A high energy consumption load resource adjustment method based on virtual power plant

Abstract

The present invention relates to the technical field of power load management, and discloses a high-energy-consuming load resource regulation method based on a virtual power plant, comprising: step 1, describing the operating behavior of each load resource in the virtual power plant by constructing a high-energy-consuming load resource dynamic scheduling model of the virtual power plant, and establishing the relationship between the state variable and the control input during the load regulation process through the corresponding dynamic equation. The high-energy-consuming load resource dynamic scheduling model enables the real-time state and control parameters of the load regulation to be described when modeling the operating process of each load resource. By introducing optimal control theory and scheduling optimization algorithm, combined with the penalty term of power output change, the regulation process of high-energy-consuming load is optimized to avoid frequent fluctuations, and by punishing power changes, load fluctuations are reduced, load regulation is achieved, and the effect of improving the stability of the power grid is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of power load management, and in particular to a high-energy-consuming load resource regulation method based on a virtual power plant. Background Art

[0002] The transformation of energy structures and the large-scale integration of renewable energy have led to significant load fluctuations in power grids. Virtual power plants, as systems integrating multiple distributed energy resources, energy storage devices, and load resources, aim to balance grid loads through intelligent scheduling and optimization, improving the grid's regulation capabilities and stability. High-energy-consuming load resources, a crucial component of virtual power plants, pose significant scheduling challenges due to their high volatility and impact on the grid.

[0003] Current load regulation methods mostly rely on scheduling strategies based on experience or simple rules. These methods often suffer from low scheduling accuracy, high scheduling delays, and large load fluctuations when regulating large, energy-intensive loads. Furthermore, traditional load resource scheduling methods lack precise control over equipment startup and shutdown. Frequent startup and shutdown operations affect power system stability, increase equipment maintenance costs, and reduce energy efficiency.

[0004] Previous research has attempted to schedule load resources through optimization algorithms and control theory, but most methods still fail to meet the requirements of real-time performance, accuracy, and system stability when faced with complex power grid environments and large-scale load resources. Therefore, how to improve the efficiency of regulating high-energy-consuming load resources and the stability of the power grid through more advanced control theory, optimization algorithms, and real-time scheduling technologies has become a key technical issue that needs to be addressed in the field of virtual power plants.

[0005] The present invention introduces optimal control theory and scheduling optimization algorithm, combines mixed integer programming algorithm and parallel computing technology, and proposes a high-energy-consuming load resource regulation method based on virtual power plant to solve the above-mentioned problems. Summary of the Invention

[0006] In view of the deficiencies in the prior art, the present invention provides a high-energy-consuming load resource regulation method based on a virtual power plant to solve the problems raised in the above background technology.

[0007] To achieve the above objectives, the present invention is implemented through the following technical solutions: 1. A high-energy-consuming load resource adjustment method based on a virtual power plant, comprising:

[0008] Step 1: A dynamic scheduling model for high-energy-consuming load resources in a virtual power plant is constructed to describe the operating behavior of each load resource in the virtual power plant, and the relationship between state variables and control inputs in the load regulation process is established through corresponding dynamic equations. The dynamic scheduling model for high-energy-consuming load resources enables the real-time state and control parameters of load regulation to be described when modeling the operating process of each load resource.

[0009] Step 2: Based on the load resource scheduling data obtained from the high-energy-consuming load resource dynamic scheduling model, define the scheduling optimization goal and set the constraints;

[0010] In step 2, the scheduling optimization target of the load resources is expressed as:

[0011]

[0012] Where N is the total number of high energy consumption load resources in the virtual power plant,

[0013] T is the number of time periods considered in the scheduling process,

[0014] C i (P i (t)) represents the scheduling cost of the i-th load resource at time t,

[0015] P i (t) represents the power output of the i-th load resource at time t,

[0016] λ i is the penalty factor, μ i is the cost of starting and stopping the i-th load resource,

[0017] |P i (t)-P i (t-1)| represents the absolute value of the power output change of the i-th load resource between time t and t-1,

[0018] z i (t) is the on / off state of the i-th load resource at time t;

[0019] The constraints of the scheduling optimization include:

[0020] Grid power balance constraints:

[0021] Among them, P i (t) represents the power output of the i-th load resource at time t, N is the total number of high-energy-consuming load resources in the virtual power plant, P grid (t) is the power demand of the grid at time t, It is used for time point t;

[0022] Load regulation capability constraint: P min.i ≤P i (t)≤P max.i ,

[0023] Among them, P min.i and P max.i is the minimum and maximum power output of the i-th load resource, P i (t) represents the power output of the i-th load resource at time t, Represents all load resources i and time t;

[0024] Step 3: Based on the scheduling optimization objective, constraints are set and a control algorithm is used to optimize the scheduling of load resources. The control algorithm adjusts the power output of the load resources to minimize the scheduling cost of the virtual power plant. The control algorithm considers the on / off status of the load devices and optimizes the start and stop operations of the load resources through mixed integer programming.

[0025] Step 4: Analyze the defined dispatch model using the calculus of variations and optimal control theory to obtain the optimal dispatch strategy. By solving the optimization equation, the power output and switching state of each load resource at different time steps are obtained, providing an operation plan for the virtual power plant dispatch.

[0026] Step 5: Perform stability analysis on the obtained scheduling scheme and use Lyapunov stability theory to analyze the stability of the scheduling system;

[0027] Step 6: Combine parallel computing technology to accelerate the optimization process. Through the real-time data acquisition system, according to the load fluctuation of the power grid, dynamically adjust the power output of each high-energy-consuming load resource in the virtual power plant to achieve efficient load regulation.

[0028] Preferably, in step 3, the control algorithm adopts optimal control theory to perform scheduling optimization, and the optimal control equation is expressed as:

[0029]

[0030] Among them, L is the Lagrangian function, B is the bias term,

[0031] P i (t) represents the power output of the i-th load resource at time t,

[0032] z i (t) represents the switch state of the i-th load resource at time t,

[0033] If z i (t)=1, load resource is turned on,

[0034] If zi (t) = 0, the load resource is closed.

[0035] Preferably, the scheduling algorithm uses mixed integer programming to optimize the start and stop operations of load resources, and the constraints of the start and stop operations are expressed as:

[0036]

[0037] Among them, z i (t) represents the switch state of the i-th load resource at time t,

[0038] If z i (t)=1, load resource is turned on,

[0039] If z i (t) = 0, the load resource is closed.

[0040] Preferably, the high-energy-consuming load resource regulation method includes analyzing the defined scheduling model using the variational method and optimal control theory to obtain an optimal scheduling strategy, which is expressed as follows by solving the optimization equation:

[0041] P i (t) = f i (P previous (t), z i (t)),

[0042] Among them, P previous (t) represents the power output of the previous time step, z i (t) represents the switch state of the i-th load resource at time t, f i (·) is the optimal dispatching strategy derived from the dynamic model of load regulation.

[0043] Preferably, in step 5, the scheduling scheme is analyzed by Lyapunov stability theory to ensure that the system will not become unstable during the load regulation process. The Lyapunov function is defined as:

[0044]

[0045] Among them, P ref.i is the target power output, P i (t) represents the power output of the i-th load resource at time t, V(P1(t),…,P N (t)) represents the Lyapunov function, and N is the total number of high-energy-consuming load resources in the virtual power plant.

[0046] Preferably, the high-energy-consuming load resource adjustment method solves the optimal scheduling by combining the gradient descent method with dynamic programming to obtain the optimal scheduling strategy for each load resource. The optimal scheduling strategy is expressed as:

[0047]

[0048] in, is the gradient of the Lagrangian function, δP i (t) represents the adjustment of the power output of the load resource at each time step, Represents all load resources i and time t.

[0049] Preferably, the high-energy-consuming load resource adjustment method improves computing efficiency through parallel computing technology, and performs parallel optimization on the scheduling of each load resource. The parallel optimization calculation process is performed by the following formula:

[0050]

[0051] in, As the overall optimization goal, is the parallel calculation of the i-th load resource at time t, N is the total number of high-energy-consuming load resources in the virtual power plant, P i (t) represents the power output of the i-th load resource at time t, Denotes the time point t.

[0052] Preferably, the high-energy-consuming load resource adjustment method includes a real-time data acquisition and feedback control system, which monitors grid load fluctuations in real time and dynamically adjusts the power output of each high-energy-consuming load resource in the virtual power plant based on real-time load data. The dynamic adjustment process is controlled by the following formula:

[0053] Pi(t)=Adjust(P current (t), P target (t)),

[0054] Among them, P current (t) is the power output of load resource i at the current moment, P target (t) is the target power output, Adjust(·) is the control function for adjusting the load output according to real-time data, P i (t) represents the power output of the i-th load resource at time t.

[0055] The present invention provides a method for regulating high-energy-consuming load resources based on a virtual power plant. It has the following beneficial effects:

[0056] 1. The present invention introduces optimal control theory and scheduling optimization algorithm, combined with the penalty term of power output change, to optimize the regulation process of high-energy-consuming loads, avoid frequent fluctuations, and reduce load fluctuations by punishing power changes, thereby achieving load regulation and improving the stability of the power grid.

[0057] 2. The present invention optimizes the start and stop operations of load resources by adopting a mixed integer programming algorithm, ensures the coordination between the on and off status of the load equipment and the regulation strategy, reduces unnecessary start and stop operations, and improves the operating efficiency of the virtual power plant by reasonably arranging the start and stop of load resources, thereby reducing the frequent start and stop of equipment and extending the service life of the equipment.

[0058] 3. The present invention improves the computational efficiency of the scheduling optimization process by combining parallel computing technology, ensuring that the virtual power plant can adjust load resources in real time when the grid load fluctuates, and by accelerating the scheduling process, improves the response speed of the system, thereby achieving the effect of being able to quickly respond to changes in grid load and maintain grid balance. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 Flowchart of the present invention. DETAILED DESCRIPTION

[0060] To help those skilled in the art understand the present invention, the following will provide a clear and complete description of the technical solutions in the embodiments of the present invention, in conjunction with the accompanying drawings. Obviously, the described embodiments are only partial embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0061] The present invention is described in detail below with reference to the accompanying drawings:

[0062] Example:

[0063] Please see the attached Figure 1 The embodiment of the present invention provides a method for regulating high-energy-consuming load resources based on a virtual power plant, comprising:

[0064] Step 1: By constructing a dynamic scheduling model for high-energy-consuming load resources in the virtual power plant, the operating behavior of each load resource in the virtual power plant is described, and the relationship between the state variables and the control input during the load regulation process is established through the corresponding dynamic equations. The dynamic scheduling model for high-energy-consuming load resources enables the real-time state and control parameters of load regulation to be described when modeling the operating process of each load resource;

[0065] Step 2: Based on the load resource scheduling data obtained from the high-energy-consuming load resource dynamic scheduling model, define the scheduling optimization goal and set the constraints;

[0066] Step 3: Based on the scheduling optimization goal, constraints are set and a control algorithm is used to optimize the scheduling of load resources. The control algorithm adjusts the power output of load resources to minimize the scheduling cost of the virtual power plant. The control algorithm considers the on / off status of load devices and optimizes the start and stop operations of load resources through mixed integer programming.

[0067] Step 4: Analyze the defined dispatch model using the calculus of variations and optimal control theory to obtain the optimal dispatch strategy. By solving the optimization equation, the power output and switching state of each load resource at different time steps are obtained, providing an operation plan for the virtual power plant dispatch.

[0068] Step 5: Perform stability analysis on the obtained scheduling scheme and use Lyapunov stability theory to analyze the stability of the scheduling system;

[0069] Step 6: Combine parallel computing technology to accelerate the optimization process. Through the real-time data acquisition system, according to the load fluctuation of the power grid, dynamically adjust the power output of each high-energy-consuming load resource in the virtual power plant to achieve efficient load regulation.

[0070] Benefits of Step 1: By building a dynamic dispatch model, this step enables the virtual power plant to describe the operating behavior and regulation characteristics of each high-energy-consuming load resource. Dynamic equations are used to establish the relationship between state variables and control inputs during load regulation, ensuring real-time dispatch and precise control in complex power grid environments, improving dispatch reliability and accuracy. This step provides a modeling foundation for subsequent dispatch optimization, enabling flexible response to various load regulation requirements.

[0071] Benefits of Step 2: By defining scheduling optimization objectives and setting constraints based on the scheduling data obtained from the model, we ensure that the scheduling objectives match actual demand, avoiding over-adjustment and failure to meet grid demand. This step ensures that load resource adjustment can be carried out within a reasonable power range while meeting the grid's power balance requirements, thus ensuring stable grid operation.

[0072] Benefits of Step 3: The introduction of a control algorithm minimizes virtual power plant scheduling costs by adjusting the power output of load resources while meeting the grid's load demand and stability requirements. A mixed integer programming algorithm optimizes the start and stop operations of load resources, avoiding frequent equipment startups and shutdowns, reducing operating costs, and improving equipment efficiency. This step maximizes economic benefits while reducing unnecessary energy consumption and equipment wear.

[0073] Benefits of Step 4: Variational methods and optimal control theory enable in-depth analysis of the dispatch model, identifying the optimal dispatch strategy that maintains grid stability and meets load demand. By solving the optimization equations, the optimal power output and switching states of each load resource at different points in time are determined, ensuring the optimality of the dispatch results. This provides a basis for scheduling decisions in the virtual power plant and optimizes the system's resource allocation.

[0074] Benefits of Step 5: By analyzing the dispatch plan's stability and applying Lyapunov stability theory, we ensure stable system operation during load regulation, avoiding excessive fluctuations and grid frequency instability during load regulation. This step provides theoretical support for the system, ensuring the safety and stability of the regulation process for high-energy-consuming load resources and guaranteeing long-term stable grid operation.

[0075] Benefits of Step 6: Incorporating parallel computing technology, this step significantly improves the computational efficiency of the dispatch optimization process, enabling the virtual power plant to quickly respond to grid load fluctuations. The real-time data acquisition system ensures dynamic monitoring of grid load fluctuations and enables precise adjustment of high-energy-consuming load resources within the virtual power plant. This improves system response speed and ensures grid load balance and stability.

[0076] In step 2, the scheduling optimization objective of load resources is expressed as:

[0077]

[0078] Where N is the total number of high energy consumption load resources in the virtual power plant,

[0079] T is the number of time periods considered in the scheduling process,

[0080] C i (P i (t)) represents the scheduling cost of the i-th load resource at time t,

[0081] P i (t) represents the power output of the i-th load resource at time t,

[0082] λ i is the penalty factor, μ i is the cost of starting and stopping the i-th load resource,

[0083] |P i (t)-P i (t-1)| represents the absolute value of the power output change of the i-th load resource between time t and t-1,

[0084] z i (t) is the on / off state of the i-th load resource at time t.

[0085] The constraints for scheduling optimization include:

[0086] Grid power balance constraints:

[0087] Among them, P i (t) represents the power output of the i-th load resource at time t, N is the total number of high-energy-consuming load resources in the virtual power plant, P grid (t) is the power demand of the grid at time t, It is used for time point t;

[0088] Load regulation capability constraints:

[0089] Among them, P min.i and P max.i is the minimum and maximum power output of the i-th load resource, P i (t) represents the power output of the i-th load resource at time t, Represents all load resources i and time t.

[0090] By defining a dispatch optimization objective function, this step systematizes the dispatch costs and constraints for all high-energy-consuming load resources in the virtual power plant, providing a mathematical foundation for subsequent dispatch algorithms. This objective function incorporates dispatch costs, load fluctuations, and startup and shutdown costs, comprehensively considering both economic benefits and system stability to ensure that the dispatch strategy balances costs and grid demand.

[0091] By introducing penalty factors, particularly penalties for absolute value changes in load output, this step can effectively reduce frequent fluctuations in load resource regulation, avoiding grid instability caused by frequent fluctuations in traditional scheduling methods. This ensures grid frequency stability and improves load resource regulation accuracy.

[0092] In this step, by considering the start-up and shutdown costs of load resources, we rationally schedule the start-up and shutdown of load equipment, reducing unnecessary start-up and shutdown operations. This rational start-up and shutdown schedule improves the operating efficiency of the virtual power plant, reduces equipment wear and tear caused by frequent starts and stops, and extends equipment life, thereby improving the economic efficiency of the virtual power plant.

[0093] While defining the dispatch objectives, this step ensures that constraints are met during the dispatch optimization process, such as grid power balance and load regulation capacity constraints. Grid power balance constraints ensure that the power output of load resources is consistent with grid load demand, avoiding overloading or underloading. Load regulation capacity constraints ensure that the power output of each load resource remains within the adjustable range, avoiding adjustments that exceed the physical limitations of the equipment and ensuring the safety and feasibility of dispatch operations.

[0094] In step 3, the control algorithm uses optimal control theory to perform scheduling optimization, and the optimal control equation is expressed as:

[0095]

[0096] Among them, L is the Lagrangian function, B is the bias term,

[0097] P i (t) represents the power output of the i-th load resource at time t,

[0098] z i (t) represents the switch state of the i-th load resource at time t,

[0099] If z i (t)=1, load resource is turned on,

[0100] If z i (t) = 0, the load resource is closed.

[0101] The scheduling algorithm uses mixed integer programming to optimize the start and stop operations of load resources. The constraints of the start and stop operations are expressed as:

[0102]

[0103] Among them, z i (t) represents the switch state of the i-th load resource at time t,

[0104] If z i (t)=1, load resource is turned on,

[0105] If z i (t) = 0, the load resource is closed.

[0106] By employing optimal control theory and combining it with Lagrangian function optimization to optimize the dispatching process, the load resources within the virtual power plant can be precisely regulated. The optimal control equation optimizes the power output of each load resource, ensuring grid load balance while maximizing the economic benefits of the power system. Compared to traditional dispatching methods, the optimal control algorithm achieves precise load regulation, improves grid stability, and reduces energy waste.

[0107] This step optimizes the start and stop operations of load resources by introducing a mixed integer programming algorithm. Properly scheduling start and stop operations can reduce the frequent start and stop of load equipment, lowering energy loss and wear during startup, and extending equipment life. By optimizing the on / off states, the operating costs of the virtual power plant can be reduced, while improving the overall efficiency of the system and avoiding the resource waste caused by the frequent start and stop operations in traditional methods.

[0108] By optimizing the switching state through control algorithms, we can achieve the proper regulation of various load resources in different time periods, avoiding over-regulation or non-regulation of load resources. Especially during peak load periods, proper regulation of the switching state of load equipment can balance load demand and grid capacity, improving the grid's regulation capabilities.

[0109] The high-energy-consuming load resource regulation method includes using the variational method and optimal control theory to analyze the defined scheduling model to obtain the optimal scheduling strategy. The optimal scheduling strategy is expressed as follows by solving the optimization equation:

[0110] P i (t) = f i (P previous (t), z i (t)),

[0111] Among them, P previous (t) represents the power output of the previous time step, z i (t) represents the switch state of the i-th load resource at time t, f i (·) is the optimal dispatching strategy derived from the dynamic model of load regulation.

[0112] By applying the calculus of variations and optimal control theory, we can mathematically derive the optimal scheduling strategy for high-energy-consuming load resources in virtual power plants. This approach minimizes scheduling errors and energy losses, ensuring grid load balance while reducing fluctuations and instabilities in the scheduling process. The optimal scheduling strategy precisely calculates the power output and switching state of each load resource at different time steps, providing an efficient load regulation solution for the grid.

[0113] By introducing the power output P of the previous time step previous (t) and the switch state z at the current time step i (t), achieving dynamic adjustment and real-time optimization. The optimal scheduling strategy considers the current state of load resources and integrates past power output, making the load regulation process highly continuous and adaptable, and enabling real-time response to grid load fluctuations. Dynamic modeling and optimization enable virtual power plants to efficiently handle complex and variable load demands.

[0114] By solving the optimal scheduling equation, we can ensure that the power output and switching states of each load resource during the scheduling process meet the grid's requirements, while also avoiding excessive fluctuations in load resources and frequent equipment startups and shutdowns. This improves the system's regulatory capabilities, ensures stable grid operation, and mitigates issues with grid frequency instability caused by excessive fluctuations. Furthermore, the optimal scheduling strategy helps improve the operating efficiency of virtual power plants and reduce scheduling costs.

[0115] In step 5, the scheduling plan is analyzed by Lyapunov stability theory to ensure that the system will not become unstable during the load regulation process. The Lyapunov function is defined as:

[0116]

[0117] Among them, P ref.i is the target power output, P i (t) represents the power output of the i-th load resource at time t, V(P1(t),…,P N (t)) represents the Lyapunov function, and N is the total number of high-energy-consuming load resources in the virtual power plant.

[0118] This method analyzes scheduling plans using Lyapunov stability theory to ensure that virtual power plants do not experience instability during load regulation. By defining the Lyapunov function, the stability of the power grid system can be analyzed and monitored in real time, preventing instability caused by excessive fluctuations and unreasonable regulation strategies. This method ensures stability and provides digital support for scheduling plans during load regulation, thereby enhancing the reliability of the virtual power plant and the security of the power grid. Through this process, the system can maintain efficient and stable operation in the face of uncertainty and load fluctuations.

[0119] The high-energy-consuming load resource regulation method solves the optimal scheduling by combining the gradient descent method with dynamic programming to obtain the optimal scheduling strategy for each load resource. The optimal scheduling strategy is expressed as:

[0120]

[0121] in, is the gradient of the Lagrangian function, δP i (t) represents the adjustment of the power output of the load resource at each time step, Represents all load resources i and time t.

[0122] This invention effectively solves the optimal scheduling problem for regulating high-energy-consuming load resources in virtual power plants by combining the gradient descent method with dynamic programming. The gradient descent method adjusts power output by optimizing the gradient of the objective function, while dynamic programming ensures a globally optimal solution. This combination improves the efficiency of the scheduling process and enhances the system's scheduling accuracy and stability. It also enables rapid response to grid load fluctuations, ensuring the efficient and stable operation of the virtual power plant. This approach offers significant advantages in addressing complex grid load fluctuations and diverse demands, providing optimized support for the operation of virtual power plants.

[0123] The high-energy-consuming load resource adjustment method improves computing efficiency through parallel computing technology and performs parallel optimization on the scheduling of each load resource. The parallel optimization calculation process is performed through the following formula:

[0124]

[0125] in, As the overall optimization goal, is the parallel calculation of the i-th load resource at time t, N is the total number of high-energy-consuming load resources in the virtual power plant, P i (t) represents the power output of the i-th load resource at time t, Denotes the time point t.

[0126] By adopting parallel computing technology, the present invention effectively solves the problem of low computational efficiency of traditional load resource scheduling methods when facing large-scale power grids and high-energy-consuming load regulation. Parallel optimization speeds up the scheduling process, shortens scheduling time, and improves resource allocation efficiency, especially when dealing with power grid load fluctuations, and can achieve real-time response. By simultaneously optimizing the scheduling of multiple load resources, the system can quickly adapt to changes in power grid load and ensure the load balance and stability of the power grid. This method provides an efficient and flexible scheduling solution for virtual power plants, significantly improving the operating efficiency and response speed of the system.

[0127] The high-energy-consuming load resource adjustment method includes a real-time data acquisition and feedback control system. The real-time data acquisition and feedback control system monitors the grid load fluctuation in real time and dynamically adjusts the power output of each high-energy-consuming load resource in the virtual power plant based on the real-time load data. The dynamic adjustment process is controlled by the following formula:

[0128] P i (t)=Adjust(P current (t), P target (t)),

[0129] Among them, P curent (t) is the power output of load resource i at the current moment, P target (t) is the target power output, Adjust(·) is the control function for adjusting the load output according to real-time data, P i (t) represents the power output of the i-th load resource at time t.

[0130] By introducing real-time data acquisition and feedback control, this invention significantly improves the real-time responsiveness and regulation accuracy of virtual power plants during load regulation. By monitoring grid load fluctuations in real time and dynamically adjusting the power output of load resources based on real-time data, it ensures stable grid operation and effectively avoids load imbalances. This feedback control mechanism enables flexible and precise load regulation, improving the operational efficiency and system stability of the virtual power plant, and supporting the smooth operation of the power system.

[0131] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A high energy consumption load resource adjustment method based on a virtual power plant, characterized in that: include: Step 1: A dynamic scheduling model for high-energy-consuming load resources in a virtual power plant is constructed to describe the operating behavior of each load resource in the virtual power plant, and the relationship between state variables and control inputs in the load regulation process is established through corresponding dynamic equations. The dynamic scheduling model for high-energy-consuming load resources enables the real-time state and control parameters of load regulation to be described when modeling the operating process of each load resource. Step 2: Based on the load resource scheduling data obtained from the high-energy-consuming load resource dynamic scheduling model, define the scheduling optimization goal and set the constraints; In step 2, the scheduling optimization target of the load resources is expressed as: Where N is the total number of high energy consumption load resources in the virtual power plant, T is the number of time periods considered in the scheduling process, C i (P i (t)) represents the scheduling cost of the i-th load resource at time t, P i (t) represents the power output of the i-th load resource at time t, λ i is the penalty factor, μ i is the cost of starting and stopping the i-th load resource, |P i (t)-P i (t-1)| represents the absolute value of the power output change of the i-th load resource between time t and t-1, z i (t) is the on / off state of the i-th load resource at time t; The constraints of the scheduling optimization include: Grid power balance constraints: Among them, P i (t) represents the power output of the i-th load resource at time t, N is the total number of high-energy-consuming load resources in the virtual power plant, P grid (t) is the power demand of the grid at time t, It is used for time point t; Load regulation capability constraints: Among them, P min.i and P max.i is the minimum and maximum power output of the i-th load resource, P i (t) represents the power output of the i-th load resource at time t, Represents all load resources i and time t; Step 3: Based on the scheduling optimization objective, constraints are set and a control algorithm is used to optimize the scheduling of load resources. The control algorithm adjusts the power output of the load resources to minimize the scheduling cost of the virtual power plant. The control algorithm considers the on / off status of the load devices and optimizes the start and stop operations of the load resources through mixed integer programming. Step 4: Analyze the defined dispatch model using the calculus of variations and optimal control theory to obtain the optimal dispatch strategy. By solving the optimization equation, the power output and switching state of each load resource at different time steps are obtained, providing an operation plan for the virtual power plant dispatch. Step 5: Perform stability analysis on the obtained scheduling scheme and use Lyapunov stability theory to analyze the stability of the scheduling system; Step 6: Combine parallel computing technology to accelerate the optimization process. Through the real-time data acquisition system, according to the load fluctuation of the power grid, dynamically adjust the power output of each high-energy-consuming load resource in the virtual power plant to achieve efficient load regulation.

2. The method for regulating high energy-consuming load resources based on a virtual power plant according to claim 1, characterized in that: In step 3, the control algorithm uses optimal control theory to perform scheduling optimization, and the optimal control equation is expressed as: Among them, L is the Lagrangian function, B is the bias term, P i (t) represents the power output of the i-th load resource at time t, z i (t) represents the switch state of the i-th load resource at time t, If z i (t)=1, load resource is turned on, If z i (t) = 0, the load resource is closed.

3. The method for regulating high energy-consuming load resources based on a virtual power plant according to claim 1, characterized in that: The scheduling algorithm uses mixed integer programming to optimize the start and stop operations of load resources. The constraints of the start and stop operations are expressed as: Among them, z i (t) represents the switch state of the i-th load resource at time t, If z i (t)=1, load resource is turned on, If z i (t) = 0, the load resource is closed.

4. The method for regulating high energy-consuming load resources based on a virtual power plant according to claim 1, characterized in that: The high-energy-consuming load resource regulation method includes analyzing the defined scheduling model using the variational method and optimal control theory to obtain the optimal scheduling strategy. The optimal scheduling strategy is expressed as follows by solving the optimization equation: P i (t)=f i (P previous (t),z i (t)), Among them, P previous (t) represents the power output of the previous time step, z i (t) represents the switch state of the i-th load resource at time t, f i (·) is the optimal dispatching strategy derived from the dynamic model of load regulation.

5. The method for regulating high energy-consuming load resources based on a virtual power plant according to claim 1, characterized in that: In step 5, the scheduling scheme is analyzed by Lyapunov stability theory to ensure that the system will not become unstable during the load regulation process. The Lyapunov function is defined as: Among them, P ref.i is the target power output, P i (t) represents the power output of the i-th load resource at time t, V(P1(t),…,P N (t)) represents the Lyapunov function, and N is the total number of high-energy-consuming load resources in the virtual power plant.

6. The method for regulating high energy-consuming load resources based on a virtual power plant according to claim 1, characterized in that: The high-energy-consuming load resource adjustment method solves the optimal scheduling by combining the gradient descent method with dynamic programming to obtain the optimal scheduling strategy for each load resource. The optimal scheduling strategy is expressed as: in, is the gradient of the Lagrangian function, δP i (t) represents the adjustment of the power output of the load resource at each time step, Represents all load resources i and time t.

7. The method for regulating high energy-consuming load resources based on a virtual power plant according to claim 1, characterized in that: The high-energy-consuming load resource adjustment method improves computing efficiency through parallel computing technology and performs parallel optimization on the scheduling of each load resource. The parallel optimization calculation process is performed by the following formula: in, As the overall optimization goal, is the parallel calculation of the i-th load resource at time t, N is the total number of high-energy-consuming load resources in the virtual power plant, P i (t) represents the power output of the i-th load resource at time t, Denotes the time point t.

8. The method for regulating high energy-consuming load resources based on a virtual power plant according to claim 1, characterized in that: The high-energy-consuming load resource adjustment method includes a real-time data acquisition and feedback control system that monitors grid load fluctuations in real time and dynamically adjusts the power output of each high-energy-consuming load resource in the virtual power plant based on the real-time load data. The dynamic adjustment process is controlled by the following formula: P i (t)=Adjust(P current (t),P target (t)), Among them, P current (t) is the power output of load resource i at the current moment, P target (t) is the target power output, Adjust(·) is the control function for adjusting the load output according to real-time data, P i (t) represents the power output of the i-th load resource at time t.

Images