of the cloud-edge-terminal collaboration framework, Figure 9 the effect of the distributed algorithm without the cloud-edge-terminal framework is shown, which requires more iterations to converge to the same accuracy.

[0125] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present invention. Various modifications to these embodiments will be obvious to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to these embodiments shown herein, but rather to the broadest scope consistent with the principles and novel features disclosed herein.

Claims

1. A virtual power plant scheduling method based on cloud-edge-end architecture, characterized in that: The following steps are involved: S1. Establish a virtual power plant information-physical system model. The physical layer of the information-physical system is the equipment of the virtual power plant. Each equipment corresponds to an intelligent agent. The communication network between intelligent agents constitutes the information layer in the information-physical system. A multi-agent system is constructed, and a topological diagram of the multi-agent system is constructed. S2. Divide the virtual power plant into multiple areas according to the topology diagram, each area is controlled by an edge server, and all edge servers are connected to the cloud server to build a cloud-edge collaborative architecture of the multi-agent system; S3. Establish equipment model of virtual power plant; S4. According to the equipment model of the virtual power plant, the optimization goal is to minimize the overall expected operating cost while meeting the supply and demand balance and the power limit of each equipment; S5. According to the optimization objectives, based on the cloud-edge collaborative architecture of the multi-agent system, a fixed-time consistent optimization algorithm is used to schedule virtual power plants.

2. According to a virtual power plant scheduling method based on cloud-edge-end architecture according to claim 1, it is characterized in that: The topology of a multi-agent system is represented by two tuples. Indicates that represents a node set, each element in the node set represents an agent, ε is an edge set, and the adjacency matrix A is defined as [a ij ], if the i-th node and the j-th node communicate with each other, then a ij =1, otherwise a ij =0.

3. According to a virtual power plant scheduling method based on cloud-edge-end architecture according to claim 1, it is characterized in that: The equipment of a virtual power plant includes: power generation equipment, energy storage equipment and loads.

4. A virtual power plant scheduling method based on cloud-edge-end architecture according to claim 3, characterized in that: The equipment models of the virtual power plant include: Power generation equipment cost function: in, represents the cost function of the ith power generation equipment, represents the power of the i-th power generation equipment, represents the different cost coefficients of the i-th power generation equipment; Energy storage equipment cost function: in, represents the cost function of the i-th energy storage device, represents the power of the i-th energy storage device, represents the different cost coefficients of the i-th energy storage device; Energy storage equipment charging and discharging constraints: Among them, SoC i (η) is the current power of the i-th energy storage device at time η, Γ i is the charge or discharge time, B i is the capacity of the energy storage device, θ i is the charge and discharge efficiency, Represents SoC i The minimum value of (η), Represents SoC i The maximum value of (η); Load cost function: in, represents the cost function of the ith load, represents the power of the i-th load, and are all non-negative cost coefficients of the i-th load; Active power balance constraints: in, represents the set of neighboring nodes of the i-th node, P ji and P ik Respectively represent the active power input and output of the i-th power node, B ji and Bik represent the admittance of current transmission line ji and line ik respectively, θ i ,θ j and θ k Represent the phase angles of the i-th, j-th and k-th nodes respectively, P i represents the active power generated by the i-th node; The power balance formula of the virtual power plant is: Among them, P l represents constant load, nw, ns, and nu represent the number of power generation equipment, energy storage equipment, and load, respectively.

5. A virtual power plant scheduling method based on cloud-edge-end architecture according to claim 4, characterized in that: The optimization objective function is: Among them, τ represents time, the unit is 1 hour, T sum Indicates the total time; The power constraints to be met are as follows: in, and Respectively represent the minimum power of power generation equipment, energy storage equipment and load, and Respectively represent the maximum power of power generation equipment, energy storage equipment and load.

6. A virtual power plant scheduling method based on cloud-edge-end architecture according to claim 5, characterized in that: Based on the fixed-time consistent optimization algorithm, the optimization problem is described as: x i ∈Ω Where f(x, τ) represents the sum of all equipment costs at time τ, x i represents the power generated by the i-th device, N represents the number of devices, and d i represents the power of the i-th load, and Ω represents the linear inequality constraint set.

7. A virtual power plant scheduling method based on cloud-edge-end architecture according to claim 6, characterized in that: Based on the fixed-time consistent optimization algorithm, the distributed algorithm of each agent is described as: x [k] =sign(x)|x| k Among them, x i represents the state of the ith agent, u i represents the parameters used by the ith agent to obtain the optimal solution within a fixed time, Indicates the driving state x i Parameters close to the optimal value, represents the set of constraints for which xi converges, k1, k2, m1, m2 are positive constants, 0<a<1, b>1, a+b=2, e i represents the parameter for the total load demand constraint of the ith intelligent implementation, represents the gradient, d i represents the power of the ith load, f j (x j ) represents the equipment cost of the j-th agent.

8. A virtual power plant dispatching system based on cloud-edge-end architecture, characterized in that: include: The multi-agent system building module is used to establish a virtual power plant cyber-physical system model. The physical layer of the cyber-physical system is the equipment of the virtual power plant. Each equipment corresponds to an agent. The communication network between agents constitutes the information layer in the cyber-physical system, builds a multi-agent system, and builds a topological diagram of the multi-agent system. A cloud-edge-end collaborative architecture building module is used to divide the virtual power plant into multiple areas according to the topology diagram, each area is controlled by an edge server, and all edge servers are connected to the cloud server to build a cloud-edge-end collaborative architecture for a multi-agent system; Equipment model building module, used to build equipment models of virtual power plants; An optimization target setting module is used to set the optimization target of minimizing the overall expected operating cost while satisfying the supply and demand balance and the power limit of each device according to the device model of the virtual power plant; The virtual power plant scheduling module is used to schedule virtual power plants according to the optimization objectives and based on the cloud-edge collaborative architecture of the multi-agent system, using a fixed-time consistent optimization algorithm.

9. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

10. An electronic device, characterized in that: The method comprises a processor and a memory, wherein the processor and the memory are interconnected, wherein the memory is used to store a computer program, the computer program comprises computer-readable instructions, and the processor is configured to call the computer-readable instructions to execute the method according to any one of claims 1 to 7.

Images