Method and device for optimizing virtual power plant scheduling model and readable storage medium

Abstract

The invention provides a method and device for optimizing a virtual power plant scheduling model and a readable storage medium, and the method comprises the steps: building a virtual power plant operation income objective function based on the comprehensive satisfaction of a user in a wind power combined output scene, and obtaining a comprehensive optimal decision through a particle swarm optimization algorithm. According to the method, modeling is carried out on the uncertainty of wind and light output, the Mahalanobis distance is adopted as measurement, the number of scenes of wind and light combined output is reduced by adopting an improved iterative self-organizing data analysis algorithm, and covariance of data is considered, so that a clustering result can better reflect the internal relation of the data; and the accuracy and reliability of wind and light output uncertainty modeling are improved. Meanwhile, according to the method, a comprehensive satisfaction model of the user is constructed from the two dimensions of economy and low-carbon property, the double requirements of the user on economic cost and environmental protection concept are fully considered, and the recognition degree of the user on the virtual power plant service is improved.

Description

Technical Field

[0001] The present application relates to the technical field of power plant resource scheduling, and in particular to a method, device and readable storage medium for optimizing a virtual power plant scheduling model. Background Art

[0002] Virtual power plants (VPPs) face many challenges in their actual operation and optimal scheduling. Among them, wind and solar energy output is significantly affected by natural conditions and has a high degree of uncertainty. Traditional methods for dealing with wind and solar output uncertainty are difficult to meet actual needs in terms of accuracy and efficiency.

[0003] In terms of user satisfaction, previous studies on virtual power plant scheduling have mostly only started from a single dimension, such as focusing only on the user's electricity cost and ignoring other important factors. From an economic perspective alone, it is impossible to fully measure the actual user satisfaction, resulting in a deviation between the virtual power plant scheduling plan and the user's real needs, affecting the user's enthusiasm and initiative to participate in the virtual power plant.

[0004] Therefore, how to improve the accuracy and efficiency of obtaining the uncertainty of wind and solar power output while meeting user needs and obtaining the optimal scheduling decision for VPP has become a problem that needs to be solved. Summary of the invention

[0005] The embodiment of the present application provides a method, device and readable storage medium for optimizing a virtual power plant scheduling model, which can improve the accuracy and reliability of modeling uncertainty of wind and solar power outputs while considering the dual requirements of economic cost and environmental protection concept, and obtain the comprehensive optimal decision for VPP scheduling. The technical solution is as follows: In the first aspect, a method for optimizing a virtual power plant dispatching model is provided, including: based on the characteristics of wind and solar power output uncertainty, establishing a wind and solar power output uncertainty model, the wind and solar power output uncertainty model is used to obtain the wind and solar power output uncertainty through an iterative self-organizing data analysis algorithm ISODATA wind power joint output scenario, the is a positive integer; based on the needs of economic cost and environmental protection concept, a user demand model is established, and the user demand model is used to obtain the comprehensive user satisfaction; based on the wind and solar output uncertainty model and the user demand model, a virtual power plant scheduling model is established, and the virtual power plant scheduling model is used to solve the The optimal decision considering the user's comprehensive satisfaction in a wind power combined output scenario is made; the particle swarm algorithm is used to calculate the optimal solution of the virtual power plant scheduling model, and the optimal solution is determined as the scheduling decision of the virtual power plant.

[0006] In combination with the first aspect, the wind and solar power output uncertainty model includes a wind and solar power output probability density function And wind and solar output joint distribution function ,Should The calculation formula is: ; Where: The samples collected during the sampling period are wind and solar power output. is the i-th sample value of wind and solar power output, n is the number of historical days of wind and solar power output data, is the bandwidth, i is a positive integer; The calculation formula is: ; Where: and From the ,Should The wind power output is the sample collected during the sampling period. is the cumulative distribution function of wind power output, is the sample collected by the photoelectric output during the sampling period, is the cumulative distribution function of photovoltaic output; For With the The relevance of It is in the range of -1 to 1 and is not equal to 0.

[0007] In combination with the first aspect, in some implementations of the first aspect, the iterative self-organizing data analysis algorithm ISODATA includes the following calculation steps: Step (101). Obtain M samples , determine the initial cluster center and record the Expected minimum distance , i is in the range of 1 to M, and M is a positive integer; Step (102). Calculate the M The probability of being selected as the next cluster center , based on this Determine The cluster center is denoted as ,Should is a positive integer, the The calculation formula is: ; Step (103). Calculate each To the cluster center The Mahalanobis distance , each of the Assign to this shortest among The corresponding cluster In and the The calculation formulas are: ; ; Where: is the weight matrix, ; If the The number of samples in is less than the minimum number of samples , then remove the and the The corresponding , the In Assign to the remaining cluster centers and shortest among The corresponding cluster In this paper, based on the In Recalculate the ; Step (104). Calculate the The maximum variance in and standard deviation , when the Greater than , or the number of clusters currently in the cluster satisfy When the number of iterations is an odd number, a split operation is performed. Split out new cluster centers and , the number of cluster centers Add 1, the specific calculation formula for the split operation is: ; Step (105). Calculate the This When the Mahalanobis distance between two cluster centers is less than the threshold, or satisfy When the number of iterations is even, a merge operation is performed to merge the two cluster centers into a new cluster center. , the number of cluster centers Subtract 1, the specific calculation formula for the merge operation is: ( ); In the formula: the left side of the equation is the new cluster center, and the right side of the equation With the are the two cluster centers, the and the All are sample sets; Step (106). Repeat the iterative calculation from step (103) to step (105) until the number of iterative calculations reaches the maximum number of iterations, and then stop the calculation. Determined as the current number of cluster centers .

[0008] In combination with the first aspect, in some implementations of the first aspect, the user demand model includes an economic indicator and a low-carbon indicator, and the user comprehensive satisfaction is constructed by linearly weighting the economic indicator and the low-carbon indicator.

[0009] In combination with the first aspect, in some implementations of the first aspect, the economic indicator includes policy subsidy benefits, response comfort loss, response production efficiency loss and electricity cost, and the low-carbon indicator includes carbon dioxide emission reduction, sulfur dioxide emission reduction and nitrogen oxide emission reduction.

[0010] In combination with the first aspect, in some implementations of the first aspect, the virtual power plant scheduling model includes a virtual power plant operation revenue objective function ,Should The optimal solution is in the The optimal decision considering the comprehensive satisfaction of the user in the scenario of wind power combined output is The calculation formula is: ; In the formula, The revenue from the purchase and sale of electricity for the virtual power plant; The operating cost of the gas turbine includes power generation cost, operating cost and downtime cost; For in this The operating cost of the energy storage system in a wind power combined output scenario, including the charging and discharging costs at each time period; for demand response costs; For the The wind power generation cost under this wind power joint output scenario includes operation and maintenance cost and wind abandonment cost. is a positive integer and is less than or equal to ; For the The cost of photovoltaic power generation in this wind power combined output scenario includes operation and maintenance costs and abandoned light costs.

[0011] In combination with the first aspect, in some implementations of the first aspect, the virtual power plant scheduling model further includes a gas turbine constraint condition, an energy storage system constraint condition, a wind and solar output constraint condition, and a power balance constraint condition, and the gas turbine constraint condition is: ; In the formula, is the minimum power output of the jth gas turbine at time t, is the maximum power output of the j-th gas turbine at time t, where j is a positive integer; The energy storage system constraints are: ; In the formula, and They are respectively the charging and discharging power of the energy storage system during normal period at time t; and They are the maximum charging and discharging power of the energy storage system during normal periods; and They are the charging and discharging power of the energy storage system during the peak and valley periods at time t; and are the maximum charging and discharging power of the energy storage system during peak and valley periods respectively; is the capacity status of the energy storage system; and are the minimum and maximum values ​​of the energy storage system capacity respectively; The wind and solar output constraints are: , ; In the formula, is the power that wind power should generate at time t; is the maximum value of wind power generation at time t; For the The actual power that should be generated by wind power in this wind power combined output scenario; For the The abandoned wind power of the wind turbine generator set at time t in the wind power joint output scenario; is the power that photovoltaic power generation should generate at time t; is the maximum value of photovoltaic power generation at time t; For the The actual power that photovoltaic power generation should generate under the wind power combined output scenario; For the The abandoned wind power of the photovoltaic generator set at time t in the wind power combined output scenario; The power balance constraint is: ; In the formula, Purchase power from the market for virtual power plants, Selling power to the market for virtual power plants; It is the power consumption of the load in the virtual power plant.

[0012] In combination with the first aspect, in some implementations of the first aspect, the particle swarm algorithm includes the following calculation steps: Step (201). Determine the calculation parameters, randomly generate the initial position and speed of the particles in the search space, the calculation parameters include the population size and the maximum number of iterations, each particle represents a candidate solution in the search space, and the speed is the direction and step size of the particle's movement; Step (202) determines the fitness function and calculates the fitness value of each particle's current position, which is used to judge the quality of the particle position; Step (203). Compare the fitness value of the current position of each particle with the fitness value of the optimal position, and update the optimal position of each particle; determine the position with the best fitness value in the optimal position of each particle as the global optimal position; Step (204). Based on the speed and the optimal position update formula of each particle, combined with the update parameters, the speed and the optimal position of each particle are updated, the update parameters include inertia weight, learning factor and random number, and the speed is within the specified range based on the boundary processing; Step (205). Re-execute the iterative calculation from step (202) to step (204) based on the updated speed and the optimal position until the number of iterations reaches the maximum number of iterations or the calculation is terminated when the convergence condition is met, and the convergence condition includes that the change value of the global optimal solution of multiple consecutive iterations is less than a threshold value; Step (206). Output the global optimal position and the corresponding optimal target value, where the optimal target value is the optimal solution determined by the particle swarm algorithm.

[0013] In a second aspect, a computer-readable storage medium is provided, on which computer instructions are stored. When the computer instructions are executed by a processor, the first aspect or any one of the implementation methods of the first aspect is implemented.

[0014] In a third aspect, a computer program product is provided, comprising computer program instructions, which, when executed on a computer, cause the computer to execute the first aspect or any one of the implementation methods of the first aspect.

[0015] The above method, device and readable storage medium for optimizing the virtual power plant dispatch model can reflect the internal connection of data by considering the covariance of data, selecting data cluster centers that are far apart and using Mahalanobis distance as a measure when clustering, and improve the accuracy and reliability of modeling the uncertainty of wind and solar output. At the same time, the dispatch decision of the virtual power plant takes into account the dual needs of users in terms of economic cost and environmental protection concept, which can better balance the interests of all parties and improve users' recognition of virtual power plant services. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 A schematic diagram of a virtual power plant scheduling model and its connection relationship provided in an embodiment of the present application; Figure 2 A flow chart of a method for optimizing a virtual power plant scheduling model provided in an embodiment of the present application; Figure 3 A flow chart of a method for reducing wind power combined output scenarios by using the ISODATA algorithm provided in an embodiment of the present application; Figure 4 A flow chart of a method for establishing a user demand model provided in an embodiment of the present application; Figure 5 A flow chart of a method for establishing a virtual power plant scheduling model provided in an embodiment of the present application; Figure 6 A flow chart of a method for solving a virtual power plant scheduling model using a PSO algorithm provided in an embodiment of the present application; Figure 7 A schematic diagram of a device provided in an embodiment of the present application; Figure 8 An internal structure diagram of a computer-readable storage medium provided in an embodiment of the present application. DETAILED DESCRIPTION

[0017] The terms used in the following embodiments of the present application are only for the purpose of describing specific embodiments and are not intended to be limiting of the present application. The technical solutions in the embodiments of the present application will be clearly and comprehensively described below in conjunction with the accompanying drawings. Among them, in the description of the embodiments of the present application, unless otherwise specified, " / " means or, for example, A / B can mean A or B; "and / or" in the text is only a description of the association relationship of associated objects, indicating that three relationships can exist, for example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone.

[0018] In the following, the terms "first" and "second" are used for descriptive purposes only and are not to be understood as suggesting or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features, and in the description of the embodiments of the present application, unless otherwise specified, "multiple" means two or more.

[0019] For ease of understanding, some relevant concepts of the embodiments of the present application are first explained as follows: A virtual power plant (VPP) is a system that integrates decentralized, small-scale power generation resources, energy storage equipment, and controllable loads into a unified, flexibly controllable power system through advanced information and communication technologies and energy management systems. Its core goal is to provide stable power supply or regulation services to the outside world like a traditional power plant, but without relying on centralized physical power generation facilities. In an embodiment of the present application, VPP obtains the optimal scheduling decision through a virtual power plant scheduling model.

[0020] Iterative self-organizing data analysis algorithm (ISODATA) is a dynamic clustering algorithm that optimizes data grouping by automatically adjusting the number and structure of clusters. In the embodiment of the present application, ISODATA is used to reduce the number of wind power combined output scenarios that need to be calculated, thereby improving the computational efficiency of the virtual power plant scheduling model while retaining the random characteristics in the wind power combined output scenario.

[0021] Particle Swarm Optimization (PSO) is a heuristic optimization algorithm based on swarm intelligence, which simulates the social behavior of bird flocks or fish schools and searches for the optimal solution in the solution space through collaboration between particles. In the embodiment of the present application, PSO is used to calculate the optimal solution for the virtual power plant scheduling model, which is the optimal scheduling decision of the VPP.

[0022] The above are some main related concepts involved in the embodiments of this application.

[0023] The present application provides a method for optimizing a virtual power plant dispatch model, which establishes a VPP operating benefit objective function based on the user's comprehensive satisfaction in the scenario of wind power combined output, and obtains the comprehensive optimal decision through the PSO algorithm. Among them, the method models the uncertainty of wind and solar power output, uses the Mahalanobis distance as a measure, and uses the improved ISODATA algorithm to reduce the number of scenarios of wind power combined output, considers the covariance of the data, so that the clustering results can better reflect the internal connection of the data, and improves the accuracy and reliability of modeling the uncertainty of wind and solar power output. At the same time, the method constructs a comprehensive user satisfaction model from the two dimensions of economy and low carbon, fully considering the dual needs of users in terms of economic cost and environmental protection concept, and improving the user's recognition of VPP services.

[0024] The following describes the method, device and readable storage medium for optimizing the virtual power plant scheduling model involved in the embodiments of the present application through Examples 1 to 3. Among them, Example 1 is used to describe the model involved in obtaining the virtual power plant scheduling plan; Example 2 is used to describe the specific method for optimizing the virtual power plant scheduling model; Example 3 is used to describe the device and readable storage medium for optimizing the virtual power plant scheduling model.

[0025] Example 1 For example, Figure 1 A schematic diagram of the virtual power plant scheduling model and its connection relationship provided in an embodiment of the present application is shown. The models involved in the VPP scheduling scheme include: a wind and solar energy uncertainty model, a user demand model and a virtual power plant scheduling model. Among them, the wind and solar energy uncertainty model is used to calculate the uncertainty of wind and solar output in the VPP, and determine the wind power joint output scenario in the virtual power plant scheduling model; the user demand model is constructed from two dimensions: economic indicators and low-carbon indicators, and is used to calculate the user's satisfaction with the VPP scheduling scheme, and determine the user's comprehensive dissatisfaction cost in the virtual power plant scheduling model; the virtual power plant scheduling model constructs a VPP operation benefit objective function based on the user's comprehensive dissatisfaction cost and the wind power joint output scenario, and solves the comprehensive optimal decision of the VPP scheduling scheme.

[0026] In the embodiment of the present application, the wind and solar energy uncertainty model models the uncertainty of wind and solar output by selecting cluster centers that are far apart and using Mahalanobis distance as a measure during clustering. Compared with the traditional method of using a simple probability distribution hypothesis to describe the uncertainty of wind and solar output, and using Euclidean distance as a measure during cluster analysis, the present application can fully consider the covariance of the data, so that the clustering results can better reflect the internal connection of the data, and improve the accuracy and reliability of the model in calculating the uncertainty of wind and solar output.

[0027] It should be noted that the wind-solar energy uncertainty model uses the ISODATA algorithm to reduce the wind power combined output scenarios that need to be calculated, which improves the calculation efficiency of the model and retains the random characteristics of the wind power combined output scenarios.

[0028] In the embodiment of the present application, the user demand model includes multiple indicators, and the comprehensive user satisfaction is constructed by linearly weighting each indicator. The comprehensive user satisfaction is the comprehensive user dissatisfaction cost in the virtual power plant scheduling model. Among them, the multiple indicators can be divided into two categories: economic indicators and low-carbon indicators. The economic indicators include user policy subsidy income indicators, user electricity cost indicators, user response comfort loss indicators, and user response production efficiency loss indicators; the low-carbon indicators are user multiple pollutant emission reduction indicators, including carbon dioxide emission reduction indicators, sulfur dioxide emission reduction indicators, and nitrogen oxide emission reduction indicators.

[0029] In the embodiment of the present application, the virtual power plant dispatch model constructs a VPP operation revenue objective function, which is composed of the VPP online purchase and sale revenue minus the VPP operation cost, and calculates the optimal solution of the VPP operation revenue objective function. Among them, the VPP operation cost includes the gas turbine operation cost, the energy storage system operation cost in the wind power combined output scenario, the user's comprehensive dissatisfaction cost, the wind power generation cost in the wind power combined output scenario, and the photovoltaic power generation cost.

[0030] It should be noted that VPP needs to meet multiple constraints during operation, including gas turbine constraints, energy storage system constraints, wind and solar output constraints, and power balance constraints. The virtual power plant dispatch model needs to calculate the optimal solution of the VPP operation benefit objective function under multiple constraints.

[0031] It should be noted that the virtual power plant scheduling model calculates the optimal solution of the VPP operation profit objective function through the PSO algorithm, and this optimal solution is the comprehensive optimal decision of the VPP.

[0032] In the embodiments of the present application, based on the wind and solar energy uncertainty model, the user demand model and the virtual power plant scheduling model, compared with the traditional method of processing wind and solar output uncertainty, the present application can obtain wind and solar output uncertainty with higher accuracy and reliability, and can consider the user's dual needs in economic cost and environmental protection concept, and make a comprehensive optimal decision for VPP.

[0033] Example 2 Figure 2 The method flow of optimizing the virtual power plant scheduling model is shown, which is applied to Figure 1 The wind and solar energy uncertainty model, user demand model and virtual power plant scheduling model are shown and detailed as follows.

[0034] S101. Establish a wind and solar power output uncertainty model.

[0035] In the embodiment of the present application, based on the non-parametric kernel density estimation method, the Gaussian function is used as the kernel function to generate the wind and solar power output probability density function. ,Should The specific calculation formula is: ; Where: The samples collected during the sampling period are wind and solar power output. is the i-th sample value of wind and solar power output, n is the number of historical days of wind and solar power output data, is the bandwidth, and i is a positive integer.

[0036] In the embodiment of the present application, based on Calculate the cumulative distribution function of wind power output and the cumulative distribution function of photovoltaic output , and based on and Solving the joint distribution function of wind and solar power output ,Should The specific calculation formula is: ; Where: and From , The wind power output is the sample collected during the sampling period. The samples collected during the sampling period are the photoelectric output; for and The relevance of Belongs to the interval range of -1 to 1 and is not equal to 0, where: A positive value indicates a positive correlation, whereas a negative correlation.

[0037] It should be noted that when calculating specific wind power combined output scenarios based on the above-mentioned wind-solar output joint distribution function calculation formula, due to the large number of wind power combined output scenarios, it is necessary to reduce and limit the number of wind power combined output scenarios.

[0038] In the embodiment of the present application, the ISODATA algorithm is used to reduce the number of wind power joint output scenarios and retain the random characteristics of the scenarios. The ISODATA algorithm includes the following calculation steps: Step (101). Obtain M samples , determine the initial cluster center and record the Expected minimum distance , i ranges from 1 to M, M is a positive integer; Step (102). Calculate the M The probability of being selected as the next cluster center , based on this Determine The cluster center is denoted as ,in, is a positive integer, The calculation formula is: ; Step (103). Calculate each To cluster center The Mahalanobis distance , each Assign to shortest among The corresponding cluster middle, and The calculation formulas are: , ; Where: is the weight matrix, , among which, if The number of samples in is less than the minimum number of samples , then remove the and the Corresponding ,Will In Assign to the remaining cluster centers and shortest among The corresponding cluster In, based on various In Recalculate the ; Step (104): Calculate The maximum variance in and standard deviation ,when Greater than , or the number of clusters in the current cluster satisfy When the number of iterations is an odd number, the split operation is performed. Split out new cluster centers and , the number of cluster centers Add 1, the specific calculation formula for the split operation is: ; Step (105). Calculate indivual When the Mahalanobis distance between two cluster centers is less than the threshold, or satisfy When the number of iterations is even, a merge operation is performed to merge the two cluster centers into a new cluster center. , the number of cluster centers Subtract 1, the specific calculation formula for the merge operation is: ; Where: The left side of the equation is the new cluster center, and the right side of the equation and are two cluster centers, and All are sample sets; Step (106). Repeat the iterative calculation from step (103) to step (105) until the number of iterative calculations reaches the maximum number of iterations, and then stop the calculation. Determined as the number of current cluster centers .

[0039] It should be noted that the ISODATA algorithm selects cluster centers that are far apart and uses the Mahalanobis distance as a measure when clustering, taking into account the covariance of the data, so that the clustering results reflect the internal connection of wind and solar power output data, thereby improving the accuracy and reliability of the calculation of wind and solar power output uncertainty.

[0040] In an embodiment of the present application, the wind-solar output uncertainty model reduces the wind power combined output scenario based on the ISODATA algorithm, and inputs the reduced wind power combined output scenario into the virtual power plant scheduling model.

[0041] S102. Establish user demand model.

[0042] In the embodiment of the present application, the user demand model includes economic indicators and low-carbon indicators. The user demand model uses a linear weighting method to construct user comprehensive satisfaction for different indicators. Among them, the economic indicators include user policy subsidy benefits , user response comfort loss, user response production efficiency loss and user electricity cost; low-carbon indicators include carbon dioxide emissions reduction, sulfur dioxide emissions reduction and nitrogen oxide emissions reduction.

[0043] In the embodiment of this application, the user policy subsidy income is the economic subsidy obtained by user i after participating in demand-side response. The specific calculation formula is:

[0044] Where N is the number of times user i participates in the demand side response during the demand side response time period, is the jth response power of user i, is the duration of user i's jth participation in demand-side response, is the subsidy unit price for user i’s j-th participation in demand-side response.

[0045] In the embodiment of the present application, the user response comfort loss is an important factor affecting whether the user participates in the VPP dispatch response and the degree of response. When the user does not participate in the VPP dispatch response, the user's electricity comfort is the highest; when the user participates in the VPP dispatch response, the user's electricity comfort will decrease as the electricity consumption mode changes. Among them, after the user participates in the VPP dispatch demand response, the electricity load is changed. , the specific calculation formula is:

[0046] In the formula, is the change in electric load after user i participates in VPP dispatch demand response in period t; The value range is between 0 and 1. The larger the value, the higher the user's electricity comfort.

[0047] In the embodiment of the present application, the user response production efficiency loss refers to the user's own utility loss caused by the load participating in the VPP scheduling demand response, such as the reduction in production efficiency due to load reduction, resulting in a reduction in the number of products during the VPP scheduling demand response time period.

[0048] In the embodiment of the present application, the user's electricity cost is used to influence the user's electricity utility. , the specific calculation formula is:

[0049] In the formula, The user's electricity utility, the typical marginal electricity utility of the user is usually negatively correlated with the electricity consumption; Costs closely related to electricity consumption include the user's electricity purchase cost under the system time-of-use electricity price and the real-time electricity price; The electricity cost is closely related to the electricity consumption period, including the user's peak shifting cost.

[0050] In the embodiment of the present application, the carbon dioxide emission reduction , sulfur dioxide emission reduction and nitrogen oxide reduction The specific calculation formulas are as follows: , , ; In the formula, To reduce power generation, is the amount of carbon dioxide emitted per unit of electricity generated, is the amount of sulfur dioxide emitted per unit of electricity generated, The amount of nitrogen oxides emitted per unit of electricity generated.

[0051] It should be noted that, since the magnitudes and units of the indicators included in the user demand model are different, the above indicators need to be normalized to obtain the measurement value of each indicator.

[0052] In the embodiment of the present application, based on the measured value of each indicator, the user's comprehensive satisfaction is calculated by linear weighting. , the specific calculation formula is:

[0053] In the formula: S is the number of indicators, indicating that there are S indicators in total; is the weight value of the sth indicator, is the measurement value of the sth indicator; It is the comprehensive user satisfaction, also known as the equivalent cost of comprehensive user satisfaction.

[0054] In the embodiment of the present application, the user demand model is based on The calculation formula obtains the equivalent cost of user comprehensive satisfaction and inputs it into the virtual power plant scheduling model.

[0055] S103. Establish a virtual power plant scheduling model.

[0056] In the embodiment of the present application, the virtual power plant dispatch model constructs a VPP operation benefit objective function based on the wind power joint output scenario obtained in step S101 and the user comprehensive satisfaction equivalent cost obtained in step S102. , solve for The comprehensive optimal decision under the wind power joint output scenario. The specific calculation formula is:

[0057] Where: The income from the VPP’s online purchase and sale of electricity; The operating cost of the gas turbine includes power generation cost, operating cost and downtime cost; for The operating cost of the energy storage system in a wind power combined output scenario, including the charging and discharging costs at each time period; The equivalent cost of user comprehensive satisfaction obtained in step S102, also known as demand response cost; For the The cost of wind power generation in a wind power combined output scenario, including operation and maintenance costs and wind curtailment costs; For the The cost of photovoltaic power generation in a wind power combined output scenario includes operation and maintenance costs and abandoned light costs.

[0058] In the embodiment of the present application, the revenue from electricity purchase and sale is , Gas turbine operating costs , Energy storage system operating costs , Wind power generation costs and photovoltaic power generation costs The specific calculation formulas are as follows: 1. Profits from electricity purchase and sale :

[0059] Where: The price at which VPP purchases electricity from the market. The price at which VPP sells electricity to the market; Purchase power from the market for VPP, Selling power to the market for VPP; 2. Gas turbine operating costs :

[0060] Where: is the output power of gas turbine j at time t; is the number of gas turbine units; and They are the quadratic coefficient and the linear coefficient of the gas turbine operating cost respectively; 3. Energy storage system operating costs :

[0061] Where: is the charging and discharging cost coefficient; is the charging power of the energy storage system during normal periods, is the discharge power of the energy storage system during normal periods; is the charging power during peak and valley periods, is the discharge power during peak and valley periods; 4. Wind power generation costs :

[0062] Where: is the operation and maintenance cost coefficient of wind power generation, is the wind abandonment cost coefficient of wind power generation; is the actual power of the wind turbine generator set at time t, is the abandoned wind power of the wind turbine generator set at time t; 5. Photovoltaic power generation costs :

[0063] Where: is the operation and maintenance cost coefficient of photovoltaic power generation, is the wind curtailment cost coefficient of photovoltaic power generation; is the actual power of the photovoltaic generator set at time t, is the abandoned wind power of the photovoltaic generator set at time t.

[0064] In the embodiment of the present application, the virtual power plant scheduling model needs to meet multiple constraints during operation, including gas turbine constraints, energy storage system constraints, wind and solar output constraints, and power balance constraints. The specific constraints are as follows: 1. Gas turbine constraints:

[0065] In the formula, is the minimum power output of the jth gas turbine at time t, is the maximum power output of the jth gas turbine at time t; 2. Energy storage system constraints:

[0066] In the formula, and They are the maximum charging and discharging power of the energy storage system during normal periods; and are the maximum charging and discharging power of the energy storage system during peak and valley periods respectively; is the capacity status of the energy storage system; and are the minimum and maximum values ​​of the energy storage system capacity respectively; 3. Wind and solar power output constraints: , ; In the formula, is the maximum value of wind power generation at time t; For the The actual power that should be generated by wind power in the combined output scenario of wind power; is the maximum value of photovoltaic power generation at time t; For the The actual power that photovoltaic power generation should generate in the scenario of combined output of wind power; 4. Power balance constraints:

[0067] In the formula, It is the internal load power of VPP.

[0068] In the embodiment of the present application, the virtual power plant scheduling model is constructed and constraints, and calculate under the constraints The optimal solution is the comprehensive optimal decision for VPP scheduling.

[0069] S104. Use PSO algorithm to solve the virtual power plant scheduling model.

[0070] In the embodiment of the present application, the profit objective function in the virtual power plant scheduling model is Including multiple wind power joint output scenarios, parameters and constraints. The optimal solution can be calculated by PSO algorithm.

[0071] In the embodiment of the present application, the PSO algorithm includes the following calculation steps: Step (201). Determine the calculation parameters, randomly generate the initial position and speed of the particles in the search space, the calculation parameters include the population size and the maximum number of iterations, each particle represents a candidate solution in the search space, and the speed is the direction and step size of the particle's movement; Step (202) determines the fitness function and calculates the fitness value of each particle's current position, which is used to judge the quality of the particle position; Step (203). Compare the fitness value of the current position of each particle with the fitness value of the optimal position, and update the optimal position of each particle; determine the position with the best fitness value in the optimal position of each particle as the global optimal position; Step (204). Based on the speed and the optimal position update formula of each particle, combined with the update parameters, the speed and the optimal position of each particle are updated, the update parameters include inertia weight, learning factor and random number, and the speed is within the specified range based on the boundary processing; Step (205). Re-execute the iterative calculation from step (202) to step (204) based on the updated speed and the optimal position until the number of iterations reaches the maximum number of iterations or the calculation is terminated when the convergence condition is met, and the convergence condition includes that the change value of the global optimal solution of multiple consecutive iterations is less than a threshold value; Step (206). Output the global optimal position and the corresponding optimal target value, where the optimal target value is the optimal solution determined by the PSO algorithm, and the optimal solution is the comprehensive optimal decision for VPP scheduling.

[0072] In the embodiment of the present application, the virtual power plant scheduling model obtains the comprehensive optimal decision through the PSO algorithm, and the VPP can perform scheduling based on the comprehensive optimal decision, so that the scheduling decision of the VPP can better balance the interests of the VPP users and suppliers, and improve users' recognition of the VPP service.

[0073] Figure 3 The method flow of reducing the wind power combined output scenario through the ISODATA algorithm is shown, and the details are as follows.

[0074] S201. Obtain a sample set and determine the expected shortest distance between the sample and the initial cluster center.

[0075] In the embodiment of the present application, the samples of the ISODATA algorithm include samples collected from wind and solar power output, and the sample set with M samples is recorded as .

[0076] In the embodiment of the present application, one or more samples in the sample set can be used as the initial cluster center, and the distance between each sample and each cluster center is calculated to determine the cluster center with the shortest distance to the sample. The shortest distance is the expected shortest distance, which is recorded as .

[0077] S202. Determine the number of cluster centers.

[0078] In the embodiment of the present application, the number of cluster centers needs to be recalculated, considering that each sample may be selected as a cluster center, so the sample set needs to be calculated. The probability of each sample in being selected as the next cluster center ,according to Determine in order from largest to smallest samples as cluster centers, denoted as .in, is a positive integer, which can be the preset value of the wind and solar power output uncertainty model. The calculation formula of is shown in step S101 and will not be repeated here.

[0079] S203. Classify the sample set into the nearest cluster center.

[0080] In the embodiment of the present application, the sample set is calculated Each sample in each cluster center The Mahalanobis distance is denoted as ,in, and The calculation formula of is shown in step S101 and will not be repeated here.

[0081] In the embodiment of the present application, each sample is based on the shortest , added to the shortest Cluster Center The corresponding cluster In this way, each sample is classified to the nearest cluster center.

[0082] It should be noted that each cluster center has a corresponding cluster. If the number of samples in the cluster is less than the minimum number of samples , it indicates that the number of samples corresponding to the cluster center is too small and is not suitable as a cluster center. Therefore, remove the cluster and cluster centers , for the original cluster The samples contained in reselect the cluster center corresponding to the shortest Mahalanobis distance and add them to the cluster cluster corresponding to the cluster center.

[0083] S204. Perform iterative calculation on the cluster centers.

[0084] In the embodiment of the present application, step S203 obtains Cluster Centers And the corresponding clusters , determine each cluster center Whether to perform splitting or merging operations and iteratively calculate cluster centers .

[0085] In the embodiment of the present application, the cluster center is determined The conditions for splitting operations include: The corresponding cluster Calculate the maximum variance and standard deviation ,like Greater than , then it means that the cluster center A split operation needs to be performed, wherein the formula of the split operation is shown in step S101 and will not be repeated here.

[0086] In the embodiment of the present application, the cluster center is determined The conditions for merging operations include: calculating two cluster centers and If the Mahalanobis distance is less than the preset threshold, it means that the cluster center With cluster center A merging operation needs to be performed, wherein the formula for the merging operation is shown in step S101 and will not be repeated here.

[0087] In the embodiment of the present application, the cluster center is determined The conditions for performing the split operation can also include: the number of iterations is an odd number, or the number of clusters in the current cluster is satisfy Quantitative relationship; Determine the cluster center The conditions for performing the merge operation can also include: the number of iterations is an even number, or the number of clusters in the current cluster is satisfy quantitative relationship.

[0088] It should be noted that the cluster centers are iteratively calculated, such as looping through steps S203-S204. After a split operation, the new cluster center is obtained and Re-execute steps S203-S204.

[0089] S205. When the iterative calculation reaches the maximum number of iterations, the calculation is stopped, and the number of cluster centers is determined to be the number of wind power combined output scenarios.

[0090] In the embodiment of the present application, the number of iterative calculations is counted until the number of iterative calculations reaches the maximum number of iterations, and the calculation is stopped, indicating that the sample set has been determined. The clustering result of . At this time, the sample set The clustering results correspond to the wind power joint output scenario. The number of wind power joint output scenarios The number of current cluster centers equal.

[0091] Figure 4 The method flow for establishing a user demand model is shown and is described in detail as follows.

[0092] S301. Determine the user policy subsidy benefit indicator.

[0093] In the embodiment of the present application, the user demand model includes economic indicators and low-carbon indicators, among which the economic indicators mainly affect the user's enthusiasm for participating in the VPP scheduling response, including user benefit indicators (such as user policy subsidy benefit indicators), user loss indicators (such as user response comfort loss indicators and user response production efficiency loss indicators) and user cost indicators (such as user electricity cost indicators). It is necessary to comprehensively consider the above-mentioned multiple indicators to establish an accurate user demand model.

[0094] For a detailed description of the user policy subsidy income indicator in the embodiment of the present application, please refer to the aforementioned Figure 2 Step S102 will not be described in detail here.

[0095] S302. Determine a user response comfort loss index and a user response production efficiency loss index.

[0096] In the embodiment of the present application, the specific description of the user response comfort loss index and the user response production efficiency loss index can be referred to the aforementioned Figure 2 Step S102 will not be described in detail here.

[0097] S303. Determine the user's electricity cost index.

[0098] In the embodiment of the present application, the goal of rational users to consume electricity is to obtain the maximum electricity utility. , for electricity efficiency The specific calculation formula can refer to the above Figure 2 Step S102 will not be described in detail here.

[0099] S304. Determine the user's multiple pollutant emission reduction indicators.

[0100] In the embodiment of the present application, the VPP dispatch response to the industrial load demand side can greatly reduce the power generation on the power generation side, indirectly reducing environmental pollution. For example, thermal power generation requires a large amount of fossil energy, emitting various pollutants such as carbon dioxide, sulfur dioxide, nitrogen oxides (such as nitric oxide, nitrogen dioxide, etc.), posing a threat to the environment. Based on the reduced power generation, the emission reduction of various pollutants can be calculated to constitute a low-carbon index.

[0101] In the embodiment of the present application, the specific calculation formula for the user's multiple pollutant emission reduction indicators can refer to the aforementioned Figure 2 Step S102 will not be described in detail here.

[0102] S305. Based on the determined indicators, construct a comprehensive user satisfaction model.

[0103] In the embodiment of the present application, the specific calculation formula of the user's comprehensive satisfaction model can refer to the aforementioned Figure 2 Step S102 will not be described in detail here.

[0104] Figure 5 The method flow for establishing a virtual power plant scheduling model is shown and described in detail as follows.

[0105] S401. Determine the VPP operation benefit objective function based on the wind power combined output scenario with comprehensive user dissatisfaction cost.

[0106] In the embodiment of the present application, VPP runs the revenue objective function and solves The comprehensive optimal decision under the wind power joint output scenario. The specific calculation formula of the benefit objective function can be referred to the above Figure 2 Step S103 will not be described in detail here.

[0107] S402: Determine the constraints that the VPP operation should meet.

[0108] In the embodiment of the present application, the VPP will be constrained by objective conditions such as equipment and environment during operation. In order to prevent the optimal solution obtained by the virtual power plant scheduling model from exceeding the limitations of objective conditions such as equipment and environment, it is necessary to limit the parameters of the virtual power plant scheduling model.

[0109] In the embodiment of the present application, the virtual power plant scheduling model needs to meet multiple constraints during operation, including gas turbine constraints, energy storage system constraints, wind and solar output constraints, and power balance constraints. For specific constraints, please refer to the aforementioned Figure 2 The step S103 in will not be described in detail here.

[0110] Figure 6 The method flow of using the PSO algorithm to solve the virtual power plant scheduling model is shown and detailed as follows.

[0111] S501. Initialize the PSO algorithm and determine the parameters required by the PSO algorithm.

[0112] In the embodiment of the present application, the specific description of initializing the PSO algorithm and determining the parameters required by the PSO algorithm can refer to the aforementioned Figure 2 Step S104 will not be described in detail here.

[0113] S502. Calculate the fitness of particles in the PSO algorithm.

[0114] In an embodiment of the present application, a fitness function is determined, and the fitness value of each particle's current position is calculated. The fitness value is used to judge the quality of the particle's position. For example, if the fitness value of a particle at its current position is 5 and the fitness value of the particle at its optimal position is 4, then the particle's current position is better than the recorded optimal position.

[0115] S503. Update the optimal fitness of the particle.

[0116] In an embodiment of the present application, the fitness value of the current position of each particle is compared with the fitness value of the optimal position, and the optimal position of each particle is updated. Exemplarily, the fitness value of the particle at the current position is 0.5, and the fitness value of the particle at the optimal position is 0.4, then the optimal position of the particle is updated to the current position; the position with the best fitness value is determined in the optimal position of each particle as the global optimal position. For example, the particle group includes two particles, one particle has a fitness value of 0.4 at the optimal position, and the other particle has a fitness value of 0.6 at the optimal position, then the global optimal position is the optimal position of the particle with a fitness value of 0.6.

[0117] S504. Update the velocity and position of the particle.

[0118] For a detailed description of the speed and position of the updated particles in the embodiment of the present application, please refer to the aforementioned Figure 2 Step S104 will not be described in detail here.

[0119] S505. Iteratively calculate the global optimal position of the particle.

[0120] In the embodiment of the present application, iterative calculations from step S502 to step S504 are re-executed based on the updated particle speed and the particle optimal position to obtain the global optimal position of the particle. For example, the particle group includes two particles. In the first round of iteration, the fitness value of one particle at the optimal position is 0.4, and the fitness value of the other particle at the optimal position is 0.6. The global optimal position is the position where the particle fitness value is 0.6; in the second round of iteration, the fitness value of one particle at the optimal position is 0.8, and the fitness value of the other particle at the optimal position is 0.7. The global optimal position is updated to the position where the particle fitness value is 0.8; in the third round of iteration, the fitness value of one particle at the optimal position is 0.7, and the fitness value of the other particle at the optimal position is 0.2. The global optimal position remains at the position where the particle fitness value is 0.8.

[0121] In the embodiment of the present application, the PSO algorithm can determine the global optimal position of the particle by performing multiple iterative calculations until the convergence condition is met. The convergence condition includes that the number of iterations reaches the maximum number of iterations, and the change value of the global optimal solution of multiple consecutive iterations is less than a threshold value.

[0122] S506. Output the optimal target value.

[0123] In the embodiment of the present application, the global optimal position corresponds to the optimal target value. For a specific description of the output optimal target value, please refer to the aforementioned Figure 2 Step S104 will not be described in detail here.

[0124] Example 3 like Figure 7 The structure of a device provided by an embodiment of the present application is shown. The device 700 includes: a processor 701, a memory 702, and a computer program 703 stored in the memory 702 and executable on the processor 701. When the processor 701 executes the computer program 703, the above Figure 1-Figure 6 The method shown in .

[0125] Exemplarily, the computer program 703 may be divided into one or more units / modules, which are stored in the memory 702 and executed by the processor 701 to complete the present application. The one or more units / modules may be a series of computer program instruction segments capable of completing specific functions, which are used to describe the execution process of the computer program 703 in the device 700.

[0126] For example, the computer program 703 may be used to execute the Figure 2The method flow for optimizing the virtual power plant scheduling model shown in steps S101-S104, the specific functions or mechanisms have been described in the above embodiments and will not be repeated here.

[0127] The above-mentioned device 700 may include, but is not limited to, a processor 701 and a memory 702. Those skilled in the art will appreciate that Figure 7 It is only an example of the device 700 and does not constitute a limitation of the device 700. It may include more or fewer components than shown in the figure, or a combination of certain components, or different components. For example, the above-mentioned device 700 may also include input and output devices, network access devices, buses, etc.

[0128] The processor 701 may be a CPU, or other general-purpose processors, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc.

[0129] The memory 702 may be an internal storage unit of the apparatus 700, such as a hard disk or a memory of the apparatus 700. The memory 702 may also include both an internal storage unit of the apparatus 700 and an external storage device.

[0130] The memory 702 is used to store the computer program and other programs and data required by the apparatus 700. The memory 702 may also be used to temporarily store data that has been output or is to be output.

[0131] Those skilled in the art can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the above-mentioned device can be divided into different functional units or modules to complete all or part of the functions described above.

[0132] The functional units and modules in the embodiments may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated units may be implemented in the form of hardware or in the form of software functional units.

[0133] In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other and are not used to limit the scope of protection of the present application.

[0134] Those skilled in the art will understand that Figure 7 The structure shown in the figure is merely a block diagram of a partial structure related to the scheme of the present application, and does not constitute a limitation on the device to which the scheme of the present application is applied. The specific device may include more or fewer components than shown in the figure, or combine certain components, or have a different arrangement of components.

[0135] It should be understood that each step in the above method embodiment provided by the present application can be completed by an integrated logic circuit of hardware in a processor or by instructions in the form of software. The method steps disclosed in the embodiments of the present application can be directly embodied as being executed by a hardware processor, or by a combination of hardware and software modules in a processor.

[0136] In some embodiments, a computer device is provided. The computer device includes a memory and a processor. The memory stores a computer program. When the processor executes the computer program, the steps in any of the above embodiments are implemented.

[0137] The present application also provides a computer program product, which includes: a computer program (also referred to as code, or instruction), which enables a computer to execute a method in any one of the above embodiments when the computer program is executed.

[0138] like Figure 8 As shown, the present application also provides a computer-readable storage medium, which stores a computer program (also referred to as code or instruction). When the computer program is executed, the computer executes the method in any of the aforementioned embodiments.

[0139] The present application also provides a chip system, which includes at least one processor for implementing the functions involved in the method executed by the device in any of the above embodiments.

[0140] In one possible design, the chip system also includes a memory, which is used to store program instructions and data, and the memory is located inside or outside the processor.

[0141] The chip system may be composed of the chip, or may include the chip and other discrete devices.

[0142] In some embodiments, the processor in the chip system may be one or more. The processor may be implemented by hardware or by software. When implemented by hardware, the processor may be a logic circuit, an integrated circuit, etc. When implemented by software, the processor may be a general-purpose processor implemented by reading software code stored in a memory.

[0143] In some embodiments, the memory in the chip system may also be one or more. The memory may be integrated with the processor or may be separately arranged with the processor, which is not limited in the embodiments of the present application. For example, the memory may be a non-transient processor, such as a read-only memory ROM, which may be integrated with the processor on the same chip or may be arranged on different chips respectively. The embodiments of the present application do not specifically limit the type of memory and the arrangement of the memory and the processor.

[0144] Exemplarily, the chip system may be a field programmable gate array (FPGA), an application specific integrated circuit (ASIC), a system on chip (SoC), a central processor unit (CPU), a network processor (NP), a digital signal processor (DSP), a microcontroller unit (MCU), a programmable logic device (PLD), or other integrated chips.

[0145] The various implementation modes of the present application can be arbitrarily combined to achieve different technical effects. To make the description concise, not all possible combinations of the various technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0146] A person skilled in the art can understand that to implement all or part of the processes in the aforementioned embodiments, the processes can be completed by a computer program to instruct the relevant hardware, and the program can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the aforementioned method embodiments. The aforementioned storage medium includes: a read-only memory ROM or a random access memory RAM, a magnetic disk or an optical disk, and other media that can store program codes.

[0147] In short, the above description is only an embodiment of the technical solution of this application, and is not intended to limit the protection scope of this application. Any modification, equivalent replacement, improvement, etc. made according to the disclosure of this application shall be included in the protection scope of this application.

Claims

1. A method for optimizing a virtual power plant scheduling model, characterized in that: include: Based on the characteristics of wind and solar power output uncertainty, a wind and solar power output uncertainty model is established. The wind and solar power output uncertainty model is used to obtain the wind and solar power output uncertainty through the iterative self-organizing data analysis algorithm ISODATA. wind power joint output scenario, the is a positive integer; Based on the needs of economic cost and environmental protection concept, a user demand model is established, and the user demand model is used to obtain the comprehensive user satisfaction; Based on the wind and solar power output uncertainty model and the user demand model, a virtual power plant scheduling model is established. The virtual power plant scheduling model is used to solve the The optimal decision considering the comprehensive satisfaction of the users in the wind power combined output scenario; A particle swarm algorithm is used to calculate the optimal solution of the virtual power plant scheduling model, and the optimal solution is determined as the scheduling decision of the virtual power plant.

2. The method according to claim 1, characterized in that: The wind and solar power output uncertainty model includes a wind and solar power output probability density function And wind and solar output joint distribution function , The calculation formula is: ; Where: The samples collected during the sampling period are wind and solar power output. is the i-th sample value of wind and solar power output, n is the number of historical days of wind and solar power output data, is the bandwidth, i is a positive integer; Said The calculation formula is: ； Where: and From the said , The wind power output is the sample collected during the sampling period. is the cumulative distribution function of wind power output, is the sample collected by the photoelectric output during the sampling period, is the cumulative distribution function of photovoltaic output; For the With the The relevance of It is in the range of -1 to 1 and is not equal to 0.

3. The method according to claim 1, characterized in that: The iterative self-organizing data analysis algorithm ISODATA comprises the following calculation steps: Step (101). Obtain M samples , determine the initial cluster center and record the Expected minimum distance , i is in the range of 1 to M, and M is a positive integer; Step (102). Calculate the M The probability of being selected as the next cluster center , based on the Determine The cluster centers are denoted as , is a positive integer, The calculation formula is: ； Step (103). Calculate each of the to the cluster center The Mahalanobis distance , each of the Assigned to the shortest among The corresponding cluster In and stated The calculation formulas are: ； ； Where: is the weight matrix, ; If the The number of samples in is less than the minimum number of samples , then remove the and stated The corresponding , the In Assign to the remaining cluster centers and shortest among The corresponding cluster In, based on the various In Recalculate the ; Step (104). Calculate the The maximum variance in and standard deviation , when the Greater than , or the number of clusters in the current satisfy When the number of iterations is an odd number, a split operation is performed to split the Split out new cluster centers and , for the Add 1, the specific calculation formula of the split operation is: ； Step (105). Calculate the described When the Mahalanobis distance between two cluster centers is less than a threshold, or satisfy When the number of iterations is an even number, a merging operation is performed to merge the two cluster centers into a new cluster center. , the number of cluster centers Subtract 1, the specific calculation formula of the merging operation is: ( )； Where: The left side of the equation is the new cluster center, the right side of the equation With the are the two cluster centers, and stated All are sample sets; Step (106). Repeat the iterative calculation from step (103) to step (105) until the number of iterative calculations reaches the maximum number of iterations and then stop the calculation. Determine the number of cluster centers currently described .

4. The method according to claim 1, characterized in that: The user demand model includes an economic index and a low-carbon index, and the user comprehensive satisfaction is constructed by linearly weighting the economic index and the low-carbon index.

5. The method according to claim 4, characterized in that: The economic indicators include policy subsidy benefits, response comfort loss, response production efficiency loss and electricity costs; the low-carbon indicators include carbon dioxide emission reduction, sulfur dioxide emission reduction and nitrogen oxide emission reduction.

6. The method according to claim 1, characterized in that: The virtual power plant dispatch model includes a virtual power plant operation revenue objective function , The optimal solution is in the The optimal decision considering the comprehensive satisfaction of the user in the wind power combined output scenario is The calculation formula is: ； In the formula, The revenue from the purchase and sale of electricity for the virtual power plant; The operating cost of the gas turbine includes power generation cost, operating cost and downtime cost; For the The operating cost of the energy storage system in a wind power combined output scenario, including the charging and discharging costs at each time period; for demand response costs; For the The wind power generation cost under the wind power combined output scenario includes operation and maintenance cost and wind abandonment cost. is a positive integer and is less than or equal to the ; For the The cost of photovoltaic power generation in the combined output scenario of wind power mentioned above includes operation and maintenance cost and abandoned light cost.

7. The method according to claim 6, characterized in that: The virtual power plant scheduling model also includes gas turbine constraints, energy storage system constraints, wind and solar output constraints, and power balance constraints. The gas turbine constraints are: ； In the formula, is the minimum power output of the jth gas turbine at time t, is the maximum power output of the j-th gas turbine at the time t, where j is a positive integer; The energy storage system constraints are: ； In the formula, and They are respectively the charging and discharging power of the energy storage system during normal period at time t; and They are the maximum charging and discharging power of the energy storage system during normal periods; and They are the charging and discharging power of the energy storage system during the peak and valley periods at time t; and are the maximum charging and discharging power of the energy storage system during peak and valley periods respectively; is the capacity status of the energy storage system; and are the minimum and maximum values ​​of the energy storage system capacity respectively; The wind and solar output constraints are: ， ； In the formula, is the power that wind power should generate at time t; is the maximum value of wind power generation at time t; For the The actual power that should be generated by wind power in the wind power combined output scenario; For the The abandoned wind power of the wind turbine generator set at time t in the wind power combined output scenario; is the power that photovoltaic power generation should generate at time t; is the maximum value of photovoltaic power generation at time t; For the The actual power that photovoltaic power generation should generate under the wind power combined output scenario; For the The abandoned wind power of the photovoltaic generator set at time t in the wind power combined output scenario; The power balance constraint condition is: ； In the formula, Purchase power from the market for virtual power plants, Selling power to the market for virtual power plants; It is the power consumption of the load in the virtual power plant.

8. The method according to claim 1, characterized in that: The particle swarm algorithm includes the following calculation steps: Step (201). Determine calculation parameters, randomly generate initial positions and velocities of particles in the search space, the calculation parameters include population size and maximum number of iterations, each particle represents a candidate solution in the search space, and the velocity is the direction and step length of movement of the particle; Step (202) determines the fitness function and calculates the fitness value of each particle's current position, wherein the fitness value is used to judge the quality of the particle position; Step (203). Compare the fitness value of the current position of each particle with the fitness value of the optimal position, and update the optimal position of each particle; determine the position with the best fitness value among the optimal positions of each particle as the global optimal position; Step (204). Based on the speed and the optimal position update formula of each particle, in combination with update parameters, the speed and the optimal position of each particle are updated, the update parameters include inertia weight, learning factor and random number, and the speed is within a specified range based on boundary processing; Step (205). Re-execute the iterative calculation from step (202) to step (204) based on the updated speed and the optimal position, until the number of iterations reaches the maximum number of iterations or the calculation is terminated when the convergence condition is met, and the convergence condition includes that the change value of the global optimal solution of multiple consecutive iterations is less than a threshold value; Step (206). Output the global optimal position and the corresponding optimal target value, where the optimal target value is the optimal solution determined by the particle swarm algorithm.

9. A computer-readable storage medium having computer instructions stored thereon, characterized in that: When the computer instructions are executed by a processor, the steps of the method according to any one of claims 1 to 8 are implemented.

10. A computer program product, comprising computer program instructions, which, when executed on a computer, cause the computer to execute the method according to any one of claims 1 to 8.

Images