Virtual power plant regulation system and method for controllable loads

Abstract

This invention relates to virtual power plants, specifically to a virtual power plant control system and method for controllable loads. The system comprises a data sensing layer that collects real-time load status, environmental parameters, and grid demand information for each controllable load within the virtual power plant; a model building layer that constructs an initial population for the Lüper Fox optimization algorithm based on the load status of each controllable load, and simultaneously constructs a virtual power plant control model based on environmental parameters and grid demand information; an optimization decision layer that uses the Lüper Fox optimization algorithm to solve the virtual power plant control model, generating an optimal virtual power plant control strategy; and a load control layer that converts the optimal virtual power plant control strategy into corresponding control commands and distributes them to each controllable load, achieving real-time optimized control of the virtual power plant. The technical solution provided by this invention effectively overcomes the shortcomings of existing technologies in accurately and efficiently optimizing and controlling controllable loads within a virtual power plant.

Description

Technical Field

[0001] This invention relates to virtual power plants, and more specifically to a virtual power plant control system and method for controllable loads. Background Technology

[0002] Among the many components of a virtual power plant, controllable loads play a crucial role. Controllable loads refer to electrical loads that can be flexibly adjusted according to grid demand or user settings, such as electric vehicle charging stations, smart home appliances, and interruptible industrial loads. By effectively regulating controllable loads, the optimal allocation of power resources can be achieved while meeting users' normal electricity needs, thereby improving the overall operational efficiency of the virtual power plant.

[0003] However, the regulation of virtual power plants faces numerous challenges. On the one hand, virtual power plants contain a large number of controllable loads of different types, with varying characteristics and operating states, making it difficult to accurately and in real-time collect and perceive their load status. On the other hand, the operation of virtual power plants is also affected by environmental parameters (such as temperature and illumination) and grid demand information (such as predicted load and electricity price data). The changes in these factors are uncertain and complex, posing significant difficulties for the regulation of virtual power plants.

[0004] Traditional virtual power plant control methods often employ fixed optimization algorithms or rules for decision-making, which prove inadequate when dealing with the complex and ever-changing operating environment of virtual power plants. For example, some optimization algorithms may focus too much on global search and neglect local optimization, resulting in low solution efficiency; while others may get stuck in local optima and fail to find a globally optimal control strategy. Furthermore, traditional methods often lack in-depth exploration and utilization of the characteristics of controllable loads, failing to fully realize their regulation potential. Summary of the Invention

[0005] (a) Technical problems to be solved

[0006] In view of the above-mentioned shortcomings of the existing technology, the present invention provides a virtual power plant control system and method for controllable load, which can effectively overcome the defects of the existing technology in that it is difficult to accurately and efficiently optimize and control the controllable load in the virtual power plant.

[0007] (II) Technical Solution

[0008] To achieve the above objectives, the present invention provides the following technical solution:

[0009] A virtual power plant control system with controllable load includes:

[0010] The data perception layer collects the load status, environmental parameters, and grid demand information of each controllable load in the virtual power plant in real time.

[0011] The model building layer constructs the initial population of the Lüper Fox optimization algorithm based on the load status of each controllable load, and constructs a virtual power plant control model based on environmental parameters and grid demand information.

[0012] To optimize the decision-making level, the Lüper Fox optimization algorithm is used to solve the virtual power plant control model and generate the optimal virtual power plant control strategy.

[0013] The load control layer transforms the optimal virtual power plant control strategy into corresponding control commands and sends them to each controllable load, thereby realizing real-time optimized control of the virtual power plant.

[0014] In the Lüper Fox optimization algorithm:

[0015] During the global exploration phase, the location of the Lüper fox was updated by simulating its daytime search for prey using a 260° wide-angle vision.

[0016] During the partial development phase, location updates were performed by simulating the Lüper fox's nighttime use of its 150° rotatable auditory location to locate prey.

[0017] By simulating the exponential decay of the Lüper fox's sense of smell with distance, a fitness weight is introduced. Individual Lüper foxes that maintain a low fitness weight for a period of time will trigger a random jumping mechanism to escape the local optimum.

[0018] Preferably, the data sensing layer collects in real time the load status, environmental parameters, and grid demand information of each controllable load within the virtual power plant, including:

[0019] Through smart meters and sensor networks, the load status of each controllable load, including industrial load power, electric vehicle charging and discharging status, and remaining power of energy storage systems, is collected in real time.

[0020] It collects environmental parameters, including weather station data, in real time, as well as power grid demand information, including predicted load and electricity price data.

[0021] Preferably, the model building layer constructs an initial population for the Lüper Fox optimization algorithm based on the load status of each controllable load, and simultaneously constructs a virtual power plant control model based on environmental parameters and grid demand information, including:

[0022] The initial population for the Lüper Fox optimization algorithm is constructed based on the load status of each controllable load.

[0023] The constraints for virtual power plant regulation are determined based on environmental parameters and grid demand information. At the same time, the objective function for virtual power plant regulation is determined, and a virtual power plant regulation model is constructed by combining the objective function and constraints.

[0024] Preferably, determining the objective function for virtual power plant regulation includes:

[0025] The objective function for virtual power plant regulation is expressed by the following formula:

[0026]

[0027] Where F is the objective function of virtual power plant regulation, C is the operating cost (i.e., the economic objective), S is the voltage fluctuation penalty term (i.e., the stability objective), ω1 and ω2 are weighting coefficients, and P... j The power setpoint for the controllable load j is generated by the Lüpertzian optimization algorithm, c j Let V be the unit power cost of controllable load j, N be the number of controllable loads, and V be the power cost per unit of controllable load j. k V is the actual voltage at node k. ref ΔV is the reference voltage, ΔV is the allowable voltage deviation, and M is the number of nodes.

[0028] Preferably, the optimization decision layer uses the Lüper Fox optimization algorithm to solve the virtual power plant control model and generate the optimal virtual power plant control strategy, including:

[0029] S1. Generate an initial population for the Lüper Fox optimization algorithm within the search space. Each Lüper Fox individual in the initial population represents a potential solution, and perform parameter initialization.

[0030] S2. Determine whether to enter the global exploration phase or the local development phase based on the number of iterations:

[0031] During the global exploration phase, the location of the Lüper fox was updated by simulating its daytime search for prey using a 260° wide-angle vision.

[0032] During the partial development phase, location updates were performed by simulating the Lüper fox's nighttime use of its 150° rotatable auditory location to locate prey.

[0033] S3. By simulating the characteristic of the Lüper fox's sense of smell decaying exponentially with distance, fitness weights are introduced, and the fitness weights of all Lüper fox individuals in the current population are calculated.

[0034] S4. Individuals of Lüpertz that maintain a low fitness weight for a period of time will trigger a random jump mechanism to escape the local optimum.

[0035] S5. Use the objective function F of the virtual power plant regulation to evaluate all individuals of the Lüper fox in the current population, calculate the corresponding fitness value, record and update the global optimal solution;

[0036] S6. Determine whether the iteration termination condition is met. If the iteration termination condition is not met, return to S2. Otherwise, take the current global optimal solution as the optimal virtual power plant control strategy.

[0037] Preferably, in S2, during the global exploration phase, the location is updated by simulating the Lüper fox using a 260° wide-angle visual search for prey during the day, including:

[0038] Simulates how the Lüper fox uses its 260° wide-angle vision to search for prey and update its location during the day:

[0039]

[0040] Among them, X i (t), X i (t+1) represent the positions of individual Lüper fox i at the t-th and t+1-th iterations, respectively, X best Let R(θ) be the position of the current global optimal solution, and R(θ) be a two-dimensional rotation matrix relative to the direction vector X. best -X i (t) is rotated to generate a new search direction, enhancing the global search capability. θ is a random rotation angle, simulating the randomness of the Lüper fox's 260° wide-angle vision. rand(0,1) is a random number uniformly distributed in the range [0,1]. α is the step size decay coefficient, balancing global exploration and local development. T is the maximum number of iterations. β is a random perturbation term. rand(-1,1) is a random number uniformly distributed in the range [-1,1]. Randomness enhances the robustness of the algorithm and prevents premature convergence.

[0041] Preferably, in the local development phase of S2, location updates are performed by simulating the Lüper fox's nighttime use of 150° rotatable auditory localization to locate prey, including:

[0042] Simulates how the Lüper fox uses its 150° rotatable auditory system to locate prey and update its position at night:

[0043]

[0044] Where u and v are both standard normal distribution random numbers, which generate the random step size of the Levy flight to simulate the uncertainty of the Lüper fox's 150° rotatable auditory localization, μ is the Levy exponent, which adjusts the shape of the Levy distribution and controls the jump frequency and amplitude of the search, and λ is the scaling factor, which controls the scaling ratio of the Levy flight step size.

[0045] Preferably, in S3, fitness weights are introduced by simulating the exponential decay of the Lüper fox's sense of smell with distance, and the fitness weights of all Lüper fox individuals in the current population are calculated, including:

[0046] Fitness weights are expressed by the following formula:

[0047]

[0048] Where, ω i(t) represents the fitness weight of individual Lüper fox i at the t-th iteration. X represents the position of individual Lüper fox i at the t-th iteration. i (t) and the position X of the current global optimal solution best The square of the Euclidean distance between them, σ is the decay rate parameter, balancing global exploration and local exploitation;

[0049] In S4, Lüper fox individuals that maintain a low fitness weight for a period of time will trigger a random jump mechanism to escape local optima, including:

[0050] The random jump mechanism is represented by the following formula:

[0051]

[0052] in, Let X be the position of individual Lüper fox i after jumping in the (t+1)th iteration. min X max These are the lower and upper bounds of the search space, respectively.

[0053] Preferably, the load control layer converts the optimal virtual power plant control strategy into corresponding control commands and distributes them to each controllable load, thereby realizing real-time optimized control of the virtual power plant, including:

[0054] The optimal virtual power plant control strategy is transformed into specific control commands, which are communicated with the load controller via the Modbus / IEC 104 protocol to issue control commands to each controllable load and receive feedback signals.

[0055] If the deviation between the actual power value of the controllable load and the power setpoint exceeds a preset threshold, the optimization decision layer is triggered to re-optimize using the Lüper Fox optimization algorithm to achieve real-time optimization and control of the virtual power plant.

[0056] A virtual power plant control method for controllable load includes the following steps:

[0057] S1. Real-time collection of load status, environmental parameters and grid demand information of each controllable load in the virtual power plant;

[0058] S2. Construct the initial population of the Lüper Fox optimization algorithm based on the load status of each controllable load, and construct a virtual power plant control model based on environmental parameters and power grid demand information.

[0059] S3. The Lüper Fox optimization algorithm is used to solve the virtual power plant control model and generate the optimal virtual power plant control strategy.

[0060] S4. Transform the optimal virtual power plant control strategy into corresponding control commands and send them to each controllable load to achieve real-time optimized control of the virtual power plant.

[0061] In the Lüper Fox optimization algorithm:

[0062] During the global exploration phase, the location of the Lüper fox was updated by simulating its daytime search for prey using a 260° wide-angle vision.

[0063] During the partial development phase, location updates were performed by simulating the Lüper fox's nighttime use of its 150° rotatable auditory location to locate prey.

[0064] By simulating the exponential decay of the Lüper fox's sense of smell with distance, a fitness weight is introduced. Individual Lüper foxes that maintain a low fitness weight for a period of time will trigger a random jumping mechanism to escape the local optimum.

[0065] (III) Beneficial Effects

[0066] Compared with the prior art, the virtual power plant control system and method for controllable load provided by the present invention has the following beneficial effects:

[0067] 1) Precise simulation and regulation to improve the accuracy of regulation.

[0068] The Lüper Fox optimization algorithm is adopted at the optimization decision level. By simulating the Lüper Fox's global search for prey during the day using a 260° wide-angle visual search and its local search for prey at night using a 150° rotatable auditory search, the algorithm can comprehensively and meticulously search the solution space. This unique search mechanism enables the algorithm to find the optimal solution from different angles and ranges, fully considering the complex characteristics of each controllable load in the virtual power plant, as well as the diversity of environmental parameters and grid demand information. This generates a more accurate optimal virtual power plant control strategy, effectively improving the accuracy of virtual power plant control and better meeting actual operation needs.

[0069] 2) Efficiently escape local optima and improve solution efficiency

[0070] By simulating the exponential decay of the Lüper fox's sense of smell with distance, a fitness weight is introduced, and a random jumping mechanism is triggered for Lüper fox individuals that maintain a low fitness weight for a period of time. This innovative design enables the algorithm to keenly detect situations where it gets trapped in local optima during the solution process, and to jump out of the local optimal solution region in time to continue searching for the global optimal solution in a wider solution space. Compared with the traditional optimization algorithm, which is prone to getting trapped in local optima, the Lüper fox optimization algorithm adopted in this invention greatly improves the solution efficiency, can find the optimal virtual power plant control strategy in a short time, reduce the time cost of control decision-making, and improve the real-time response capability of the virtual power plant.

[0071] 3) Adapt flexibly to complex environments and enhance system adaptability.

[0072] The data perception layer collects real-time load status, environmental parameters, and grid demand information for each controllable load within the virtual power plant, providing comprehensive and accurate data support for the model building layer and the optimization decision layer. Based on this real-time data, the model building layer constructs the initial population for the Lüper Fox optimization algorithm and the virtual power plant control model, ensuring the model accurately reflects the current operating status of the virtual power plant. The optimization decision layer, based on real-time data and the continuously updated model, dynamically generates the optimal virtual power plant control strategy using the Lüper Fox optimization algorithm. This dynamic control method based on real-time data enables the system to flexibly adapt to the uncertainty and complexity of environmental parameters and grid demand information during virtual power plant operation, enhancing the system's adaptability and stability under different operating conditions and ensuring the safe and efficient operation of the virtual power plant. Attached Figure Description

[0073] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without any creative effort.

[0074] Figure 1 This is a schematic diagram of the system of the present invention;

[0075] Figure 2 This is a schematic diagram of the process of using the Lüperfox optimization algorithm in this invention to solve the virtual power plant control model and generate the optimal virtual power plant control strategy. Detailed Implementation

[0076] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0077] The functional layered architecture of the virtual power plant control system for controllable load provided by this invention is described below with specific examples (e.g., Figure 1 As shown in the diagram, and considering the technical effects, the functional layered architecture includes:

[0078] The data perception layer collects the load status, environmental parameters, and grid demand information of each controllable load in the virtual power plant in real time.

[0079] The model building layer constructs the initial population of the Lüper Fox optimization algorithm based on the load status of each controllable load, and constructs a virtual power plant control model based on environmental parameters and grid demand information.

[0080] To optimize the decision-making level, the Lüper Fox optimization algorithm is used to solve the virtual power plant control model and generate the optimal virtual power plant control strategy.

[0081] The load control layer transforms the optimal virtual power plant control strategy into corresponding control commands and sends them to each controllable load, thereby achieving real-time optimized control of the virtual power plant.

[0082] The technical solution of this application, in the Lüperfox optimization algorithm:

[0083] During the global exploration phase, the location of the Lüper fox was updated by simulating its daytime search for prey using a 260° wide-angle vision.

[0084] During the partial development phase, location updates were performed by simulating the Lüper fox's nighttime use of its 150° rotatable auditory location to locate prey.

[0085] By simulating the exponential decay of the Lüper fox's sense of smell with distance, a fitness weight is introduced. Individual Lüper foxes that maintain a low fitness weight for a period of time will trigger a random jumping mechanism to escape the local optimum.

[0086] I. Data Perception Layer

[0087] The data perception layer collects in real time the load status, environmental parameters, and grid demand information of each controllable load within the virtual power plant, including:

[0088] Through smart meters and sensor networks, the load status of each controllable load, including industrial load power, electric vehicle charging and discharging status, and remaining power of energy storage systems, is collected in real time.

[0089] It collects environmental parameters, including weather station data, in real time, as well as power grid demand information, including predicted load and electricity price data.

[0090] II. Model Building Layer

[0091] The model building layer constructs an initial population for the Lüper Fox optimization algorithm based on the load status of each controllable load, and simultaneously constructs a virtual power plant control model based on environmental parameters and grid demand information, including:

[0092] The initial population for the Lüper Fox optimization algorithm is constructed based on the load status of each controllable load.

[0093] The constraints for virtual power plant regulation are determined based on environmental parameters and grid demand information. At the same time, the objective function for virtual power plant regulation is determined, and a virtual power plant regulation model is constructed by combining the objective function and constraints.

[0094] Specifically, the objective function for virtual power plant regulation is determined, including:

[0095] The objective function for virtual power plant regulation is expressed by the following formula:

[0096]

[0097] Where F is the objective function of virtual power plant regulation, C is the operating cost (i.e., the economic objective), S is the voltage fluctuation penalty term (i.e., the stability objective), ω1 and ω2 are weighting coefficients, and P... j The power setpoint for the controllable load j is generated by the Lüpertzian optimization algorithm, c j Let V be the unit power cost of controllable load j, N be the number of controllable loads, and V be the power cost per unit of controllable load j. k V is the actual voltage at node k. ref ΔV is the reference voltage, ΔV is the allowable voltage deviation, and M is the number of nodes.

[0098] III. Optimize the decision-making level

[0099] The optimization decision-making layer uses the Lüper Fox optimization algorithm to solve the virtual power plant control model and generate the optimal virtual power plant control strategy, such as... Figure 2 As shown, it includes:

[0100] S1. Generate an initial population for the Lüper Fox optimization algorithm within the search space. Each Lüper Fox individual in the initial population represents a potential solution, and perform parameter initialization.

[0101] S2. Determine whether to enter the global exploration phase or the local development phase based on the number of iterations:

[0102] During the global exploration phase, the location of the Lüper fox was updated by simulating its daytime search for prey using a 260° wide-angle vision.

[0103] During the partial development phase, location updates were performed by simulating the Lüper fox's nighttime use of its 150° rotatable auditory location to locate prey.

[0104] S3. By simulating the characteristic of the Lüper fox's sense of smell decaying exponentially with distance, fitness weights are introduced, and the fitness weights of all Lüper fox individuals in the current population are calculated.

[0105] S4. Individuals of Lüpertz that maintain a low fitness weight for a period of time will trigger a random jump mechanism to escape the local optimum.

[0106] S5. Use the objective function F of the virtual power plant regulation to evaluate all individuals of the Lüper fox in the current population, calculate the corresponding fitness value, record and update the global optimal solution;

[0107] S6. Determine whether the iteration termination condition is met. If the iteration termination condition is not met, return to S2. Otherwise, take the current global optimal solution as the optimal virtual power plant control strategy.

[0108] Specifically, in S2, during the global exploration phase, the location of the Lüper fox is updated by simulating its 260° wide-angle visual search for prey during the day, including:

[0109] Simulates how the Lüper fox uses its 260° wide-angle vision to search for prey and update its location during the day:

[0110]

[0111] Among them, X i (t), X i (t+1) represent the positions of individual Lüper fox i at the t-th and t+1-th iterations, respectively, X best Let R(θ) be the position of the current global optimal solution, and R(θ) be a two-dimensional rotation matrix relative to the direction vector X. best -X i (t) is rotated to generate a new search direction, enhancing the global search capability. θ is a random rotation angle, simulating the randomness of the Lüper fox's 260° wide-angle vision. rand(0,1) is a random number uniformly distributed in the range [0,1]. α is the step size decay coefficient, balancing global exploration and local development. T is the maximum number of iterations. β is a random perturbation term. rand(-1,1) is a random number uniformly distributed in the range [-1,1]. Randomness enhances the robustness of the algorithm and prevents premature convergence.

[0112] Specifically, in the partial development phase of S2, location updates are achieved by simulating the Lüper fox's ability to locate prey at night using its 150° rotatable auditory system, including:

[0113] Simulates how the Lüper fox uses its 150° rotatable auditory system to locate prey and update its position at night:

[0114]

[0115] Where u and v are both standard normal distribution random numbers, which generate the random step size of the Levy flight to simulate the uncertainty of the Lüper fox's 150° rotatable auditory localization, μ is the Levy exponent, which adjusts the shape of the Levy distribution and controls the jump frequency and amplitude of the search, and λ is the scaling factor, which controls the scaling ratio of the Levy flight step size.

[0116] Specifically, in S3, by simulating the exponential decay of the Lüper fox's sense of smell with distance, a fitness weight is introduced, and the fitness weight of all Lüper fox individuals in the current population is calculated, including:

[0117] Fitness weights are expressed by the following formula:

[0118]

[0119] Where, ω i (t) represents the fitness weight of individual Lüper fox i at the t-th iteration. X represents the position of individual Lüper fox i at the t-th iteration. i (t) and the position X of the current global optimal solution best The square of the Euclidean distance between them, σ is the decay rate parameter, which balances global exploration and local development.

[0120] Specifically, in S4, Lüper fox individuals that maintain a low fitness weight for a period of time will trigger a random jump mechanism to escape local optima, including:

[0121] The random jump mechanism is represented by the following formula:

[0122]

[0123] in, Let X be the position of individual Lüper fox i after jumping in the (t+1)th iteration. min X max These are the lower and upper bounds of the search space, respectively.

[0124] The aforementioned technical solution employs the Lüper Fox optimization algorithm at the decision-making level. By simulating the Lüper Fox's global search for prey during the day using a 260° wide-angle visual approach and its localized search for prey at night using a 150° rotatable auditory approach, the algorithm can comprehensively and meticulously search the solution space. This unique search mechanism enables the algorithm to find the optimal solution from different angles and ranges, fully considering the complex characteristics of each controllable load in the virtual power plant, as well as the diversity of environmental parameters and grid demand information. This generates a more accurate optimal virtual power plant control strategy, effectively improving the accuracy of virtual power plant control and better meeting actual operational needs.

[0125] By simulating the exponential decay of the Lüper fox's sense of smell with distance, a fitness weight is introduced, and a random jump mechanism is triggered for Lüper fox individuals that maintain a low fitness weight for a period of time. This innovative design enables the algorithm to keenly detect situations where it gets trapped in local optima during the solution process, and to jump out of the local optimum region in time to continue searching for the global optimum in a wider solution space. Compared with the traditional optimization algorithm, which is prone to getting trapped in local optima, the Lüper fox optimization algorithm adopted in this invention greatly improves the solution efficiency, can find the optimal virtual power plant control strategy in a short time, reduce the time cost of control decision-making, and improve the real-time response capability of the virtual power plant.

[0126] IV. Load Control Layer

[0127] The load control layer translates the optimal virtual power plant control strategy into corresponding control commands and distributes them to each controllable load, realizing real-time optimized control of the virtual power plant, including:

[0128] The optimal virtual power plant control strategy is transformed into specific control commands, which are communicated with the load controller via the Modbus / IEC 104 protocol to issue control commands to each controllable load and receive feedback signals.

[0129] If the deviation between the actual power value of the controllable load and the power setpoint exceeds a preset threshold, the optimization decision layer is triggered to re-optimize using the Lüper Fox optimization algorithm to achieve real-time optimization and control of the virtual power plant.

[0130] In this technical solution, the data perception layer collects the load status, environmental parameters, and grid demand information of each controllable load within the virtual power plant in real time, providing comprehensive and accurate data support for the model building layer and the optimization decision layer. Based on this real-time data, the model building layer constructs the initial population for the Lüper Fox optimization algorithm and the virtual power plant control model, enabling the model to accurately reflect the current operating status of the virtual power plant. The optimization decision layer, based on real-time data and the continuously updated model, uses the Lüper Fox optimization algorithm to dynamically generate the optimal virtual power plant control strategy. This dynamic control method based on real-time data allows the system to flexibly adapt to the uncertainty and complexity of environmental parameters and grid demand information during the operation of the virtual power plant, enhancing the system's adaptability and stability under different operating conditions and ensuring the safe and efficient operation of the virtual power plant.

[0131] Based on the aforementioned virtual power plant control system for controllable loads, this application also discloses a virtual power plant control method for controllable loads, comprising the following steps:

[0132] S1. Real-time collection of load status, environmental parameters and grid demand information of each controllable load in the virtual power plant;

[0133] S2. Construct the initial population of the Lüper Fox optimization algorithm based on the load status of each controllable load, and construct a virtual power plant control model based on environmental parameters and power grid demand information.

[0134] S3. The Lüper Fox optimization algorithm is used to solve the virtual power plant control model and generate the optimal virtual power plant control strategy.

[0135] S4. The optimal virtual power plant control strategy is converted into corresponding control commands and sent to each controllable load to realize real-time optimization control of the virtual power plant.

[0136] In the Lüper Fox optimization algorithm:

[0137] During the global exploration phase, the location of the Lüper fox was updated by simulating its daytime search for prey using a 260° wide-angle vision.

[0138] During the partial development phase, location updates were performed by simulating the Lüper fox's nighttime use of its 150° rotatable auditory location to locate prey.

[0139] By simulating the exponential decay of the Lüper fox's sense of smell with distance, a fitness weight is introduced. Individual Lüper foxes that maintain a low fitness weight for a period of time will trigger a random jumping mechanism to escape the local optimum.

[0140] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions will not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A virtual power plant control system with controllable load, characterized in that: include: The data perception layer collects the load status, environmental parameters, and grid demand information of each controllable load in the virtual power plant in real time. The model building layer constructs the initial population of the Lüper Fox optimization algorithm based on the load status of each controllable load, and constructs a virtual power plant control model based on environmental parameters and grid demand information. To optimize the decision-making level, the Lüper Fox optimization algorithm is used to solve the virtual power plant control model and generate the optimal virtual power plant control strategy. The load control layer transforms the optimal virtual power plant control strategy into corresponding control commands and sends them to each controllable load, thereby realizing real-time optimized control of the virtual power plant. In the Lüper Fox optimization algorithm: During the global exploration phase, the location of the Lüper fox was updated by simulating its daytime search for prey using a 260° wide-angle vision. During the partial development phase, location updates were performed by simulating the Lüper fox's nighttime use of its 150° rotatable auditory location to locate prey. By simulating the exponential decay of the Lüper fox's sense of smell with distance, a fitness weight is introduced. Individual Lüper foxes that maintain a low fitness weight for a period of time will trigger a random jumping mechanism to escape the local optimum.

2. The virtual power plant control system for controllable load according to claim 1, characterized in that: The data perception layer collects in real time the load status, environmental parameters, and grid demand information of each controllable load within the virtual power plant, including: Through smart meters and sensor networks, the load status of each controllable load, including industrial load power, electric vehicle charging and discharging status, and remaining power of energy storage systems, is collected in real time. It collects environmental parameters, including weather station data, in real time, as well as power grid demand information, including predicted load and electricity price data.

3. The virtual power plant control system for controllable load according to claim 1, characterized in that: The model building layer constructs an initial population for the Lüper Fox optimization algorithm based on the load status of each controllable load, and simultaneously constructs a virtual power plant control model based on environmental parameters and grid demand information, including: The initial population for the Lüper Fox optimization algorithm is constructed based on the load status of each controllable load. The constraints for virtual power plant regulation are determined based on environmental parameters and grid demand information. At the same time, the objective function for virtual power plant regulation is determined, and a virtual power plant regulation model is constructed by combining the objective function and constraints.

4. The virtual power plant control system for controllable load according to claim 3, characterized in that: The objective function for determining the virtual power plant control includes: The objective function for virtual power plant regulation is expressed by the following formula: Where F is the objective function of virtual power plant regulation, C is the operating cost (i.e., the economic objective), S is the voltage fluctuation penalty term (i.e., the stability objective), ω1 and ω2 are weighting coefficients, and P... j The power setpoint for the controllable load j is generated by the Lüpertzian optimization algorithm, c j Let V be the unit power cost of controllable load j, N be the number of controllable loads, and V be the power cost per unit of controllable load j. k V is the actual voltage at node k. ref ΔV is the reference voltage, ΔV is the allowable voltage deviation, and M is the number of nodes.

5. The virtual power plant control system for controllable load according to claim 4, characterized in that: The optimization decision layer uses the Lüper Fox optimization algorithm to solve the virtual power plant control model and generate the optimal virtual power plant control strategy, including: S1. Generate an initial population for the Lüper Fox optimization algorithm within the search space. Each Lüper Fox individual in the initial population represents a potential solution, and perform parameter initialization. S2. Determine whether to enter the global exploration phase or the local development phase based on the number of iterations: During the global exploration phase, the location of the Lüper fox was updated by simulating its daytime search for prey using a 260° wide-angle vision. During the partial development phase, location updates were performed by simulating the Lüper fox's nighttime use of its 150° rotatable auditory location to locate prey. S3. By simulating the characteristic of the Lüper fox's sense of smell decaying exponentially with distance, fitness weights are introduced, and the fitness weights of all Lüper fox individuals in the current population are calculated. S4. Individuals of Lüpertz that maintain a low fitness weight for a period of time will trigger a random jump mechanism to escape the local optimum. S5. Use the objective function F of the virtual power plant regulation to evaluate all individuals of Lüper fox in the current population, calculate the corresponding fitness value, record and update the global optimal solution; S6. Determine whether the iteration termination condition is met. If the iteration termination condition is not met, return to S2. Otherwise, take the current global optimal solution as the optimal virtual power plant control strategy.

6. The virtual power plant control system for controllable load according to claim 5, characterized in that: In S2, during the global exploration phase, the location of the Lüper fox is updated by simulating its 260° wide-angle visual search for prey during the day, including: Simulates how the Lüper fox uses its 260° wide-angle vision to search for prey and update its location during the day: Among them, X i (t), X i (t+1) represent the positions of individual Lüper fox i at the t-th and t+1-th iterations, respectively, X best Let R(θ) be the position of the current global optimal solution, and R(θ) be a two-dimensional rotation matrix relative to the direction vector X. best -X i (t) is rotated to generate a new search direction, enhancing the global search capability. θ is a random rotation angle, simulating the randomness of the Lüper fox's 260° wide-angle vision. rand(0,1) is a random number uniformly distributed in the range [0,1]. α is the step size decay coefficient, balancing global exploration and local development. T is the maximum number of iterations. β is a random perturbation term. rand(-1,1) is a random number uniformly distributed in the range [-1,1]. Randomness enhances the robustness of the algorithm and prevents premature convergence.

7. The virtual power plant control system for controllable load according to claim 6, characterized in that: In the partial development phase of S2, location updates were performed by simulating the Lüper fox's nighttime use of its 150° rotatable auditory targeting to locate prey, including: Simulates how the Lüper fox uses its 150° rotatable auditory system to locate prey and update its position at night: Where u and v are both standard normal distribution random numbers, which generate the random step size of the Levy flight to simulate the uncertainty of the Lüper fox's 150° rotatable auditory localization, μ is the Levy exponent, which adjusts the shape of the Levy distribution and controls the jump frequency and amplitude of the search, and λ is the scaling factor, which controls the scaling ratio of the Levy flight step size.

8. The virtual power plant control system for controllable load according to claim 7, characterized in that: In S3, by simulating the exponential decay of the Lüper fox's sense of smell with distance, a fitness weight is introduced, and the fitness weight of all Lüper fox individuals in the current population is calculated, including: Fitness weights are expressed by the following formula: Where, ω i (t) represents the fitness weight of individual Lüper fox i at the t-th iteration. X represents the position of individual Lüper fox i at the t-th iteration. i (t) and the position X of the current global optimal solution best The square of the Euclidean distance between them, σ is the decay rate parameter, balancing global exploration and local exploitation; In S4, Lüper fox individuals that maintain a low fitness weight for a period of time will trigger a random jump mechanism to escape local optima, including: The random jump mechanism is represented by the following formula: in, Let X be the position of individual Lüper fox i after jumping in the (t+1)th iteration. min X max These are the lower and upper bounds of the search space, respectively.

9. The virtual power plant control system for controllable load according to claim 1, characterized in that: The load control layer transforms the optimal virtual power plant control strategy into corresponding control commands and distributes them to each controllable load, thereby achieving real-time optimized control of the virtual power plant, including: The optimal virtual power plant control strategy is transformed into specific control commands, which are communicated with the load controller via the Modbus / IEC 104 protocol to issue control commands to each controllable load and receive feedback signals. If the deviation between the actual power value of the controllable load and the power setpoint exceeds a preset threshold, the optimization decision layer is triggered to re-optimize using the Lüper Fox optimization algorithm to achieve real-time optimization and control of the virtual power plant.

10. A virtual power plant control method for controllable loads, applied to the virtual power plant control system for controllable loads as described in claim 1, characterized in that: Includes the following steps: S1. Real-time collection of load status, environmental parameters and grid demand information of each controllable load in the virtual power plant; S2. Construct the initial population of the Lüper Fox optimization algorithm based on the load status of each controllable load, and construct a virtual power plant control model based on environmental parameters and power grid demand information. S3. The Lüper Fox optimization algorithm is used to solve the virtual power plant control model and generate the optimal virtual power plant control strategy. S4. Transform the optimal virtual power plant control strategy into corresponding control commands and send them to each controllable load to achieve real-time optimized control of the virtual power plant. In the Lüper Fox optimization algorithm: During the global exploration phase, the location of the Lüper fox was updated by simulating its daytime search for prey using a 260° wide-angle vision. During the partial development phase, location updates were performed by simulating the Lüper fox's nighttime use of its 150° rotatable auditory location to locate prey. By simulating the exponential decay of the Lüper fox's sense of smell with distance, a fitness weight is introduced. Individual Lüper foxes that maintain a low fitness weight for a period of time will trigger a random jumping mechanism to escape the local optimum.

Images