A method for coordinated optimal dispatching of upper and lower level power grids based on alternating direction method of multipliers

Abstract

The application discloses a kind of upper and lower grid collaborative optimization scheduling methods based on alternating direction multiplier method, and relate to power system dispatching technical field.The method includes separately establishing upper and lower grid optimization scheduling model according to grid loss, load side demand response and light abandonment penalty;Nash negotiation theory is combined with upper and lower grid optimization scheduling model to form upper and lower grid collaborative optimization Nash negotiation model;Upper and lower grid collaborative optimization Nash negotiation model is solved, and the result is optimized according to the result of solving scheduling.It can consider the benefit of each subject and alliance of upper and lower grid and protect the privacy of each subject of upper and lower grid by introducing Nash negotiation theory and using alternating direction multiplier method.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of power system dispatching technology, and in particular to a method for coordinated optimization dispatching of upper and lower level power grids based on an alternating direction multiplier method. BACKGROUND

[0002] Under the drive of the "double carbon" goal, more and more new energy power is connected to the power grid. Although large-scale new energy grid connection greatly reduces the carbon emissions of the power system, the uncertainty and limited predictability thereof increase the pressure of power consumption, and the power system needs to balance the energy supply and demand between the randomly fluctuating load demand and the randomly fluctuating power supply. The structure, operation control mode, and planning, construction and management thereof will undergo fundamental changes, forming a new generation of power system mainly composed of new energy power production, transmission and consumption.

[0003] As an important part of renewable energy, photovoltaic power generation is developing rapidly in China. After large-scale photovoltaic grid connection, the randomness and intermittency of photovoltaic output have caused certain influence on the safe operation and power quality of the power grid. Therefore, it is necessary to study and propose an active power dispatching method suitable for large-scale photovoltaic access in combination with the characteristics of photovoltaic output, so as to meet the safety and stability requirements of the power system after photovoltaic access.

[0004] At present, the research on photovoltaic system generation dispatching mainly focuses on photovoltaic power prediction and dispatching model, and there are many related researches on power system dispatching methods containing photovoltaic systems. However, most of these researches are based on photovoltaic consumption and unit dispatching within a single region, and there are few researches on coordinated consumption and dispatching of photovoltaic power between two or more levels. With the increase of photovoltaic access capacity to the power system, photovoltaic dispatching based on a single region cannot meet the safety and stability requirements of the power system.

[0005] Therefore, in order to improve the photovoltaic consumption capacity of the system and ensure the stability and economy of the power system, it is necessary to study the optimization dispatching problem of coordinated consumption of photovoltaic power between two or more levels. SUMMARY

[0006] In order to overcome the shortcomings in the existing research and improve the overall photovoltaic consumption rate of the power grid, the present application provides a method for coordinated optimization dispatching of upper and lower level power grids based on an alternating direction multiplier method.

[0007] In the first aspect, the method for coordinated optimization dispatching of upper and lower level power grids provided by the present application adopts the following technical scheme:

[0008] A method for coordinated optimization dispatching of upper and lower level power grids based on an alternating direction multiplier method, comprising:

[0009] According to the grid loss, load side demand response and light penalty, the upper and lower grid optimization scheduling models are respectively established;

[0010] The Nash negotiation theory is combined with the upper and lower grid optimization scheduling models to form an upper and lower grid collaborative optimization Nash negotiation model;

[0011] The upper and lower grid collaborative optimization Nash negotiation model is solved, and the scheduling is optimized according to the solving result.

[0012] Further technical solutions are that when the upper and lower grid collaborative optimization Nash negotiation model is solved, the following steps are included:

[0013] The upper and lower grid collaborative optimization Nash negotiation model is converted into an upper and lower grid alliance benefit maximization model and an electric energy transaction payment negotiation model;

[0014] The Lagrange multiplier and the penalty factor are introduced to obtain an augmented Lagrange function of the alliance benefit maximization model objective function of the upper and lower grid alliance benefit maximization model objective function;

[0015] According to the ADMM algorithm principle, the augmented Lagrange function of the alliance benefit maximization model objective function is decomposed to obtain a distributed optimization operation model of the lower grid and the upper grid, and the optimal expected transaction electric quantity of the upper grid and each lower grid is obtained after iterative calculation;

[0016] The optimal expected transaction electric quantity of the upper grid and each lower grid is substituted into the electric energy transaction payment negotiation model, the Lagrange multiplier and the penalty factor are introduced to obtain an augmented Lagrange function of the electric energy transaction payment negotiation model objective function;

[0017] According to the ADMM algorithm principle, the augmented Lagrange function of the electric energy transaction payment negotiation model objective function is decomposed to obtain a distributed optimization model of the electric energy transaction price of each subject of the upper grid and the lower grid, and the optimal transaction electric price of the upper grid and each lower grid is obtained after iterative calculation.

[0018] Further technical solutions are that the upper and lower grid optimization scheduling model includes an upper grid subject benefit maximization operation model and a lower grid subject benefit maximization operation model;

[0019] The objective function of the upper grid subject benefit maximization operation model is:

[0020] maxU up =U d1,g +U d2,g -C up,m -C up,loss

[0021] In the formula, U d1,g and U d2,gThe electricity sales revenue of the subordinate power grid 1 and the subordinate power grid 2 respectively; C up,m The operation cost of the superior power grid; C up,loss The network loss cost of the superior power grid;

[0022] The objective function of the subordinate power grid main benefit maximization operation model is:

[0023] maxU di = -U di,g -C di,m -C di,loss (i = 1, 2)

[0024] In the formula, i = 1, 2 respectively represent the subordinate power grid 1 and the subordinate power grid 2; U di,g , C di,m , C di,loss The electricity purchase cost, the operation cost and the network loss cost of the subordinate power grid i to the superior power grid respectively.

[0025] Further technical solutions are that the objective function of the superior and subordinate power grid cooperative optimization Nash negotiation model is:

[0026]

[0027] In the formula, The optimal operation benefit of the non-main body of the superior and subordinate power grid when not cooperating, that is, the Nash negotiation breaking point, is a constant.

[0028] Further technical solutions are that the objective function of the superior and subordinate power grid cooperative optimization Nash negotiation model is:

[0029] maxω up + ω d1 + ω d2

[0030] The objective function of the electricity transaction payment negotiation model is:

[0031]

[0032] In the formula, The optimal solution of the superior and subordinate power grid cooperative optimization Nash negotiation model.

[0033] Further technical solutions are that the augmented Lagrange function of the coalition benefit maximization model objective function is:

[0034]

[0035] The objective function of the subordinate power grid distributed optimization operation model is:

[0036]

[0037] The objective function of the upper-level power grid distributed optimization operation model is:

[0038]

[0039] Further, the augmented Lagrange function of the electric energy transaction payment negotiation model objective function is:

[0040]

[0041] The objective function of the lower-level power grid electric energy transaction price distributed optimization model is:

[0042]

[0043] The objective function of the upper-level power grid electric energy transaction price distributed optimization model is:

[0044]

[0045] In a second aspect, the present application provides a scheduling method using the following technical solution:

[0046] A scheduling system comprises:

[0047] A first modeling unit is configured to separately establish upper-level and lower-level power grid optimization scheduling models according to power grid loss, load side demand response, and light abandonment penalty.

[0048] A second modeling unit is configured to combine Nash negotiation theory with the upper-level and lower-level power grid optimization scheduling models to form an upper-level and lower-level power grid collaborative optimization Nash negotiation model.

[0049] A solving unit is configured to solve the upper-level and lower-level power grid collaborative optimization Nash negotiation model and optimize scheduling according to the solution.

[0050] In a third aspect, the present application provides an intelligent terminal using the following technical solution:

[0051] An intelligent terminal comprises a memory and a processor, and the memory stores a computer program capable of being loaded and executed by the processor to perform any one of the first aspect-based upper-level and lower-level power grid collaborative optimization scheduling methods based on the alternating direction multiplier method.

[0052] In a fourth aspect, the present application provides a computer readable storage medium using the following technical solution:

[0053] A computer readable storage medium stores a computer program capable of being loaded and executed by a processor to perform any one of the first aspect-based upper-level and lower-level power grid collaborative optimization scheduling methods based on the alternating direction multiplier method.

[0054] The beneficial effects produced by the above technical solutions are as follows:

[0055] 1. In view of the limitation of photovoltaic consumption based on a single area in the prior research, a top-down power grid collaborative optimization scheduling method based on an alternating direction multiplier method is proposed, so as to realize coordinated consumption of photovoltaic power in a multi-level power grid and improve the overall photovoltaic consumption rate.

[0056] 2. The Nash negotiation theory is introduced to establish a top-down power grid collaborative optimization Nash negotiation model, and the model is converted into two sub-problems, so as to simultaneously consider the individual and alliance overall benefits of the top-down power grid.

[0057] 3. The alternating direction multiplier method (ADMM) is used to solve the two sub-problems, so as to protect the privacy of the top-down power grid individual while solving the model. BRIEF DESCRIPTION OF DRAWINGS

[0058] Figure 1 is a method flowchart provided by an embodiment of the present application;

[0059] Figure 2 is another method flowchart provided by an embodiment of the present application;

[0060] Figure 3 is a top-down power grid interactive operation framework diagram provided by an embodiment of the present application. DETAILED DESCRIPTION

[0061] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. The description of the at least one exemplary embodiment is actually only illustrative, but not as any limitation on the present application and its application or use. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work belong to the scope of protection of the present application.

[0062] In the following description, many specific details are set forth in order to provide a thorough understanding of the present application, but the present application can also be implemented in other ways different from those described herein, and those skilled in the art can make similar generalizations without departing from the concept of the present application, so the present application is not limited to the specific embodiments disclosed below.

[0063] Embodiment 1

[0064] The present application discloses a top-down power grid collaborative optimization scheduling method based on an alternating direction multiplier method, including the following steps,

[0065] Please refer to Figure 1 :

[0066] Step S100: According to the grid loss, load side demand response and curtailment penalty, an upper and lower grid optimization scheduling model is established separately.

[0067] Compared with the traditional passive distribution network, the active distribution network contains a large number of small distributed power sources. With the development of active distribution network, the functions of two-level grid are further integrated, and power interaction occurs between upper and lower grid, such as Figure 3 The upper and lower grid interaction operation framework in the embodiment. Through the interaction between the upper and lower grid, a multi-level management system of grid dispatching is established, which can reduce the grid loss and maximize the consumption of renewable energy power, and improve the overall new energy consumption rate.

[0068] The distributed power source of the lower grid is a gas turbine and photovoltaic. At the same time, the demand response resources of the lower grid are rich, and reasonable use of demand response resources can increase the flexibility of the system, including transferable load and reducible load. In addition, the active line loss caused by line transmission cannot be ignored.

[0069] Due to the strong uncertainty of photovoltaic output of the upper and lower grid, the random planning method is adopted in the embodiment. First, 1000 photovoltaic output scenarios of each grid are simulated by Monte Carlo random sampling technology, then the k-means clustering method is used to reduce the photovoltaic output scenarios to 6 typical scenarios, and the probability of each scenario is calculated to simulate the uncertainty of photovoltaic output of each grid.

[0070] The objective function of the lower grid main benefit maximization operation model is:

[0071] maxU di =-U di,g -C di,m -C di,loss (i=1,2) (1)

[0072] In the formula, i=1, 2 represent lower grid 1 and lower grid 2 respectively; U di,g , C di,m , C di,loss are the purchase cost of lower grid i to upper grid, the operation cost of lower grid and the grid loss cost of lower grid.

[0073] The constraint condition of the lower grid main benefit maximization operation model is:

[0074]

[0075]

[0076]

[0077]

[0078]

[0079]

[0080] wherein, is s di the probability of the scenario; S di is the total number of photovoltaic output scenarios of the subordinate grid i; is the electricity purchase price of the subordinate grid i to the superior grid at time t; is the electricity purchase quantity of the subordinate grid i to the superior grid at time t; are respectively the gas turbine power generation quantity, the photovoltaic predicted power generation quantity, the photovoltaic actual output, the translatable load quantity and the reducible load quantity of the subordinate grid i at time t; k gt , are respectively the unit power cost coefficient of the gas turbine and the light abandonment penalty coefficient; k dr1 , k dr2 is the demand response coefficient, and are respectively the translatable load compensation coefficient and the reducible load compensation coefficient; is the active power loss of the subordinate grid i at time t; is the loss penalty coefficient; is the load power after the demand response of the subordinate grid; are respectively the minimum output and the maximum output of the gas turbine; T is the number of time periods for dividing a day, and is 24.

[0081] The superior grid includes coal-fired units and photovoltaics, and includes operation cost and superior grid loss cost; the operation benefit is the electricity sale benefit of the subordinate grid 1 and the subordinate grid 2 subjects. The objective function of the superior grid subject benefit maximization operation model can be expressed as:

[0082] max U up = U d1,g + U d2,g - C up,m - C up,loss (8)

[0083] wherein, U d1,g and U d2,g are respectively the electricity sale benefit of the subordinate grid 1 and the subordinate grid 2; C up,m is the operation cost of the superior grid; C up,loss is the loss cost of the superior grid.

[0084] The constraint condition of the superior grid subject benefit maximization operation model is:

[0085]

[0086]

[0087]

[0088]

[0089] In the formula: tau up The probability of s up The scene; S up The total number of upper grid photovoltaic output scenes; The jth coal-fired unit output at t time; a j , b j , c j The jth coal-fired unit operation cost coefficient; The upper grid photovoltaic predicted output and actual output at t time; The upper grid light penalty coefficient; The upper grid active power loss and loss cost coefficient, respectively; The s up The upper grid main body sells electricity to the lower grid 1, 2 main body at t time; The load power after the upper grid demand response.

[0090] Step S200: Combine Nash negotiation theory with upper and lower grid optimization scheduling model to form upper and lower grid cooperative optimization Nash negotiation model.

[0091] It is assumed that the upper and lower grids belong to different interest subjects, that is, they are independent and rational individuals. Each subject of the upper and lower grid can reach an agreement through bargaining negotiation, so that the benefits of each subject are improved.

[0092] The basic principles of Nash negotiation are described here. The embodiment of the application assumes that each interest subject is an independent rational individual, and each interest subject hopes to seek a balanced strategy through negotiation to maximize the benefits of each interest subject. As shown in formula (13), Nash negotiation theory belongs to the category of cooperative game, and can take into account individual and collective interests. The solution that maximizes Nash product is the balanced solution of Nash negotiation game, and the Nash negotiation solution can make the participants of the cooperative alliance obtain Pareto optimal benefits.

[0093]

[0094] In formula (13): N is the total number of negotiation subjects; U n The benefit of negotiation subject n; The negotiation breaking point, that is, the benefit before the participating subjects cooperate in negotiation; Indicates the benefit improvement value obtained by negotiation participating subject n through cooperation. The goal of Nash negotiation is to maximize the benefit improvement value of all cooperative subjects.

[0095] For the cooperative operation problem of each subject of the upper and lower power grids, a Nash negotiation model for the cooperative operation of the upper and lower power grids can be obtained from the Nash negotiation theory, and the objective function is expressed as follows:

[0096]

[0097] In the formula, the formula under s.t. is the constraint condition of the objective function; are the optimal operation benefits of the non-subjects of the upper and lower power grids when there is no cooperation, i.e., the Nash negotiation breakdown point, which is a constant, and the solving model is shown in formulas (15) and (16):

[0098] (1) Upper and lower power grid 1, 2 individual operation income model

[0099]

[0100] In the formula, is the income when the lower power grid i is operated individually; are the operation cost and network loss cost when the lower power grid i is operated individually; are the gas turbine output, actual photovoltaic output, translatable load amount, and reducible load amount when the lower power grid i is operated individually; is the active network loss when the lower power grid i is operated individually.

[0101] (2) Upper power grid individual operation income model

[0102]

[0103] In the formula, is the income when the upper power grid is operated individually; are the operation cost and network loss cost when the upper power grid is operated individually; are the coal-fired unit output and actual photovoltaic output of the upper power grid when the upper power grid is operated individually; is the active network loss when the upper power grid is operated individually.

[0104] Step S300: Solving the upper and lower power grid cooperative optimization Nash negotiation model, and optimizing the dispatching according to the solving result.

[0105] Wherein, when solving the upper and lower power grid cooperative optimization Nash negotiation model, the following steps are specifically included.

[0106] Please refer to Figure 2 :

[0107] Step S310: Converting the upper and lower power grid cooperative optimization Nash negotiation model into an upper and lower power grid alliance benefit maximization model and an electric energy transaction payment negotiation model.

[0108] The Nash negotiation model for the cooperative operation of the upper and lower power grids is essentially a non-convex nonlinear optimization problem, which is difficult to solve directly. Therefore, the Nash negotiation model for the cooperative operation of the upper and lower power grids, i.e., formula (13), is equivalent to two sub-problems or sub-models: an upper and lower power grid alliance benefit maximization sub-model and an electric energy transaction payment negotiation sub-model.

[0109] Sub-model 1: The objective function of the upper and lower power grid alliance benefit maximization sub-model is represented by formula (17):

[0110]

[0111] Sub-model 2: The objective function of the electric energy transaction payment negotiation sub-model is represented by formula (18):

[0112]

[0113] In the formula: is the optimal solution of sub-model 1. The equivalence proof process is as follows:

[0114] It is known that the mean inequality is:

[0115]

[0116] In the formula: R+ is a set of non-negative real numbers, and the equality holds only when a1=a2=…=a m

[0117] Solving the maximum value of model (14) requires satisfying the “one positive, two fixed, and three equal” conditions of the mean inequality. Among them, “one positive” means that in model (14), the improved benefit value after the cooperation of the upper power grid and the two lower power grids, i.e., are positive values; “two fixed” means that the total improved benefit value after the cooperation of the upper power grid and the two lower power grids, i.e., is a fixed value; “three equal” means that all are equal, the objective function reaches the maximum value, i.e., the improved benefit value after cooperation is equally divided by the upper power grid, the lower power grid 1, and the lower power grid 2.

[0118] Model (14) has already satisfied the “one positive” condition. If is a fixed value, then model (14) satisfies the “two fixed” condition, and the objective function reaches the maximum value. Therefore, it is obtained that:

[0119]

[0120] Therefore, if U up +U d1 +U d2 is a fixed value, the condition for the mean inequality to reach the maximum value can be satisfied, and U​up +U d1 +U d2 The greater the value, the greater the objective function value of the model (14), let:

[0121]

[0122] Then:

[0123]

[0124] So:

[0125]

[0126] Equation (23) is to maximize the benefits of the upper and lower grid alliance.

[0127] By solving the alliance benefit maximization sub-problem 1, the optimal value is obtained: And Substituting it into equation (14) gives:

[0128]

[0129] Since the natural logarithm is a strictly monotonic increasing convex function, taking the logarithm of equation (23) will convert the original problem of maximizing into a problem of minimizing. Therefore, model (23) is equivalent to:

[0130]

[0131] Equation (25) is the electricity trading payment negotiation problem.

[0132] From the above derivation, in the upper and lower grid alliance benefit maximization model, the electricity trading amount U d1,g and U d2,g are offset in the superposition process. Therefore, if the overall model of the upper and lower grid alliance benefit maximization is directly established, the electricity trading amount of each subject in the upper and lower grid will be offset due to superposition, resulting in the inability to determine U d1,g and U d2,g Therefore, it is necessary to introduce the Nash negotiation model. By solving sub-problem 2, the electricity trading price can be determined, and then the electricity trading amount can be determined.

[0133] Step S320: Introduce the Lagrange multiplier and the penalty factor to obtain the augmented Lagrange function of the upper and lower grid alliance benefit maximization model objective function.

[0134] In order to protect the privacy of each subject of the upper and lower power grids during the negotiation, the alternating direction multiplier method (ADMM) is used to solve the upper and lower power grid alliance benefit maximization sub-problem and the power payment transaction negotiation sub-problem in a distributed manner. The ADMM algorithm has the advantages of good convergence characteristics and strong robustness, and can protect the privacy of each subject of the upper and lower power grids during the solution.

[0135] Based on the ADMM-based upper and lower power grid alliance benefit maximization sub-model solution, first, for the upper and lower power grid cooperative operation Nash negotiation model, i.e. formula (14), let

[0136]

[0137] In the formula, and respectively represent the expected amount of electricity sold by the upper power grid to the lower power grids 1 and 2 at time t; respectively represent the expected amount of electricity purchased by the lower power grids 1 and 2 from the upper power grid at time t; when , it indicates that the upper power grid, the lower power grid 1, and the lower power grid 2 have reached a consensus on electricity trading.

[0138] Take the opposite number of the upper and lower power grid alliance benefit maximization model formula target function, i.e. formula (17), and convert it into a minimization problem. Introduce the Lagrange multiplier and the penalty factor ρ d1 , ρ d2 ; the augmented Lagrange function of the upper and lower power grid alliance benefit maximization model target function can be obtained, which is represented by formula (27):

[0139]

[0140] Step S330: According to the ADMM algorithm principle, the augmented Lagrange function of the alliance benefit maximization model target function is decomposed to obtain the distributed optimization operation model of the lower power grid and the upper power grid. After iterative calculation, the optimal expected transaction electricity of the upper power grid and each lower power grid is obtained.

[0141] Among them, the distributed optimization operation model of the lower power grid 1 and the lower power grid 2 is:

[0142]

[0143] The distributed optimization operation model of the upper power grid is:

[0144]

[0145] The solution steps of the upper and lower power grid distributed optimization operation model are as follows:

[0146] Step 11): Set the maximum iteration number k max = 100 and the convergence precision ξ = 10​-5 and a penalty factor p d1 = p d2 = 10 -4 ; initialize iteration number k = 0, initial purchase quantity of subordinate power grids 1, 2 Lagrange multiplier

[0147] Step 12): for the upper power grid, receive expected purchase quantity data from the subordinate power grids 1, 2 and Solve the model model (29), and obtain expected sale quantity and

[0148] Step 13): for the subordinate power grid 1, receive expected sale quantity from the upper power grid Solve the model (28) (i = 1) to obtain expected purchase quantity

[0149] Step 14): for the subordinate power grid 2, receive expected sale quantity from the upper power grid Solve the model (28) (i = 2) to obtain expected purchase quantity

[0150] Step 15): update the Lagrange multiplier according to formula (30):

[0151]

[0152] Step 16): update the iteration number: k = k + 1;

[0153] Step 17): judge the convergence of the algorithm, if the iteration termination condition of formula (31) is met:

[0154]

[0155] the iteration is terminated, otherwise return to step 2 to repeat the calculation until the convergence condition is met or the maximum iteration number is reached.

[0156] Step S340: substitute the optimal expected transaction quantity of the upper power grid and each subordinate power grid into the electricity transaction payment negotiation model, introduce the Lagrange multiplier and the penalty factor, and obtain the augmented Lagrange function of the objective function of the electricity transaction payment negotiation model.

[0157] By solving the upper and lower power grid alliance benefit maximization sub-model 1, the optimal expected transaction quantity between the upper power grid and the subordinate power grids 1 and 2 can be obtained simultaneously: and According to formula (2), the electricity transaction amount U d1,g and U d2,g can be represented as:

[0158]

[0159] Substitute equation (32) into the electricity payment negotiation sub-model-equation (18):

[0160]

[0161] According to the principle of ADMM algorithm, the auxiliary variables and are used to decouple the electricity price between the upper grid and the lower grid 1 and the upper grid and the lower grid 2:

[0162]

[0163] In the formula: and are the electricity prices expected by the lower grid 1 and 2 to purchase electricity from the upper grid, respectively; is the electricity price expected by the upper grid to sell electricity to the lower grid 1 and 2. The augmented Lagrangian function of the objective function of model (33) is:

[0164]

[0165] In the formula: is the Lagrange multiplier; ψ d1 , ψ d2 is the penalty factor.

[0166] Step S350: According to the principle of ADMM algorithm, the augmented Lagrangian function of the objective function of the electricity transaction payment negotiation model-equation (35) is decomposed to obtain the electricity transaction price distributed optimization model of each lower grid and the upper grid, and the optimal transaction electricity price of the upper grid and each lower grid is obtained after iterative calculation.

[0167] Among them, equation (35) is decomposed, and the electricity transaction price distributed optimization model of each subject of the upper and lower grid can be obtained respectively:

[0168] (1) Electricity transaction price distributed optimization model of lower grid 1 and 2

[0169]

[0170] (2) Electricity transaction price distributed optimization model of upper grid

[0171]

[0172] The solution steps of the upper and lower grid electricity transaction payment negotiation sub-model are as follows:

[0173] Step 21): Set the maximum iteration number k max = 100, and the convergence precision ξ = 10-5 and a penalty factor ψ d1 = ψ d2 = 10 -4 ; initialize iteration number k = 0, initial purchase electricity price of subordinate power grid 1, 2 Lagrange multiplier

[0174] Step 22): for the upper power grid, receive expected purchase electricity price data from the subordinate power grid 1, 2 and Solve the model (37) to obtain the expected selling electricity price and

[0175] Step 23): for the subordinate power grid 1, receive the expected selling electricity price from the upper power grid Solve the model (36) (i = 1) to obtain the expected purchase electricity price

[0176] Step 24): for the subordinate power grid 2, receive the expected selling electricity price from the upper power grid Solve the model (36) (i = 2) to obtain the expected purchase electricity price

[0177] Step 25): update the Lagrange multiplier according to formula (38):

[0178]

[0179] Step 26): update the iteration number: k = k + 1;

[0180] Step 27): judge the convergence of the algorithm, if the iteration termination condition of formula (39) is met:

[0181]

[0182] the iteration is terminated, otherwise return to step 2 to repeat the calculation until the convergence condition is met or the maximum iteration number is reached.

[0183] The method disclosed in the embodiment has been verified to be effective by building a reasonable simulation example.

[0184] Embodiment 2

[0185] Based on the same inventive concept, the application further discloses a scheduling system comprising:

[0186] The first modeling unit is used for separately establishing upper and lower power grid optimization scheduling models according to power grid loss, load side demand response and light abandonment penalty.

[0187] A second modeling unit is configured to combine the Nash negotiation theory with the superior-inferior power grid optimization scheduling model to form a superior-inferior power grid collaborative optimization Nash negotiation model.

[0188] A solving unit is configured to solve the superior-inferior power grid collaborative optimization Nash negotiation model and optimize scheduling according to a solving result.

[0189] Embodiment 3

[0190] Based on the same inventive concept, the application further discloses an intelligent terminal comprising a memory and a processor, wherein the memory stores a computer program capable of being loaded and executed by the processor to perform any one of the superior-inferior power grid collaborative optimization scheduling methods based on the alternating direction method of multipliers in Embodiment 1.

[0191] Embodiment 4

[0192] Based on the same inventive concept, the application further discloses a computer readable storage medium, wherein the medium stores a computer program capable of being loaded and executed by the processor to perform any one of the superior-inferior power grid collaborative optimization scheduling methods based on the alternating direction method of multipliers in Embodiment 1.

[0193] The above embodiments are only used to specifically describe the technical solutions of the application, and the above descriptions are only used to help understand the method of the application and its core idea, and should not be understood as a limitation of the application. Any changes or replacements that can be easily thought of by those skilled in the art within the technical range disclosed by the application should be covered within the protection scope of the application.

Claims

1. A method for coordinated optimal dispatching of upper and lower power grids based on an alternating direction method of multipliers, characterized in that, The application relates to a method for collaborative optimization and dispatching of upper and lower level power grids based on an alternating direction multiplier method. The method comprises the following steps: The Nash negotiation theory is combined with the upper and lower level power grid optimization dispatching model to form an upper and lower level power grid collaborative optimization Nash negotiation model; The upper and lower level power grid collaborative optimization Nash negotiation model is solved, and the upper and lower level power grids are optimized and dispatched according to the solving result; The upper and lower level power grid collaborative optimization Nash negotiation model is solved, and the upper and lower level power grids are optimized and dispatched according to the solving result; The upper and lower level power grid collaborative optimization Nash negotiation model is solved, and the upper and lower level power grids are optimized and dispatched according to the solving result; The upper and lower level power grid collaborative optimization Nash negotiation model is solved, and the upper and lower level power grids are optimized and dispatched according to the solving result; The upper and lower level power grid collaborative optimization Nash negotiation model is solved, and the upper and lower level power grids are optimized and dispatched according to the solving result; The upper and lower level power grid collaborative optimization Nash negotiation model is solved, and the upper and lower level power grids are optimized and dispatched according to the solving result; The upper and lower level power grid collaborative optimization Nash negotiation model is solved, and the upper and lower level power grids are optimized and dispatched according to the solving result; 2. The method of claim 1, wherein the method is characterized by, The upper and lower level power grid collaborative optimization Nash negotiation model is solved, and the upper and lower level power grids are optimized and dispatched according to the solving result. The upper and lower level power grid optimization dispatching model comprises an upper level power grid subject benefit maximization operation model and a lower level power grid subject benefit maximization operation model; maxU up = U d1,g + U d2,g - C up,m - C up,loss In the formula, U d1,g and U d2,g are the electricity sales revenues of the lower-level power grids 1 and 2, respectively; C up,m is the operation cost of the upper-level power grid; and C up,loss is the network loss cost of the upper-level power grid. The target function of the upper level power grid subject benefit maximization operation model is: maxU di = -U di,g - C di,m - C di,loss (i = 1, 2) In the formula, i=1, 2 respectively represent the lower power grid 1 and the lower power grid 2; U di,g , C di,m , C di,loss are respectively the purchase cost of the lower power grid i to the upper power grid, the operation cost of the lower power grid and the network loss cost of the lower power grid.

3. The method of claim 1, wherein the method is characterized by: The target function of the lower level power grid subject benefit maximization operation model is: In the formula, respectively, the optimal operation benefits of the non-main body of the upper and lower power grids when not cooperating, i.e. the Nash negotiation breaking point, are constants, up is the upper power grid, and d1 and d2 are the lower power grids.

4. The method of claim 1, wherein the method is characterized in that: The target function of the upper and lower level power grid collaborative optimization Nash negotiation model is: maxω up +ω d1 +ω d2 The target function of the upper and lower level alliance benefit maximization model is: In the formula, U is the optimal solution of the Nash negotiation model for the coordination optimization of the upper and lower power grids d1,g and U d2,g are the electricity selling benefits of the lower power grid 1 and the lower power grid 2 respectively, are the optimal operation benefits of the non-main bodies of the upper and lower power grids when not cooperating.

5. The method of claim 1, wherein, The target function of the electric energy transaction payment negotiation model is: is a Lagrange multiplier, p d1 , p d2 is a penalty factor, respectively represent the amount of electricity that the upper-level power grid expects to sell to the lower-level power grids 1, 2 at time t, respectively represent the amount of electricity that the lower-level power grids 1, 2 expect to purchase from the upper-level power grid at time t; The augmented Lagrange function of the alliance benefit maximization model target function is: S is the probability of the scenario, k di N is the total number of photovoltaic output scenarios of the subordinate grid i, k gt , is the unit power cost coefficient of the gas turbine, and is the light abandonment penalty coefficient, respectively, is the power generation of the gas turbine of the subordinate grid i, is the predicted power generation of the photovoltaic, is the actual output of the photovoltaic, is the amount of translatable load, and is the amount of reducible load of the subordinate grid i at time t, k dr1 , k dr2 is the demand response coefficient, and is the compensation coefficient of the translatable load and the compensation coefficient of the reducible load, respectively, is the active power loss of the subordinate grid i at time t, is the loss penalty coefficient, and ρ di is the penalty factor, represents the expected amount of electricity sold by the upper-level grid to the subordinate grid i at time t, represents the expected amount of electricity purchased by the subordinate grid i from the upper-level grid at time t, is the Lagrange multiplier; The target function of the lower level power grid distributed optimization operation model is: τ up S up S up S a j , b j , c j a S S S 6. The method of claim 1, wherein, The target function of the upper level power grid distributed optimization operation model is: In the formula, is the electricity price expected by the upper grid to sell electricity to the lower grid 1, 2, is the optimal expected trading electricity between the upper grid and the lower grid 1, the upper grid and the lower grid 2, respectively, the optimal operation benefit of the upper and lower grid non-main body when not cooperating, is the Lagrange multiplier, and Ψ d1 , Ψ d2 is the penalty factor, is the optimal solution of the upper and lower grid cooperative optimization Nash negotiation model; The augmented Lagrange function of the electric energy transaction payment negotiation model target function is: The target function of the lower level power grid electric energy transaction price distributed optimization model is:

7. An upper-and-lower-level power grid collaborative optimization scheduling system based on an alternating direction multiplier method, applicable to any one of the upper-and-lower-level power grid collaborative optimization scheduling methods based on the alternating direction multiplier method in claims 1-6, characterized in that, The target function of the upper level power grid electric energy transaction price distributed optimization model is: The method comprises the following steps: The first modeling unit is used for separately establishing the upper and lower level power grid optimization dispatching model according to power grid loss, load side demand response and light abandonment penalty; The second modeling unit is used for combining the Nash negotiation theory with the upper and lower level power grid optimization dispatching model to form the upper and lower level power grid collaborative optimization Nash negotiation model; 8. A smart terminal, characterized by The solving unit is used for solving the upper and lower level power grid collaborative optimization Nash negotiation model, and optimizing and dispatching the upper and lower level power grids according to the solving result. The application further discloses a computer program which can be loaded and executed by the processor to realize the method.

9. A computer-readable storage medium, characterized in that, The computer program capable of being loaded and executed by the processor and performing the super-subordinate power grid collaborative optimization scheduling method based on the alternating direction multiplier method according to any one of claims 1 to 6 is stored.

Images