Distributed wind storage power station cluster day-ahead output optimization method, dispatching method and terminal

Abstract

The application discloses a kind of distributed wind storage power station cluster day-ahead output optimization method, scheduling method, terminal, wherein optimization method includes: aggregation layer is with aggregation layer to each wind storage power station each time period formulated retail price, the total power of aggregation layer, the power of charge and discharge of energy storage system that aggregation layer possesses is decision variable, with minimum operating cost as target, establishes first optimization model;Wind storage power station layer in each wind storage power station with the power of wind storage power station, wind turbine generator power, the power of charge and discharge of energy storage, charging power is decision variable, with maximum power generation income as target, establishes second optimization model;First optimization module and second optimization module constitute master-slave game model, then conversion nonlinear programming model is solved, obtains the optimal day-ahead output plan curve of each wind storage power station.Scale distributed wind storage power station cluster is reasonably dispatched and efficiently utilized, realizes the win-win effect of grid side cost optimization and power station side maximum income.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of optimal scheduling of power systems, and particularly relates to a distributed wind storage power station cluster day-ahead output optimization method, a scheduling method, a terminal and a medium. BACKGROUND

[0002] At present, many wind farms are located in mountainous and hilly areas, and the proportion of wind power stations with an altitude of more than 500 m is high, and the main feature is to access the power grid in a dispersed manner. A large number of distributed power generation systems with intermittency and randomness are connected to the power grid, which greatly increases the complexity and control difficulty of the power grid, and has a significant impact on the safe, reliable and economic operation of the power grid.

[0003] Energy storage technology, especially new energy storage technology represented by electrochemical energy storage, has the advantages of bidirectional power regulation, fast response rate, flexible charging and discharging, etc. It can not only assist the active control of wind, light and other distributed power generation systems, smooth the power flow curve and improve the frequency characteristics of the system, but also use the reactive power capacity of the energy storage converter to provide reactive power compensation function and improve the voltage distribution of the system. In order to realize the friendly access and consumption of large-scale distributed wind farms, it is gradually required to configure a certain proportion of energy storage systems in wind farms to form wind storage power stations.

[0004] However, distributed wind storage power stations have the characteristics of a large number of points, wide surfaces, small single machine capacity and other significant distributed power clusters, and the traditional centralized scheduling and control mode will be difficult to effectively implement. Under the background of increasing renewable energy penetration, how to realize the reasonable scheduling and efficient utilization of large-scale distributed wind storage power stations will become a major issue. SUMMARY

[0005] The present application provides a distributed wind storage power station cluster day-ahead output optimization method, a scheduling method, a terminal and a medium to solve the problem that the existing centralized scheduling and control mode is difficult to realize the reasonable scheduling and efficient utilization of large-scale distributed wind storage power station clusters.

[0006] In a first aspect, a distributed wind storage power station cluster day-ahead output optimization method is provided, comprising:

[0007] The aggregation layer establishes a first optimization model with the minimum operation cost as the target, according to the day-ahead output prediction curve of all wind storage power stations in the distributed wind storage power station cluster and the day-ahead wholesale electricity price of the electricity spot market, with the retail electricity price formulated by the aggregation layer to each wind storage power station in each time period, the total sold power of the aggregation layer, and the charging and discharging power of the energy storage system owned by the aggregation layer as decision variables.

[0008] The wind storage power station layer formulates a retail electricity price according to the aggregation layer, and combines the day-ahead output prediction curve and the configuration capacity of the energy storage, to take the selling power of the wind storage power station, the power generated by the wind turbine generator in the wind storage power station, the discharging power and the charging power of the energy storage in the wind storage power station as decision variables, and to take the maximum power generation benefit as a target, to establish a second optimization model;

[0009] The first optimization module in the aggregation layer and the second optimization module in the wind storage power station layer form a master-slave game model, which is converted into a single-layer conventional nonlinear programming model through transformation of the master-slave game model, and then is solved by using a particle swarm algorithm to obtain the optimal day-ahead selling power curve of each wind storage power station, that is, the optimal day-ahead output planning curve of each wind storage power station.

[0010] Further, the first optimization model has an optimization duration of 24 hours of the next day, and a target function of minimum operation cost of the aggregation layer, and is expressed as follows:

[0011]

[0012] In the formula, F represents the operation cost of the aggregation layer, t represents the serial number of the optimization period, T represents the total number of optimization periods, ΔT represents the time length of each optimization period, i represents the number of the wind storage power station, G represents the number of the wind storage power stations, c t represents the retail electricity price formulated by the aggregation layer for each wind storage power station in each period, P i,t represents the purchasing power of the aggregation layer from the ith wind storage power station in the tth period, that is, the selling power of the ith wind storage power station in the tth period; π t represents the selling price of the aggregation layer to the power grid in the tth period, and in the model, the aggregation layer is considered as a price accepter in the day-ahead spot market of the power grid, so the selling price is a known quantity, which is called the day-ahead wholesale electricity price of the power spot market; P t represents the selling power of the aggregation layer to the power grid in the tth period;

[0013] The constraint conditions of the first optimization model and the corresponding mathematical expressions are as follows:

[0014] The upper and lower limit constraints of the retail electricity price are as follows:

[0015]

[0016] In the formula, L represents the minimum retail electricity price limit value in the tth period, represents the maximum retail electricity price limit value in the tth period;

[0017] The accumulated sum of the retail electricity price in a day is a constant constraint, and is expressed as follows:

[0018]

[0019] where C is a constant, representing the upper limit of the cumulative retail electricity price;

[0020] The upper and lower limits of the aggregated layer's selling power to the grid:

[0021] 0≤P t ≤P M ,t=1,2,3,...,T

[0022] where P M represents the maximum allowed value of the aggregated layer's selling power to the grid;

[0023] The supply-demand balance constraint of the aggregated layer:

[0024]

[0025] where P represents the discharge power of the energy storage system owned by the aggregated layer in the tth time period, represents the charge power of the energy storage system owned by the aggregated layer in the tth time period;

[0026] The operation constraint of the energy storage system owned by the aggregated layer:

[0027]

[0028] where P max represents the maximum charge and discharge power allowed by the energy storage system, u t is a 0-1 binary variable to prevent the energy storage from charging and discharging at the same time, E t represents the state of charge of the energy storage system at the end of the tth time period, η c and η d represent the charging and discharging efficiencies of the energy storage system, E max and E min represent the maximum and minimum state of charge allowed by the energy storage system, E0and E T represent the initial and final state of charge of the energy storage system in the dispatching period.

[0029] Further, the second optimization model also optimizes for 24 hours the next day, and the objective function is to maximize the generation revenue of each wind storage power station, which is represented as follows:

[0030]

[0031] where F i represents the generation revenue of the ith wind storage power station, the penalty coefficient ρ>0, represents the discharge power of the energy storage in the tth time period in the ith wind storage power station, represents the charge power of the energy storage in the tth time period in the ith wind storage power station;

[0032] The constraint conditions of the second optimization model and their corresponding mathematical expressions are as follows:

[0033] The wind storage power station power balance constraint:

[0034]

[0035] In the formula, Pwind,i(t) represents the power generated by the wind turbine in the ith wind storage power station;

[0036] The wind turbine power generation constraint:

[0037]

[0038] In the formula, Pwind,i(t) represents the predicted power generated by the wind turbine in the ith wind storage power station at the tth time period;

[0039] The energy storage operation constraint:

[0040]

[0041] In the formula, P i max Pmax,i(t) represents the maximum charging and discharging power allowed by the energy storage in the ith wind storage power station, E i,t Soc,i(t) represents the state of charge of the energy storage in the ith wind storage power station at the end of the tth time period, and respectively represent the charging and discharging efficiency of the energy storage in the ith wind storage power station, E i,max and E i,min respectively represent the maximum and minimum state of charge allowed by the energy storage in the ith wind storage power station, E i,0 and E i,T Soc,i(0) and Soc,i(T) represent the initial state of charge and the final state of charge of the energy storage in the ith wind storage power station at the beginning of the scheduling period.

[0042] Further, the transformation of the master-slave game model into a single-layer conventional nonlinear programming model includes:

[0043] Solving the KKT conditions (Karush-Kuhn-Tucker conditions) of the second optimization model;

[0044] Linearly transforming the nonlinear terms in the obtained KKT conditions by the big M method;

[0045] Taking the KKT conditions obtained after completing the linear transformation as new constraint conditions, and adding them to the first optimization model to obtain a new optimization model;

[0046] In the new optimization model, the objective function is a nonlinear function, and therefore, a particle swarm algorithm is used to solve it.

[0047] Further, the KKT condition of the second optimization model is solved, and the solving comprises:

[0048] The second optimization model is rewritten in the following form:

[0049]

[0050] s.t.

[0051]

[0052] The matrix form expression is written as:

[0053] minK i x i

[0054] s.t.

[0055]

[0056] In the formula, K i is a coefficient of the objective function, that is, the corresponding coefficient matrix when the objective function -F i is written in the matrix form; x i represents a vector composed of decision variables, and specifically, In the constraint condition, formula (a) represents an inequality linear equation group composed of inequalities in the constraint condition, and is uniformly written in the form of less than or equal to the right side, and only the constant term is moved to the right side of the inequality, A i is a coefficient matrix corresponding to the inequality linear equation group when written in the matrix form, and s i represents a constant column vector corresponding to the inequality linear equation group when written in the matrix form, λ i is a dual variable corresponding to the inequality equation group, which is a column vector, and the vector length is the same as the number of rows of A i , that is, the equation number of the corresponding inequality equation group; formula (b) represents an equality linear equation group composed of equalities in the constraint condition, and all the non-zero terms are moved to the left side of the equality, B i is a coefficient matrix corresponding to the equality linear equation group when written in the matrix form, μ i is a dual variable corresponding to the inequality equation group, which is a column vector, and the vector length is the same as the number of rows of B i , that is, the equation number of the corresponding equality equation group;

[0057] According to the abstract expression of the second optimization model, the corresponding KKT condition can be written as:

[0058]

[0059] In the formula, the symbol ∇ represents partial derivative, and the superscript "'" represents the transpose of a vector or a matrix.

[0060] Further, λ i '(A i x i -s i ) = 0 is a nonlinear constraint, and is linearized by a large M method, and the transformation process is as follows:

[0061]

[0062] In the formula, z i is a column vector composed of 0-1 variables, and M is a preset sufficiently large positive number.

[0063] Further, the KKT condition obtained after the linear transformation is taken as a new constraint condition, and is added to the first optimization model to obtain a new optimization model as follows:

[0064]

[0065] s.t.

[0066]

[0067]

[0068] 0≤P t ≤P M ,t = 1, 2, 3,..., T

[0069]

[0070]

[0071]

[0072]

[0073] In the objective function of the new optimization model, c t P i,t is a nonlinear term, and is solved by using a particle swarm intelligent optimization algorithm to obtain the optimal day-ahead power curve of the price curve and each wind storage power station, wherein the optimal day-ahead power curve of each wind storage power station is the optimal day-ahead output planning curve of each wind storage power station.

[0074] In a second aspect, a distributed wind storage power station cluster scheduling method is provided, comprising:

[0075] The optimal day-ahead output planning curve of each wind storage power station is obtained by using the distributed wind storage power station cluster day-ahead output optimization method.

[0076] The output power of each wind storage power station is controlled according to the optimal day-ahead output planning curve of each wind storage power station.

[0077] In a third aspect, an electronic terminal is provided, comprising:

[0078] A memory storing a computer program;

[0079] A processor for loading and executing the computer program to implement the distributed wind storage power station cluster day-ahead output optimization method or the distributed wind storage power station cluster scheduling method.

[0080] In a fourth aspect, a computer readable storage medium storing a computer program is provided, and the computer program further implements the distributed wind storage power station cluster day-ahead output optimization method or the distributed wind storage power station cluster scheduling method when executed by a processor.

[0081] The present application provides a distributed wind storage power station cluster day-ahead output optimization method, a scheduling method, a terminal and a medium. From the perspective of wind storage power station and power grid scheduling, the characteristics of distributed wind storage power station cluster, such as decentralized arrangement and small single machine capacity, are combined to introduce the concept of wind storage power station cluster aggregation layer. An optimization model is established in the aggregation layer based on wind storage power station cluster, and an optimization model is established in the wind storage power station layer based on each wind storage power station. Then, a master-slave game model is constructed and solved by combining the two, to obtain the optimal day-ahead output planning curve of each wind storage power station. This not only improves the income of each wind storage power station in the day-ahead real-time energy market, but also reduces the power purchase cost of the grid side, and realizes the reasonable scheduling and efficient utilization of the large-scale distributed wind storage power station cluster. BRIEF DESCRIPTION OF DRAWINGS

[0082] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor based on these drawings.

[0083] Figure 1 It is a distributed wind storage power station cluster day-ahead output optimization method flowchart provided by the embodiments of the present application. DETAILED DESCRIPTION

[0084] In order to make the objects, technical solutions and advantages of the present application clearer, the technical solutions of the present application will be described in detail below. Obviously, the described embodiments are only some of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative work belong to the protection scope of the present application.

[0085] The present application also starts from the perspective of wind storage power station and power grid dispatching, introduces the concept of wind storage power station cluster aggregation layer in combination with the characteristics of the dispersed arrangement of distributed wind storage power station cluster and small single machine capacity, proposes a day-ahead output optimization scheme and dispatching scheme for the large-scale distributed wind storage power station cluster, which can not only improve the income of each wind storage power station in the day-ahead real-time energy market, but also reduce the power purchase cost of the power grid side, and realize the reasonable dispatching and efficient utilization of the large-scale distributed wind storage power station cluster. The aggregation layer is equivalent to an intermediary, which is a large wholesaler; and then each wind storage power station is equivalent to each terminal consumer. If each energy storage power station directly participates in the spot trading of the power grid, one is that the capacity of a single power station is too small, and the other is that the number is too large, so that the power grid side cannot realize the solution. The aggregation layer has its own energy storage system, which plays a role of more charging and less supplementing. The aggregation layer has a total power purchase amount from each wind storage station at each moment, and also has a total power purchase amount from each wind storage station, and the difference between the two purchase and sale values is charged by the energy storage system. If the purchase is greater than the sale, the energy storage system charges, and vice versa. The technical solutions of the present application will be described in detail below in combination with specific embodiments.

[0086] The embodiment of the present application provides a distributed wind storage power station cluster day-ahead output optimization method, mainly including two levels. The first level is called the wind storage power station cluster aggregation layer, and the main task is to develop a reasonable retail electricity price curve for each wind storage power station in combination with the day-ahead output prediction curve of all wind storage power stations in the wind storage power station cluster, so as to promote each wind storage power station to respond to the demand of the power grid through the price signal. The second level is the wind storage power station layer, and the purpose is that each wind storage power station develops a reasonable day-ahead output planning curve in combination with the retail electricity price curve provided by the aggregation layer according to the day-ahead output prediction curve of its own wind power and the capacity configuration of the energy storage, so as to maximize its own income. The specific steps will be described below.

[0087] A distributed wind storage power station cluster day-ahead output optimization method, comprising:

[0088] S1: The aggregation layer develops a retail electricity price for each wind storage power station in the distributed wind storage power station cluster according to the day-ahead output prediction curve of all wind storage power stations in the distributed wind storage power station cluster and the day-ahead wholesale electricity price of the electricity spot market, and takes the total power purchase of the aggregation layer, the total power sale of the aggregation layer, and the charging and discharging power of the energy storage system owned by the aggregation layer as the decision variables, and takes the minimum operation cost as the target, to establish a first optimization model.

[0089] Specifically, the first optimization model optimizes a time length of 24 hours of the next day, a time of 0:00-23:59 of the next day, and a target function of minimum running cost of the aggregation layer, and is expressed as follows:

[0090]

[0091] wherein F represents the running cost of the aggregation layer, t represents the serial number of the optimization period, T represents the total number of optimization periods, ΔT represents the time length of each optimization period, the product of T and ΔT is 24 hours, for example, when T = 24, the time length of each optimization period is 1 hour, i represents the number of the wind storage power station, G represents the number of the wind storage power stations, c t is a decision variable, represents the retail electricity price of each period formulated by the aggregation layer for each wind storage power station, P i,t is a decision variable, represents the power purchase of the aggregation layer from the i-th wind storage power station in the t-th period, that is, the power sale of the i-th wind storage power station in the t-th period, and is also the output power of the i-th wind storage power station in the t-th period, π t represents the power sale price of the aggregation layer to the power grid in the t-th period, in the model, the aggregation layer is considered as a price accepter in the power day-ahead spot market of the power grid, therefore, the power sale price is a known quantity, which is called the day-ahead wholesale power price of the power spot market, P t represents the power sale power of the aggregation layer to the power grid in the t-th period.

[0092] The first optimization model needs to satisfy the following constraint conditions:

[0093] Retail electricity price upper and lower limit constraint:

[0094]

[0095] wherein, represents the minimum retail electricity price limit value in the t-th period, represents the maximum retail electricity price limit value in the t-th period;

[0096] The cumulative sum of the retail electricity price in a day is a constant constraint:

[0097]

[0098] wherein C is a constant, represents the limit value of the cumulative sum of the retail electricity price, and is mainly used to limit the market power of the aggregator and protect the interests of each wind storage power station in the wind storage power station layer;

[0099] The power sale power upper and lower limit constraint of the aggregation layer to the power grid:

[0100] 0≤P t ≤P M ,t=1,2,3,...,T

[0101] In the formula, P M represents the maximum allowed value of the power sold to the grid by the aggregation layer;

[0102] Supply-demand balance constraint of the aggregation layer:

[0103]

[0104] In the formula, represents the discharging power of the energy storage system owned by the aggregation layer in the tth time period; t c represents the charging power of the energy storage system owned by the aggregation layer in the tth time period;

[0105] Operation constraint of the energy storage system owned by the aggregation layer:

[0106]

[0107] In the formula, the first two rows represent the upper and lower limit constraints of the discharging and charging power of the energy storage system, and the complementary constraint of charging and discharging, the third row represents the state of charge transfer equation of the energy storage system, the fourth row represents the upper and lower limit constraint of the state of charge of the energy storage system, and the last row represents that the state of charge of the energy storage system remains equal at the beginning and end of the scheduling period; P max represents the maximum charging and discharging power allowed by the energy storage system; u t is a 0-1 binary variable used to prevent the energy storage from charging and discharging at the same time; E t represents the state of charge of the energy storage system at the end of the tth time period; η c and η d respectively represent the charging and discharging efficiency of the energy storage system; E max and E min respectively represent the maximum and minimum state of charge allowed by the energy storage system; E0and E T respectively represent the initial state of charge and the final state of charge of the energy storage system in the scheduling period.

[0108] S2: Each wind-storage power station in the wind-storage power station layer formulates a retail electricity price according to the retail electricity price formulated by the aggregation layer, and combines the day-ahead output prediction curve and the configuration capacity of the energy storage, to establish a second optimization model with the discharging power of the energy storage in the wind-storage power station, the charging power of the energy storage in the wind-storage power station, the power generation revenue of the wind-storage power station as the decision variables, and the maximum power generation revenue as the target.

[0109] Specifically, the optimization time length of the second optimization model is also 24 hours of the next day, the time is also 0:00-23:59 of the next day, and the target function is the maximum power generation revenue of each wind-storage power station, which is represented as follows:

[0110]

[0111] In the formula, F irepresents the generation revenue of the ith wind storage power station, the first term on the right side of the equation represents the revenue of the wind storage power station from selling electricity, and the second term on the right side of the equation is a penalty term for avoiding simultaneous charging and discharging of the energy storage, and the penalty coefficient ρ > 0, represents the discharging power of the energy storage in the ith wind storage power station at the tth time period, represents the charging power of the energy storage in the ith wind storage power station at the tth time period.

[0112] The second optimization model needs to satisfy the following constraint conditions:

[0113] The wind storage power station power balance constraint:

[0114]

[0115] In the formula, represents the power generation of the wind turbine in the ith wind storage power station;

[0116] The wind turbine power generation constraint:

[0117]

[0118] In the formula, represents the predicted power generation of the wind turbine in the ith wind storage power station at the tth time period, which is obtained according to the day-ahead output prediction curve of the wind storage power station;

[0119] The energy storage operation constraint:

[0120]

[0121] In the formula, the first two rows represent the upper and lower limit constraints of the discharging and charging power of the energy storage, the third row represents the state of charge transfer equation of the energy storage, the fourth row represents the upper and lower limit constraints of the state of charge of the energy storage, and the last row represents that the state of charge of the energy storage at the beginning and end of the dispatching period is equal; P i max represents the maximum charging and discharging power of the energy storage in the ith wind storage power station, E i,t represents the state of charge of the energy storage in the ith wind storage power station at the end of the tth time period, and respectively represent the charging and discharging efficiency of the energy storage in the ith wind storage power station, E i,max and E i,min respectively represent the maximum and minimum state of charge of the energy storage in the ith wind storage power station, E i,0 and E i,T represents the initial state of charge and the final state of charge of the energy storage in the ith wind storage power station at the dispatching period.

[0122] S3: the first optimization module in the aggregation layer and the second optimization module in the wind storage power station layer constitute a master-slave game model, through transformation of the master-slave game model, a single-layer conventional nonlinear programming model is converted, and then a particle swarm algorithm is used for solving, so that the optimal day-ahead power selling curve of each wind storage power station is obtained, that is, the optimal day-ahead output planning curve of each wind storage power station, and a win-win effect of optimal grid side cost and maximum power station side benefit is realized.

[0123] Wherein, the transformation of the master-slave game model to the single-layer conventional nonlinear programming model comprises:

[0124] S31: solving the KKT condition of the second optimization model, specifically comprising:

[0125] The second optimization model is rewritten as follows:

[0126]

[0127] s.t.

[0128]

[0129] The matrix form expression is:

[0130] minK i x i

[0131] s.t.

[0132]

[0133] In the formula, K i is the coefficient of the objective function, that is, the corresponding coefficient matrix when the objective function-F i is written in matrix form; x i represents a vector composed of decision variables, specifically In the constraint condition, formula (a) represents an inequality linear equation group composed of inequalities in the constraint condition, and is uniformly written in the form of left less than or equal to right, and only the constant term is moved to the right side of the inequality, A i is the coefficient matrix corresponding to the inequality linear equation group when written in matrix form, and s i represents the constant column vector corresponding to the inequality linear equation group when written in matrix form, and λ i is the dual variable corresponding to the inequality equation group, which is a column vector, and the vector length is the same as the number of rows of A i , that is, the number of equations of the corresponding inequality equation group; formula (b) represents an equality linear equation group composed of equalities in the constraint condition, and all non-zero terms are moved to the left side of the equation, B iFor writing in matrix form, the coefficient matrix of the equation system of linear equations, μ i For the corresponding dual variable of the inequality equation system, B i is a column vector, the length of the vector is the same as the number of rows of B

[0134] According to the abstract expression of the second optimization model, the corresponding KKT condition can be written as:

[0135]

[0136] In the formula, The symbol represents the partial derivative, and the superscript "'" represents the transpose of the vector or matrix.

[0137] S32: Linear transformation of the nonlinear term in the obtained KKT condition by the big M method. Specifically, it includes:

[0138] The last row of the KKT condition λ i '(A i x i -s i ) = 0 is a nonlinear constraint, so it is linearized by the big M method, and the transformation process is as follows:

[0139]

[0140] In the formula, z i is a column vector composed of 0-1 variables, and M is a sufficiently large positive number, taking an empirical value.

[0141] S33: Take the KKT condition obtained after linear transformation as a new constraint condition, and add it to the first optimization model to obtain a new optimization model. The new optimization model is as follows:

[0142]

[0143] s.t.

[0144]

[0145]

[0146] 0≤P t ≤P M ,t=1,2,3,...,T

[0147]

[0148]

[0149]

[0150]

[0151] S34: In the new optimization model, the objective function is a nonlinear function, so the particle swarm algorithm is used for solving.

[0152] In the objective function of the new optimization model, c t P i,t is a nonlinear term, so it is difficult to directly use a mature commercial solver for accurate solution, so the particle swarm intelligent optimization algorithm is used for solving, to obtain the retail electricity price curve and the optimal day-ahead power selling curve of each wind storage power station, wherein the optimal day-ahead power selling curve of each wind storage power station is the optimal day-ahead output planning curve of each wind storage power station.

[0153] It should be noted that steps S1 and S2 can be executed synchronously, or step S2 can be executed first and then step S1. The distributed wind storage power station cluster day-ahead output optimization method provided in the above embodiment simultaneously starts from the perspectives of wind storage power stations and grid dispatching, combines the characteristics of distributed wind storage power station clusters, such as decentralized arrangement and small single-machine capacity, introduces the concept of a wind storage power station cluster aggregation layer, establishes an optimization model based on wind storage power station clusters in the aggregation layer, establishes an optimization model based on each wind storage power station in the wind storage power station layer, then combines the two to build a master-slave game model and solve it, to obtain the optimal day-ahead output planning curve of each wind storage power station, which not only improves the income of each wind storage power station in the day-ahead real-time energy market, but also reduces the power purchase cost of the grid side, to realize reasonable dispatching and efficient utilization of the large-scale distributed wind storage power station cluster.

[0154] Based on the distributed wind storage power station cluster day-ahead output optimization method provided in the above embodiment, the embodiment of the present application further provides a distributed wind storage power station cluster dispatching method, which comprises:

[0155] Step 1: obtaining the optimal day-ahead output planning curve of each wind storage power station by using the distributed wind storage power station cluster day-ahead output optimization method as described above;

[0156] Step 2: controlling the output power of each wind storage power station according to the optimal day-ahead output planning curve of each wind storage power station, to realize the win-win effect of optimal cost of the grid side and maximum income of the power station side, and to realize reasonable dispatching and efficient utilization of the large-scale distributed wind storage power station cluster.

[0157] The embodiment of the present application further provides an electronic terminal, which comprises:

[0158] a memory storing a computer program;

[0159] a processor for loading and executing the computer program to realize the distributed wind storage power station cluster day-ahead output optimization method or the distributed wind storage power station cluster dispatching method as described above.

[0160] The embodiment of the present application also provides a computer readable storage medium storing a computer program, which further implements the distributed wind storage power station cluster day-ahead output optimization method or the distributed wind storage power station cluster scheduling method when executed by a processor.

[0161] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a terminal or a computer program product. Therefore, the present application can adopt a completely hardware embodiment, a completely software embodiment or an embodiment combining software and hardware aspects. Moreover, the present application can adopt a computer program product in the form of being implemented on one or more computer usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer usable program codes.

[0162] The present application is described with reference to flowcharts and / or block diagrams of the methods, devices (systems) and computer program products according to the embodiments of the present application. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and the combination of the flows and / or blocks in the flowcharts and / or block diagrams can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device produce a device implemented in the flowcharts and / or block diagrams. Figure 1 The function specified in one flow or multiple flows and / or blocks Figure 1 The device for implementing the function specified in one block or multiple blocks.

[0163] These computer program instructions can also be stored in a computer readable storage medium capable of guiding the computer or other programmable data processing device to work in a specific way, so that the instructions stored in the computer readable storage medium produce a product including instruction devices, which implement the flowcharts and / or block diagrams. Figure 1 The function specified in one flow or multiple flows and / or blocks Figure 1 The device for implementing the function specified in one block or multiple blocks.

[0164] These computer program instructions can also be loaded into a computer or other programmable data processing device, so that a series of operation steps are performed on the computer or other programmable device to produce a computer implemented process, so that the instructions executed on the computer or other programmable device provide a process for implementing the flowcharts and / or block diagrams. Figure 1 The function specified in one flow or multiple flows and / or blocks Figure 1 The device for implementing the function specified in one block or multiple blocks.

[0165] It can be understood that the same or similar parts in the above-mentioned embodiments can be mutually referenced, and the contents not described in detail in some embodiments can refer to the same or similar contents in other embodiments.

[0166] Although the embodiments of the present application have been shown and described above, it can be understood that the above-mentioned embodiments are exemplary and cannot be understood as limiting the present application, and the ordinary skilled in the art can make changes, modifications, replacements and variations to the above-mentioned embodiments within the scope of the present application.

Claims

1. A method for day-ahead power output optimization of a distributed wind storage power station cluster, characterized in that, The first optimization model of the aggregation layer has an optimization time length of 24 hours of the next day, and a target function of minimum operation cost of the aggregation layer, and is expressed as follows: The constraint conditions of the first optimization model and corresponding mathematical expressions thereof are as follows: Retail electricity price upper and lower limit constraint: The cumulative sum of the retail electricity price in a day is a constant constraint:

2. The distributed wind storage power station cluster day-ahead output optimization method according to claim 1, characterized in that, In the formula, C is a constant, representing the limit value of the cumulative sum of the retail electricity price; ； Wherein, F represents the operation cost of the aggregation layer; t represents the serial number of the optimization period; T represents the total number of optimization periods; ΔT represents the time length of each optimization period, i represents the number of the wind storage power station; G represents the number of the wind storage power station; represents the retail electricity price of each period formulated by the aggregation layer for each wind storage power station, represents the electricity purchase power of the aggregation layer to the ith wind storage power station in the tth period, that is, the electricity sale power of the ith wind storage power station in the tth period; represents the electricity sale price of the aggregation layer to the power grid in the tth period, represents the electricity sale power of the aggregation layer to the power grid in the tth period; The upper and lower limits of the power sold by the aggregation layer to the power grid constraint: The supply and demand balance constraint of the aggregation layer: ； wherein LRTVtrepresents the lowest retail electricity price limit for the tth time period, HRTVtrepresents the highest retail electricity price limit for the tth time period; The operation constraint of the energy storage system owned by the aggregation layer: ； The second optimization model also has an optimization time length of 24 hours of the next day, and a target function of maximum power generation income of each wind storage power station, and is expressed as follows: The constraint conditions of the second optimization model and corresponding mathematical expressions thereof are as follows: ； In the formula, represents the maximum allowed value of the power sold to the grid by the aggregation layer. The wind storage power station power balance constraint: ； wherein denotes the discharge power of the energy storage system owned by the aggregation layer at the t-th time interval, denotes the charge power of the energy storage system owned by the aggregation layer at the t-th time interval; The wind turbine power generation power constraint: ； wherein, Pmax, tdenotes the maximum charge and discharge power allowed by the energy storage system, is a 0-1 binary variable, S0denotes the state of charge of the energy storage system at the end of the tth time period, and P0and P1denote the charge and discharge efficiency of the energy storage system, respectively, and Sminand Smaxdenote the minimum and maximum state of charge allowed by the energy storage system, respectively, and S0and S1denote the initial and final state of charge of the energy storage system for the dispatch period, respectively.

3. The distributed wind storage plant cluster day-ahead output optimization method according to claim 2, characterized in that, The energy storage operation constraint: ； wherein, represents the generation revenue of the ith wind storage power station, and the penalty coefficient ρ > 0, represents the discharging power of the energy storage in the ith wind storage power station at the t th time period, represents the charging power of the energy storage in the ith wind storage power station at the t th time period; The transformation of the master-slave game model into a single-layer conventional nonlinear programming model includes: Solving the KKT condition of the second optimization model; ； In the formula, Pi represents the power generated by the wind turbine in the ith wind storage power station; Linearly transforming the nonlinear terms in the obtained KKT condition by the big M method; ； In the formula, represents the predicted power generation of the wind turbine in the ith wind storage power station in the tth time period. Taking the KKT condition obtained after the linear transformation as a new constraint condition, and adding it to the first optimization model to obtain a new optimization model; ； wherein, denotes the maximum charge and discharge power allowed for the energy storage in the ith wind storage power plant, denotes the state of charge of the energy storage in the ith wind storage power plant at the end of the t-th time period, and denote the charge and discharge efficiency of the energy storage in the ith wind storage power plant, respectively, and denote the maximum and minimum state of charge allowed for the energy storage in the ith wind storage power plant, respectively, and denote the initial state of charge and the final state of charge of the energy storage in the ith wind storage power plant at the beginning and at the end of the dispatching period.

4. The distributed wind storage plant cluster day-ahead output optimization method according to claim 3, characterized in that, In the new optimization model, the target function is a nonlinear function, and therefore, the particle swarm algorithm is used for solving. The solving of the KKT condition of the second optimization model includes: Rewriting the second optimization model into the following form: The matrix form expression is as follows: According to the abstract expression of the second optimization model, the corresponding KKT condition can be written as:

5. The distributed wind storage plant cluster day-ahead output optimization method according to claim 4, characterized in that, Taking the KKT condition obtained after the linear transformation as a new constraint condition, and adding it to the first optimization model to obtain a new optimization model as follows: The method comprises the following steps: ； Obtaining the optimal day-ahead output planning curve of each wind storage power station by using the method according to any one of claims 1 to 7; ； where K i is the coefficient of the objective function, i.e., the objective function-F i is written in matrix form; x i represents a vector composed of decision variables, specifically , t = 1, 2, 3, …, T; formula (a) in the constraint condition represents an inequality linear equation system composed of inequalities in the constraint condition, and is uniformly written in the form of less than or equal to the right side, and only the constant term is moved to the right side of the inequality, A i is the coefficient matrix corresponding to the inequality linear equation system when written in matrix form, s i represents the constant column vector corresponding to the inequality linear equation system when written in matrix form, λ i is the dual variable corresponding to the inequality equation system, which is a column vector, and the vector length is the same as the number of rows of A i ; formula (b) represents an equality linear equation system composed of equalities in the constraint condition, and all non-zero terms are moved to the left side of the equal sign, B i is the coefficient matrix corresponding to the equality linear equation system when written in matrix form, μ i is the dual variable corresponding to the inequality equation system, which is a column vector, and the vector length is the same as the number of rows of B i ; Controlling the output power of each wind storage power station according to the optimal day-ahead output planning curve of each wind storage power station. ； wherein The symbol denotes partial differentiation, and the superscript "T" denotes the transpose of a vector or matrix.

6. The distributed wind storage plant cluster day-ahead output optimization method according to claim 5, characterized in that, In the KKT conditions is a nonlinear constraint, so it is linearized by the large M method, and the transformation process is as follows: ； where z i is a column vector of 0-1 variables, and M is a predetermined positive number.

7. The distributed wind storage plant cluster day-ahead dispatch method of claim 6, wherein, The method comprises the following steps: ； The objective function of the new optimization model The objective function of the new optimization model The objective function of the new optimization model The objective function of the new optimization model The objective function of the new optimization model The objective function of the new optimization model The objective function of the new optimization model The objective function of the new optimization model The objective function of the new optimization model The objective function of the new optimization model The objective function of the new optimization model The objective function of the new optimization model The objective function of the new optimization model The objective function of the new 8. A distributed wind storage power station cluster scheduling method, characterized in that, A memory storing a computer program; ​ ​ 9. An electronic terminal, characterized in that ​ ​ A processor for loading and executing the computer program to implement the method of any of claims 1 to 8.

10. A computer-readable storage medium storing a computer program, characterized in that, The computer program is further processor-executed to implement the method of any of claims 1 to 8.

Images