A method for collaborative optimization dispatching of virtual power plants participating in distribution network

Abstract

The present invention discloses a method for collaborative optimization and dispatching of a distribution network in which a virtual power plant participates, and relates to the technical field of collaborative optimization of distribution networks. The present invention adopts an autoregressive sliding average model to establish a prediction model, captures the time dependence and random volatility of historical load data through time series analysis, and dynamically corrects the prediction results in combination with ambient temperature and electricity price fluctuations, thereby improving the dispatching accuracy. Based on the predicted load, a multi-objective optimization dispatching model is constructed, and the optimization goals are to minimize energy consumption, minimize electricity cost and maximize operating efficiency. A genetic algorithm is used to solve and allocate the operating strategies of chillers and cooling towers, so that the optimal power dispatching is achieved while meeting the cooling demand. The system operation mode is optimized in combination with the start-stop constraints of the chiller and the wind speed regulation characteristics of the cooling tower, and a gradient descent algorithm is used to adjust the matching relationship between the cooling tower wind speed and the chiller load rate, thereby reducing power mismatch, improving the overall heat exchange efficiency and reducing energy consumption.

Description

Technical Field

[0001] The present invention relates to the technical field of coordinated optimization of distribution networks, and in particular to a method for coordinated optimization scheduling of distribution networks in which a virtual power plant participates. Background Art

[0002] With the increase in the penetration rate of new energy, virtual power plants (VPPs), as flexible resource aggregation platforms, play a key role in balancing supply and demand and optimizing energy management in distribution network dispatching. The existing VPP dispatching methods are difficult to fully tap the potential of adjustable loads in terms of deep optimization on the load side, making it difficult to further improve the overall energy efficiency level.

[0003] For example, in the cooling system of a large commercial building or industrial park, the operating status of the chiller and cooling tower directly affects the overall energy efficiency level. When optimizing such loads, traditional VPP scheduling methods often adopt a time period adjustment strategy based on load forecasting, or set the start and stop logic according to the equipment power. This method is difficult to dynamically adapt to real-time changes in external environmental factors, equipment operating status, and electricity price signals, resulting in insufficient coordinated optimization of chillers and cooling towers. Problems such as overload operation of some equipment, reduced heat exchange efficiency, or a sudden increase in power demand may occur, affecting the overall scheduling effect of the VPP.

[0004] At the same time, the utilization rate of VSC converter stations in AC and DC distribution networks is low, and they cannot be dynamically adjusted according to the fluctuations in renewable energy output, which reduces the overall energy utilization efficiency of the system. The power flow optimization calculations that traditional VPP scheduling relies on mostly use second-order cone programming (SOCP) or mixed integer nonlinear programming (MINLP), which have high computational complexity and are difficult to meet high-frequency scheduling requirements, affecting the system's real-time optimization capabilities. Therefore, a method for virtual power plants to participate in distribution network collaborative optimization scheduling is urgently needed to solve such problems. Summary of the invention

[0005] In view of the above existing problems, the present invention is proposed.

[0006] The present invention provides a method for collaborative optimization and dispatching of a virtual power plant participating in a distribution network to solve the problems of traditional solutions: insufficient optimization of the load-side cooling system, delayed dispatching of converter stations, slow response of the power purchase and sales market, high complexity of power flow calculation, and problems affecting the VPP dispatching efficiency.

[0007] In order to solve the above technical problems, the present invention provides the following technical solutions: The embodiment of the present invention provides a method for a virtual power plant to participate in the coordinated optimization scheduling of a distribution network, which comprises: Step S1, collecting the operating data of the internal cooling system of the virtual power plant VPP, obtaining the ambient temperature and electricity price fluctuation data, and predicting the cooling load in the future scheduling cycle based on time series analysis; Step S2, based on the cooling load obtained in step S1, construct a scheduling model to obtain a joint scheduling plan for the chiller and the cooling tower; Step S3, based on the joint scheduling scheme obtained in step S2, combined with the chiller start-stop constraints and the cooling tower wind speed adjustment characteristics, optimize the operation mode of the cooling system, and adjust the matching relationship between the cooling tower wind speed and the chiller load rate to obtain the optimal operation parameters of the cooling system; Step S4, based on the optimal operating parameters of the cooling system obtained in step S3, and in combination with the operating data of step S1 and the energy storage and adjustable load of the virtual power plant, a demand response strategy is formulated, and the demand response strategy is executed during the peak period of the power grid; Step S5, based on the demand response strategy obtained in step S4 and combined with the real-time electricity price of the power grid, optimize the operation mode of the cooling system under different electricity price levels.

[0008] As a preferred solution of the method for a virtual power plant to participate in the coordinated optimization and scheduling of a distribution network described in the present invention, the operating data includes the status of the chiller, the wind speed of the cooling tower, the chilled water temperature and the power load change data.

[0009] As a preferred solution of the method for a virtual power plant to participate in the coordinated optimization scheduling of a distribution network described in the present invention, the step of predicting the cooling load in the future scheduling period based on time series analysis is as follows: Assume that the cooling load historical data sequence is : , The normalized data is : , in, represents the historical data series of cooling load, Indicates The cooling load at the moment, Represents the total number of historical data points, represents the normalized cooling load value, The autoregressive moving average model is used for modeling, and the time series model is defined as: , in, is the number of autoregressive terms, is the autoregression coefficient, is the number of sliding average items, is the sliding average coefficient, is the random error term, is the index of the autoregressive term, is the index of the sliding average item, The ARMA model is used to predict the cooling load forecast value within the scheduling period. The prediction formula is: , in, represents the prediction step length, Indicates the predicted time The cooling load value.

[0010] As an optimal solution of the method for collaborative optimization and scheduling of a distribution network involving a virtual power plant described in the present invention, the scheduling model aims to minimize energy consumption, minimize electricity cost and maximize operating efficiency, and adopts an optimization algorithm to solve and obtain a joint scheduling plan for chillers and cooling towers.

[0011] As a preferred solution of the method for collaborative optimization and dispatching of a distribution network in which a virtual power plant participates in the present invention, the steps of obtaining a joint dispatching scheme of a chiller and a cooling tower by using an optimization algorithm with the goal of minimizing energy consumption, minimizing electricity cost and maximizing operating efficiency are as follows: Construct the objective function and define it as: , in, To optimize the target value, is the total energy consumption of the cooling system, is the total electricity cost of the cooling system, For the overall operating efficiency of the cooling system, is the energy consumption weight coefficient, is the electricity cost weight coefficient, is the operating efficiency weight coefficient, Calculate energy consumption, including: The total energy consumption of the chiller is calculated as follows: , The total energy consumption of the cooling tower is calculated as follows: , in, is the total energy consumption of the chiller, For the moment The power of the chiller, is the total energy consumption of the cooling tower, For the moment Cooling tower power, is the time step, is the total number of time steps in the scheduling cycle, The total electricity cost for the cooling system is calculated as: , in, For the moment The electricity price, Calculate the operating efficiency using the following formula: , in, For the moment The cooling system's cooling output, Genetic algorithm is used for optimization and solution. The genetic algorithm steps include: Randomly generate multiple scheduling plans for chillers and cooling towers, Calculate the objective function value of each solution, Select the better individuals according to the fitness value. Perform crossover operation on the selected individuals. Random mutation of some individuals. Repeat until convergence.

[0012] As a preferred solution of the method for collaborative optimization and dispatching of a distribution network in which a virtual power plant participates in the present invention, the step of optimizing the operation mode of the cooling system by combining the start-stop constraints of the chiller and the wind speed adjustment characteristics of the cooling tower, and adjusting the matching relationship between the wind speed of the cooling tower and the load rate of the chiller to obtain the optimal operation parameters of the cooling system is as follows: Assume that the start and stop state variables of the chiller are : , in, Indicates time The operating status of the chiller. Indicates opening, Indicates closing. To avoid frequent starts and stops, the chiller needs to meet the minimum operating time bundle: , in, Indicates that the chiller is at time The start and stop signal variables, Represents startup, Represents hold or close, The minimum continuous operation time of the chiller, Assume the cooling tower wind speed is : , in, For the moment Cooling tower wind speed, Minimum cooling tower wind speed, The maximum wind speed of the cooling tower is is the wind speed adjustment ratio variable, Assume that the load rate of the chiller is : , in, For the moment The load factor of the chiller, The maximum power output of the chiller, Optimize wind speed adjustment ratio variable , the optimization goal is: , in, To match the scale factor, The gradient descent algorithm is used to update the wind speed adjustment variable, and the update formula is: , in, is the number of iteration steps, is the learning rate, is the matching error objective function.

[0013] As a preferred solution of the method for collaborative optimization and scheduling of a distribution network in which a virtual power plant participates in the present invention, the demand response strategy includes adjusting the cooling tower wind speed, reducing the power of the chiller and increasing the energy storage output power.

[0014] As a preferred solution of the method for participating in the coordinated optimization and dispatching of distribution networks by a virtual power plant described in the present invention, the step of formulating a demand response strategy in combination with the operation data of step S1 and the energy storage and adjustable load of the virtual power plant, and executing the demand response strategy during the peak period of the power grid is as follows: To model the demand response constraint, the time interval of the power grid peak period is expressed as : , in, Indicates that the electricity price is higher than the average electricity price Peak hour collection, is the average electricity price during the entire dispatch period, To dispatch the energy storage system, assume that the power output of the energy storage system is: , in, For the moment The power output of the energy storage system, The maximum output power of the energy storage system, is the discharge ratio variable of the energy storage system, The goal is to minimize the total electrical power used during peak hours, defined as: , The quadratic programming optimization method is used to solve the problem, and the constraints are: .

[0015] As a preferred solution of the method for a virtual power plant to participate in the coordinated optimization and dispatching of a distribution network described in the present invention, in step S5, the operation mode of the optimized cooling system at different electricity price levels includes: Based on electricity price forecasts, the cooling system operating power is increased during periods of low electricity prices. Reduce cooling system operating power during periods of high electricity prices.

[0016] As a preferred solution of the method for collaborative optimization and dispatching of a distribution network in which a virtual power plant participates in the present invention, the step of optimizing the operation mode of the cooling system under different electricity price levels based on the demand response strategy obtained in step S4 and in combination with the real-time electricity price of the power grid is as follows: During low electricity price periods , adopt high power operation strategy: , in, Indicates the low electricity price period when the electricity price is lower than the average electricity price. Set value for high power operation, During high electricity price periods , using a low power operation strategy: , in, Set value for low power operation, During peak electricity price periods In the system, the energy storage system releases electricity first: .

[0017] The beneficial effects of the present invention are: The present invention adopts an autoregressive moving average ARMA model to establish a prediction model, captures the time dependence and random volatility of historical load data through time series analysis, and dynamically corrects the prediction results in combination with ambient temperature and electricity price fluctuations, thereby improving the scheduling accuracy. Based on the predicted load, a multi-objective optimization scheduling model is constructed, with minimum energy consumption, minimum electricity cost and maximum operating efficiency as the optimization goals. A genetic algorithm is used to solve and allocate the operating strategies of the chiller and the cooling tower, so that the optimal power scheduling can be achieved while meeting the cooling demand.

[0018] The present invention combines the start-stop constraints of the chiller and the wind speed regulation characteristics of the cooling tower to optimize the system operation mode, and adopts a gradient descent algorithm to adjust the matching relationship between the cooling tower wind speed and the load rate of the chiller, thereby reducing power mismatch, improving the overall heat exchange efficiency, and reducing energy consumption.

[0019] In terms of demand response, the present invention combines the energy storage system and the adjustable load to optimize the power adjustment strategy of the cooling system. During the peak load period of the power grid, the power of the chiller is reduced, the wind speed of the cooling tower is adjusted, and the stored energy is released to reduce the peak power consumption of the power grid and enhance the flexibility of the virtual power plant. At the same time, in response to electricity price fluctuations, the cooling system operation mode is adjusted based on electricity price forecasts, a high-power operation strategy is adopted during low electricity price periods to improve energy utilization, and the operating power is reduced during high electricity price periods. In addition, the energy storage system charge and discharge optimization strategy is combined to reduce the overall electricity purchase cost.

[0020] In summary, the present invention enables the chiller and cooling tower to be dynamically adjusted under different load and electricity price conditions, enhances the adaptability of the AC / DC distribution network and the efficiency of new energy utilization, and improves the overall operating performance of the virtual power plant. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0022] Figure 1 It is a flow chart of the method for collaborative optimization and dispatching of distribution networks in which a virtual power plant participates in the present invention. DETAILED DESCRIPTION

[0023] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0024] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0025] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0026] Example 1, reference Figure 1 This embodiment provides a method for a virtual power plant to participate in the coordinated optimization scheduling of a distribution network, comprising the following steps: Step S1, collecting the operating data of the internal cooling system of the virtual power plant VPP, obtaining the ambient temperature and electricity price fluctuation data, and predicting the cooling load in the future scheduling cycle based on time series analysis; The operation data includes chiller status, cooling tower wind speed, chilled water temperature and power load change data; The step of predicting the cooling load in the future scheduling period based on time series analysis is: Assume that the cooling load historical data sequence is : , The normalized data is : , in, represents the historical data series of cooling load, Indicates The cooling load at the moment, Represents the total number of historical data points, represents the normalized cooling load value, The autoregressive moving average model is used for modeling, and the time series model is defined as: , in, is the number of autoregressive terms, is the autoregression coefficient, is the number of sliding average items, is the sliding average coefficient, is the random error term, is the index of the autoregressive term, is the index of the sliding average item, The ARMA model is used to predict the cooling load forecast value within the scheduling period. The prediction formula is: , in, represents the prediction step length, Indicates the predicted time The cooling load value.

[0027] Specifically, in this step, the cooling load is predicted by the time series analysis method; the autoregressive moving average ARMA model is used to establish a cooling load prediction model, and the time dependence and random volatility of the data are captured by the autoregressive term and the moving average term to predict the cooling load of the future scheduling cycle; Dynamically adjust the prediction results according to influencing factors such as ambient temperature and electricity price fluctuations to improve the scheduling accuracy of the cooling system; Step S2, based on the cooling load obtained in step S1, construct a scheduling model to obtain a joint scheduling plan for the chiller and the cooling tower; The scheduling model aims to minimize energy consumption, minimize electricity cost and maximize operating efficiency, and adopts an optimization algorithm to solve and obtain a joint scheduling plan for the chiller and the cooling tower.

[0028] The steps of obtaining the joint scheduling scheme of the chiller and the cooling tower by using the optimization algorithm to minimize energy consumption, minimize electricity cost and maximize operating efficiency are as follows: Construct the objective function and define it as: , in, To optimize the target value, is the total energy consumption of the cooling system, is the total electricity cost of the cooling system, For the overall operating efficiency of the cooling system, is the energy consumption weight coefficient, is the electricity cost weight coefficient, is the operating efficiency weight coefficient, Calculate energy consumption, including: The total energy consumption of the chiller is calculated as follows: , The total energy consumption of the cooling tower is calculated as follows: , in, is the total energy consumption of the chiller, For the moment The power of the chiller, is the total energy consumption of the cooling tower, For the moment Cooling tower power, is the time step, is the total number of time steps in the scheduling cycle, The total electricity cost for the cooling system is calculated as: , in, For the moment The electricity price, Calculate the operating efficiency using the following formula: , in, For the moment The cooling system's cooling output, Genetic algorithm is used for optimization and solution. The genetic algorithm steps include: Randomly generate multiple scheduling plans for chillers and cooling towers, Calculate the objective function value of each solution, Select the better individuals according to the fitness value. Perform crossover operation on the selected individuals. Random mutation of some individuals. Repeat until convergence.

[0029] Specifically, in this step, an optimization scheduling model is constructed, with minimum energy consumption, minimum electricity cost and maximum operating efficiency as the optimization goals, and the energy consumption, electricity cost and operating efficiency of the cooling system are comprehensively considered to achieve multi-objective optimization; The energy consumption calculation part is to accumulate the power consumption of the chiller and cooling tower respectively, and the electricity cost part is combined with the dynamic changes of electricity prices; in addition, the operating efficiency is measured by the ratio of cooling output to total energy consumption to ensure that the cooling system maintains a high efficiency state; finally, the genetic algorithm is used for optimization and solution. Through the steps of population initialization, fitness calculation, selection, crossover and mutation, the optimal scheduling scheme is gradually approached, which can effectively coordinate the load distribution of the chiller and cooling tower; Step S3, based on the joint scheduling scheme obtained in step S2, combined with the chiller start-stop constraints and the cooling tower wind speed adjustment characteristics, optimize the operation mode of the cooling system, and adjust the matching relationship between the cooling tower wind speed and the chiller load rate to obtain the optimal operation parameters of the cooling system; The step of optimizing the operation mode of the cooling system by combining the start-stop constraint of the chiller and the wind speed regulation characteristics of the cooling tower, and adjusting the matching relationship between the wind speed of the cooling tower and the load rate of the chiller to obtain the optimal operation parameters of the cooling system is as follows: Assume that the start and stop state variables of the chiller are : , in, Indicates time The operating status of the chiller. Indicates opening, Indicates closing. To avoid frequent starts and stops, the chiller needs to meet the minimum operating time bundle: , in, Indicates that the chiller is at time The start and stop signal variables, Represents startup, Represents hold or close, The minimum continuous operation time of the chiller, Assume the cooling tower wind speed is : , in, For the moment Cooling tower wind speed, Minimum cooling tower wind speed, The maximum wind speed of the cooling tower is is the wind speed adjustment ratio variable, Assume that the load rate of the chiller is : , in, For the moment The load factor of the chiller, The maximum power output of the chiller, Optimize wind speed adjustment ratio variable , the optimization goal is: , in, To match the scale factor, The gradient descent algorithm is used to update the wind speed adjustment variable, and the update formula is: , in, is the number of iteration steps, is the learning rate, is the matching error objective function.

[0030] Specifically, the start and stop status of the chiller is subject to the minimum continuous operation time constraint. To avoid damage to the equipment caused by frequent start and stop, the cooling tower wind speed adopts a linear adjustment strategy to ensure the corresponding cooling capacity under different load conditions. In addition, in order to optimize the matching relationship between the load rate of the chiller and the cooling tower wind speed, an objective function that minimizes the load error is constructed, and a gradient descent algorithm is used for optimization and adjustment, which effectively reduces the power mismatch problem between the chiller and the cooling tower, improves the overall operating efficiency of the cooling system, and further reduces energy consumption. Step S4, based on the optimal operating parameters of the cooling system obtained in step S3, and in combination with the operating data of step S1 and the energy storage and adjustable load of the virtual power plant, a demand response strategy is formulated, and the demand response strategy is executed during the peak period of the power grid; The demand response strategy includes adjusting the cooling tower wind speed, reducing the chiller power and increasing the energy storage output power.

[0031] The step of formulating a demand response strategy in combination with the operation data of step S1 and the energy storage and adjustable load of the virtual power plant, and executing the demand response strategy during the peak period of the power grid is as follows: To model the demand response constraint, the time interval of the power grid peak period is expressed as : , in, Indicates that the electricity price is higher than the average electricity price Peak hour collection, is the average electricity price during the entire dispatch period, To dispatch the energy storage system, assume that the power output of the energy storage system is: , in, For the moment The power output of the energy storage system, The maximum output power of the energy storage system, is the discharge ratio variable of the energy storage system, The goal is to minimize the total electrical power used during peak hours, defined as: , The quadratic programming optimization method is used to solve the problem, and the constraints are: .

[0032] Specifically, here, by optimizing the discharge strategy of the energy storage system, it can provide additional power during peak hours, thereby reducing the power demand of the cooling system on the grid; optimizing the operation strategy of the chiller and cooling tower during peak hours to minimize the load on the grid, which can effectively alleviate the contradiction between power supply and demand during peak hours, improve the scheduling flexibility of the virtual power plant, and reduce the overall operating cost; Step S5, based on the demand response strategy obtained in step S4 and combined with the real-time electricity price of the power grid, optimizing the operation mode of the cooling system under different electricity price levels; In step S5, the operation mode of optimizing the cooling system under different electricity price levels includes: Based on electricity price forecasts, the cooling system operating power is increased during periods of low electricity prices. Reduce cooling system operating power during periods of high electricity prices.

[0033] The step of optimizing the operation mode of the cooling system under different electricity price levels based on the demand response strategy obtained in step S4 and combining the real-time electricity price of the power grid is as follows: During low electricity price periods , adopt high power operation strategy: , in, Indicates the low electricity price period when the electricity price is lower than the average electricity price. Set value for high power operation, During high electricity price periods , using a low power operation strategy: , in, Set value for low power operation, During peak electricity price periods In the system, the energy storage system releases electricity first: .

[0034] Specifically, during periods of low electricity prices, the system adopts a high-power operation mode to maximize the use of low-cost electricity and improve the operating efficiency of the cooling system; during periods of high electricity prices, the system reduces power operation and combines with the energy storage system to release stored electricity to reduce electricity purchase costs; in addition, by dynamically adjusting the charging and discharging strategy of the energy storage system, it charges during periods of low electricity prices and discharges during periods of high electricity prices, thereby smoothing the power consumption curve of the cooling system and reducing electricity cost fluctuations.

[0035] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A method for collaborative optimization and dispatching of a distribution network in which a virtual power plant participates, characterized in that: include, Step S1, collecting the operating data of the internal cooling system of the virtual power plant VPP, obtaining the ambient temperature and electricity price fluctuation data, and predicting the cooling load in the future scheduling cycle based on time series analysis; Step S2, based on the cooling load obtained in step S1, construct a scheduling model to obtain a joint scheduling plan for the chiller and the cooling tower; Step S3, based on the joint scheduling scheme obtained in step S2, combined with the chiller start-stop constraints and the cooling tower wind speed adjustment characteristics, optimize the operation mode of the cooling system, and adjust the matching relationship between the cooling tower wind speed and the chiller load rate to obtain the optimal operation parameters of the cooling system; Step S4, based on the optimal operating parameters of the cooling system obtained in step S3, and in combination with the operating data of step S1 and the energy storage and adjustable load of the virtual power plant, a demand response strategy is formulated, and the demand response strategy is executed during the peak period of the power grid; Step S5, based on the demand response strategy obtained in step S4 and combined with the real-time electricity price of the power grid, optimize the operation mode of the cooling system under different electricity price levels.

2. A method for collaborative optimization and dispatching of a virtual power plant participating in a distribution network according to claim 1, characterized in that: The operating data includes chiller status, cooling tower wind speed, chilled water temperature and power load change data.

3. A method for collaborative optimization and dispatching of a virtual power plant participating in a distribution network as claimed in claim 2, characterized in that: The step of predicting the cooling load in the future scheduling period based on time series analysis is: Assume that the cooling load historical data sequence is : ， The normalized data is : ， in, represents the historical data series of cooling load, Indicates The cooling load at the moment, Represents the total number of historical data points, represents the normalized cooling load value, The autoregressive moving average model is used for modeling, and the time series model is defined as: ， in, is the number of autoregressive terms, is the autoregression coefficient, is the number of sliding average items, is the sliding average coefficient, is the random error term, is the index of the autoregressive term, is the index of the sliding average item, The ARMA model is used to predict the cooling load forecast value within the scheduling period. The prediction formula is: ， in, represents the prediction step length, Indicates the predicted time The cooling load value.

4. A method for collaborative optimization and dispatching of a virtual power plant participating in a distribution network as claimed in claim 3, characterized in that: The scheduling model aims to minimize energy consumption, minimize electricity cost and maximize operating efficiency, and adopts an optimization algorithm to obtain a joint scheduling plan for the chiller and the cooling tower.

5. A method for collaborative optimization and dispatching of a virtual power plant participating in a distribution network as claimed in claim 4, characterized in that: The steps of obtaining the joint scheduling scheme of the chiller and the cooling tower by using the optimization algorithm to minimize energy consumption, minimize electricity cost and maximize operating efficiency are as follows: Construct the objective function and define it as: ， in, To optimize the target value, is the total energy consumption of the cooling system, is the total electricity cost of the cooling system, For the overall operating efficiency of the cooling system, is the energy consumption weight coefficient, is the electricity cost weight coefficient, is the operating efficiency weight coefficient, Calculate energy consumption, including: The total energy consumption of the chiller is calculated as follows: ， The total energy consumption of the cooling tower is calculated as follows: ， in, is the total energy consumption of the chiller, For the moment The power of the chiller, is the total energy consumption of the cooling tower, For the moment Cooling tower power, is the time step, is the total number of time steps in the scheduling cycle, The total electricity cost for the cooling system is calculated as: ， in, For the moment The electricity price, Calculate the operating efficiency using the following formula: ， in, For the moment The cooling system's cooling output, Genetic algorithm is used for optimization and solution. The genetic algorithm steps include: Randomly generate multiple scheduling plans for chillers and cooling towers, Calculate the objective function value of each solution, Select the better individuals according to the fitness value. Perform crossover operation on the selected individuals. Random mutation of some individuals. Repeat until convergence.

6. A method for collaborative optimization and dispatching of a virtual power plant participating in a distribution network as claimed in claim 5, characterized in that: The step of optimizing the operation mode of the cooling system by combining the start-stop constraint of the chiller and the wind speed regulation characteristics of the cooling tower, and adjusting the matching relationship between the wind speed of the cooling tower and the load rate of the chiller to obtain the optimal operation parameters of the cooling system is as follows: Assume that the start and stop state variables of the chiller are : ， in, Indicates time The operating status of the chiller. Indicates opening, Indicates closing, To avoid frequent starts and stops, the chiller needs to meet the minimum operating time bundle: ， in, Indicates that the chiller is at time The start and stop signal variables, Represents startup, Represents hold or close, The minimum continuous operation time of the chiller, Assume the cooling tower wind speed is : ， in, For the moment Cooling tower wind speed, Minimum cooling tower wind speed, The maximum wind speed of the cooling tower is is the wind speed adjustment ratio variable, Assume that the load rate of the chiller is : ， in, For the moment The load factor of the chiller, The maximum power output of the chiller, Optimize wind speed adjustment ratio variable , the optimization goal is: ， in, To match the scale factor, The gradient descent algorithm is used to update the wind speed adjustment variable, and the update formula is: ， in, is the number of iteration steps, is the learning rate, is the matching error objective function.

7. A method for virtual power plant participating in coordinated optimization and dispatching of distribution network as claimed in claim 6, characterized in that: The demand response strategy includes adjusting the cooling tower wind speed, reducing the chiller power and increasing the energy storage output power.

8. A method for virtual power plant participating in distribution network coordinated optimization scheduling as claimed in claim 7, characterized in that: The step of formulating a demand response strategy in combination with the operation data of step S1 and the energy storage and adjustable load of the virtual power plant, and executing the demand response strategy during the peak period of the power grid is as follows: To model the demand response constraint, the time interval of the power grid peak period is expressed as : ， in, Indicates that the electricity price is higher than the average electricity price Peak hour collection, is the average electricity price during the entire dispatch period, To dispatch the energy storage system, assume that the power output of the energy storage system is: ， in, For the moment The power output of the energy storage system, The maximum output power of the energy storage system, is the discharge ratio variable of the energy storage system, The goal is to minimize the total electrical power used during peak hours, defined as: ， The quadratic programming optimization method is used to solve the problem, and the constraints are: 。 9. A method for virtual power plant participating in distribution network coordinated optimization scheduling as claimed in claim 8, characterized in that: In step S5, the operation mode of optimizing the cooling system under different electricity price levels includes: Based on electricity price forecasts, the cooling system operating power is increased during periods of low electricity prices. Reduce cooling system operating power during periods of high electricity prices.

10. A method for virtual power plant participating in distribution network coordinated optimization scheduling as claimed in claim 9, characterized in that: The step of optimizing the operation mode of the cooling system under different electricity price levels based on the demand response strategy obtained in step S4 and combining the real-time electricity price of the power grid is as follows: During low electricity price periods , adopt high power operation strategy: ， in, Indicates the low electricity price period when the electricity price is lower than the average electricity price. Set value for high power operation, During high electricity price periods , using a low power operation strategy: ， in, Set value for low power operation, During peak electricity price periods In the system, the energy storage system releases electricity first: 。

Images