A distributed transaction volume adjustment method for distribution network based on ADMM algorithm

Abstract

The present invention proposes a method for adjusting the distributed transaction volume of multiple subjects in a distribution network based on an ADMM algorithm, including: obtaining the transaction volume request data of the producer and consumer, generating a transaction volume matrix according to the transaction volume request data, inputting the transaction volume matrix into the transaction model to generate the consistency problem in the ADMM algorithm; decomposing the consistency problem into sub-problems of a single seller and a single buyer, obtaining the seller's auxiliary variables and the buyer's auxiliary variables by performing ADMM iteration on the sub-problems, and updating the transaction volume matrix by calculating the average values ​​of the seller's auxiliary variables and the buyer's auxiliary variables; when the difference between the seller's auxiliary variables, the buyer's auxiliary variables and the updated transaction volume matrix is ​​less than a preset limit, stopping the iteration, and adjusting the power consumption and power injection of the producer and consumer according to the updated transaction volume matrix. The present invention takes into account the situation of collecting network access fees, can realize the fully distributed transaction of power among multiple subjects in the distribution network, and effectively protects the privacy of the producer and consumer.

Description

Technical Field

[0001] The present invention belongs to the field of distributed interaction of distribution networks, and in particular relates to a method for adjusting multi-agent distributed transaction volume of distribution networks based on an ADMM algorithm. Background Art

[0002] With the massive access of distributed energy sources on the user side, such as micro gas turbines, combined heat and power units, new energy generation, and energy storage systems, the traditional distribution network is undergoing a transformation from a passive distribution network to an active distribution network. The active distribution network has advanced supervision and control systems, data acquisition systems, and various communication technologies. With the support of these intelligent technologies, the terminal users of the distribution network can effectively utilize local distributed energy and realize the transformation from consumers in the traditional sense to prosumers with the ability to produce and manage electricity. For a long time, in order to encourage prosumers to actively participate in energy exchange, the feed-in-tariff policy (FiT) has been widely used. Prosumers can sell excess electricity to power grid companies at a price higher than the feed-in tariff of conventional power sources to make profits. However, the FiT policy has the disadvantages that a large number of prosumers will access the distribution network to participate in the competition in the electricity market, which will bring long decision-making time and low clearing efficiency to traditional centralized transactions. Therefore, it is necessary to explore a new sustainable development model for the consumption of distributed energy in a market environment.

[0003] In recent years, a peer-to-peer (P2P) trading model has emerged, where distributed power generation entities and power users, as producers and consumers, can directly trade electricity. Currently, the market operator (distribution system operator, DSO) integrates the trading needs of various producers and consumers in the trading market and matches and dispatches the trading needs of various producers and consumers. However, as the number of producers and consumers participating in the P2P power trading market continues to increase, the existing distributed mechanism has the disadvantage of being difficult to protect the private information of producers and consumers. Summary of the invention

[0004] In order to solve the shortcomings and deficiencies in the prior art, the present invention proposes a distribution network multi-agent distributed transaction volume adjustment method based on the ADMM algorithm, comprising:

[0005] Obtain the transaction volume request data of the producer and consumer, generate the transaction volume matrix based on the transaction volume request data, and input the transaction volume matrix into the transaction model to generate the consistency problem in the ADMM algorithm;

[0006] The consistency problem is decomposed into sub-problems of a single seller and a single buyer. The seller's auxiliary variables and the buyer's auxiliary variables are obtained by performing ADMM iterations on the sub-problems. The transaction volume matrix is ​​updated by calculating the average values ​​of the seller's auxiliary variables and the buyer's auxiliary variables.

[0007] When the difference between the seller's auxiliary variables, the buyer's auxiliary variables and the updated transaction volume matrix is ​​less than the preset limit, the iteration is stopped, and the electricity consumption and electricity injection of the prosumer are adjusted according to the updated transaction volume matrix.

[0008] Optionally, the step of obtaining the transaction volume request data of the producer and consumer, generating a transaction volume matrix according to the transaction volume request data, and inputting the transaction volume matrix into the constructed transaction model to generate consistency problems in the ADMM algorithm includes:

[0009] Obtaining transaction volume request data of prosumers, marking prosumers as buyers or sellers according to their identity types, and forming a transaction volume matrix according to the transaction volume request data based on the corresponding relationship between buyers and sellers;

[0010] The transaction volume matrix is ​​input into the transaction model F, which is:

[0011]

[0012] Among them, C b (p b ) represents the utility function of the buyer set, p b represents the buyer’s electricity consumption, C s (p s ) represents the cost function of the seller set, p s Indicates the amount of electric energy injected by the seller, 1 s represents a unit column vector of dimension s, Representation 1 s The transpose of b represents the unit column vector of dimension b, П represents the network fee matrix, and E represents p b and p s The transaction volume matrix of the buyers and sellers is composed of П°E, where П°E represents the Hadamard product of П and E, B represents the seller set, and S represents the buyer set;

[0013] Set the constraints of transaction model F as follows:

[0014]

[0015] in, p s , They represent the upper and lower limits of the amount of electricity injected by the prosumer, p b , They represent the upper and lower limits of the electricity consumption of prosumers respectively;

[0016] The transaction model F after inputting the transaction volume matrix is ​​inverted to obtain the consistency problem in the ADMM algorithm.

[0017] Optionally, the dimension of the network fee matrix is ​​M×N, M is the total number of sellers, N is the total number of buyers, and the value ranges of M and N are both positive integers.

[0018] Optionally, the cost function C of the seller set s (p s ) is expressed as:

[0019] C s (p s )=(c s +diag(d s )×p s ) T ×p s ;

[0020] Among them, p s represents the amount of electric energy injected by the seller, c s ,d s These are the preset coefficients for the linear term and the quadratic term respectively.

[0021] Optionally, the utility function C of the buyer set b (p b ) is expressed as:

[0022] C b (p b )=(c b +diag(d b )×p b ) T ×p b ;

[0023] Among them, p b represents the buyer's electricity consumption, c b ,d b These are the preset coefficients for the linear term and the quadratic term respectively.

[0024] Optionally, decomposing the consistency problem into sub-problems of a single seller and a single buyer includes:

[0025] Decompose the first set of sub-problems F1 of the seller set from the consistency problem. The expression formula of F1 is:

[0026]

[0027] in, is the seller's auxiliary variable to be solved, and Λ represents the transaction price matrix agreed upon by the buyer and seller, ρ represents the preset penalty parameter, Denotes Λ and The Hadamard product of express and Hadamard product of;

[0028] Based on the first set of subproblems F1, we further decompose the subproblems of seller i. for:

[0029]

[0030] Among them, C s,i (p s,i ) represents the cost function of seller i, p s,i represents the amount of electric energy injected by seller i, express The i-th row of represents the transpose of the i-th row of П, e i,* represents the i-th row of E, Λ i,* represents the i-th row of Λ.

[0031] Optionally, decomposing the consistency problem into sub-problems of a single seller and a single buyer includes:

[0032] Decompose the second set of sub-problems F2 of the buyer set from the consistency problem. The expression formula of F2 is:

[0033]

[0034] in, is the buyer's auxiliary variable to be solved, and Φ represents the transaction price matrix agreed upon by the buyer and seller, ρ represents the preset penalty parameter, represents Φ and The Hadamard product of express and Hadamard product of;

[0035] Based on the second set of subproblems F2, we further decompose the subproblems of buyer j. for:

[0036]

[0037] Among them, C b,j (p b,j ) represents the utility function of buyer j, p b,j represents the electricity consumption of buyer j, express The jth column of *,j represents the transpose of the jth column of П, e *,j represents the j-th column of E, represents the transpose of the j-th column of Φ.

[0038] Optionally, the updating of the transaction volume matrix by calculating the average value of the seller's auxiliary variable and the buyer's auxiliary variable includes:

[0039] Sum each row of the seller auxiliary variable and the buyer auxiliary variable respectively, and update each row of the transaction volume matrix as follows:

[0040]

[0041] Among them, e i,* (k+1) represents the i-th row of the volume matrix in the k+1th iteration, represents the i-th row of the seller auxiliary variable in the k-th iteration, represents the i-th row of the buyer auxiliary variable in the k-th iteration.

[0042] Optionally, the updating of the transaction volume matrix by calculating the average value of the seller's auxiliary variable and the buyer's auxiliary variable includes:

[0043] Sum each column of the seller auxiliary variable and the buyer auxiliary variable respectively, and update each column of the trading volume matrix as follows:

[0044]

[0045] Among them, e *,j (k+1) represents the jth column of the volume matrix in the k+1th iteration, represents the j-th column of the seller auxiliary variable in the k-th iteration, represents the j-th column of the buyer auxiliary variable in the k-th iteration.

[0046] The beneficial effects brought by the technical solution provided by the present invention are:

[0047] Aiming at the real-time power transaction between producers and consumers on the distribution side, an adjustment and optimization method for distributed transactions in the P2P market based on the ADMM algorithm is proposed. At the same time, considering the situation of collecting transmission fees, the relevant variables of transmission fees are introduced into the transaction model. This method can realize the fully distributed transaction of power among multiple entities in the distribution network. Only the transaction volume request data of producers and consumers is needed to optimize the P2P transaction volume. There is no need for producers and consumers to provide more private information, which effectively protects the privacy of producers and consumers. At the same time, it also improves the efficiency of optimization calculation. It is suitable for distributed transaction volume adjustment for multiple distribution network entities. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] In order to more clearly illustrate the technical solution of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0049] Figure 1 A schematic diagram of a flow chart of a method for adjusting multi-agent distributed transaction volume in a distribution network based on an ADMM algorithm proposed in one embodiment of the present invention;

[0050] Figure 2 Schematic diagram of the P2P electricity trading model. DETAILED DESCRIPTION

[0051] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0052] The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the numbers used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in sequences other than those illustrated or described herein.

[0053] It should be understood that in various embodiments of the present invention, the size of the sequence number of each process does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0054] It should be understood that in the present invention, "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or apparatus comprising a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to these processes, methods, products or apparatuses.

[0055] It should be understood that in the present invention, "plurality" refers to two or more than two. "And / or" is merely a description of the association relationship of associated objects, indicating that three relationships may exist. For example, and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. The character " / " generally indicates that the objects associated before and after are in an "or" relationship. "Contains A, B and C", "Contains A, B, C" means that A, B, and C are all included, "Contains A, B or C" means that one of A, B, and C is included, and "Contains A, B and / or C" means that any one, any two, or any three of A, B, and C are included.

[0056] It should be understood that in the present invention, "B corresponding to A", "B corresponding to A", "A corresponds to B" or "B corresponds to A" means that B is associated with A and B can be determined based on A. Determining B based on A does not mean determining B based only on A, but B can also be determined based on A and / or other information. A and B match when the similarity between A and B is greater than or equal to a preset threshold.

[0057] Depending on the context, "if" as used herein may be interpreted as "when" or "when" or "in response to determining" or "in response to detecting."

[0058] The technical solution of the present invention is described in detail with specific embodiments below. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described in detail in some embodiments.

[0059] Embodiment 1

[0060] like Figure 1 As shown, this embodiment proposes a method for adjusting multi-agent distributed transaction volume in a distribution network based on an ADMM algorithm, including:

[0061] S1: Obtain the transaction volume request data of the producer and consumer, generate a transaction volume matrix based on the transaction volume request data, and input the transaction volume matrix into the transaction model to generate the consistency problem in the ADMM algorithm;

[0062] S2: Decompose the consistency problem into sub-problems of a single seller and a single buyer, obtain the seller's auxiliary variables and the buyer's auxiliary variables by performing ADMM iterations on the sub-problems, and update the transaction volume matrix by calculating the average values ​​of the seller's auxiliary variables and the buyer's auxiliary variables;

[0063] S3: When the difference between the seller's auxiliary variable, the buyer's auxiliary variable and the updated transaction volume matrix is ​​less than the preset limit, the iteration is stopped, and the electricity consumption and electricity injection of the prosumer are adjusted according to the updated transaction volume matrix.

[0064] The peer-to-peer (P2P) trading model supported by the interactive energy mechanism is an energy trading model that has emerged in recent years. It has many advantages, such as promoting distributed energy consumption, protecting the privacy and security of producers and consumers, and ensuring economic benefits. At the same time, a large number of producers and consumers accessing the distribution network to participate in the competition in the power market will bring problems such as long decision-making time and low clearing efficiency to traditional centralized transactions. P2P transactions can realize distributed computing and avoid the problem of weak centralized computing of producers and consumers. Therefore, P2P technology is expected to become the next generation of smart grid energy management technology. This embodiment uses the alternating direction method of multipliers (ADMM) algorithm to adjust distributed transactions. Compared with the current P2P trading mechanism, it takes into account the impact of network fees on each transaction, and does not require the intervention of a third party. It can fully protect the privacy information of producers and consumers participating in the P2P trading market, and at the same time solve the transaction volume and transaction price, which is conducive to the timely adjustment of P2P power trading strategies by producers and consumers from multiple entities.

[0065] like Figure 2 As shown, the left side is the seller set S: = {1,2,…,s}, and the right side is the buyer set B: = {1,2,…,b}, p s,i represents the total electric energy invested by seller i in the transaction, p b,j represents the total electricity purchased by buyer j, λ i,j represents the transaction price reached between seller i and buyer j. All transaction prices can be represented by matrices Λ or Φ, π i,j The grid-connection fee for the unit electricity transaction between seller i and seller j is expressed as the grid-connection fee that needs to be paid. The grid-connection fee can play a role in differentiating the prosumers, that is, distinguishing a certain prosumer from the rest. All grid-connection fees can be expressed as the grid-connection fee matrix П∈{π i,j |i∈S,j∈B}. All trading volumes can be represented by the trading volume matrix E∈{e i,j |i∈S,j∈B} means, e i,j Represents the amount of electricity traded between seller i and buyer j.

[0066] In a specific trading period, the identities of each producer and consumer in the market have been determined as sellers or buyers, which are marked as the seller set and the buyer set respectively. This embodiment first defines that the seller cost function and the buyer utility function can be expressed by the following quadratic function:

[0067] C x (p x )=(c x +diag(d x )×p x ) T ×p x , x∈{S,B};

[0068] In the formula, p x ≥0, x∈S represents the amount of electric energy injected by the seller, and x∈B represents the amount of electric energy consumed by the buyer. The prosumers are marked as buyers or sellers according to their identity type, forming a buyer set B and a seller set S. Therefore, in this embodiment, the seller's cost function C s (p s ) is expressed as:

[0069] C s (p s )=(c s +diag(d s )×p s ) T ×p s ;

[0070] Among them, p s represents the amount of electric energy injected by the seller, c s ,d s These are the preset coefficients for the linear term and the quadratic term respectively.

[0071] The buyer's utility function C b (p b ) is expressed as:

[0072] C b (p b )=(c b +diag(d b )×p b ) T ×p b ;

[0073] Among them, p b represents the buyer's electricity consumption, c b ,d b These are the preset coefficients for the linear term and the quadratic term respectively.

[0074] Based on the above cost function and utility function, in order to maximize the utility in the transaction, the transaction model F between the buyer and the seller is constructed as follows:

[0075]

[0076] Among them, C b (p b ) represents the utility function of the buyer set, p b represents the electricity consumption of the buyer set, C s (p s ) represents the cost function of the seller set, p s represents the amount of electric energy injected by the seller set, 1 srepresents a unit column vector of dimension s, Representation 1 s The transpose of b represents a unit column vector with dimension b, П represents the transmission fee matrix, each element of which represents the transmission fee required for unit power transactions between different buyers and sellers, which is set by the DSO and can be positive or negative. E represents the transaction volume matrix of buyers and sellers, which is a global variable, and П°E represents the Hadamard product of П and E.

[0077] Set the constraints of transaction model F as follows:

[0078]

[0079] Among them, E1 b =p s and E T 1 s =p b The power balance constraints for buyers and sellers are specified separately, that is, the electric energy injected into the network by each seller is equal to the sum of the electric energy traded with all buyers, and the same applies to buyers. p s , They represent the upper and lower limits of the amount of electricity injected by the prosumer, p b , They represent the upper and lower limits of the electricity consumption of the prosumer, respectively. E≥0 stipulates that only the seller is allowed to sell electricity to the buyer during the trading period.

[0080] In the real-time market, prosumers will bid to the DSO of the distribution network according to their preferences and operating constraints. It should be noted that due to the existence of some cross-period coupling constraints, the parameters of prosumers may change in different trading periods. In this embodiment, only the single-period situation is considered. Secondly, the change of parameters does not affect the convexity of the objective function of the trading model constructed below, nor does it affect the model solution.

[0081] The ADMM algorithm is applicable to large-scale distributed optimization problems. This method decomposes the global problem into multiple local sub-problems, and obtains the solution to the original global problem by unifying and coordinating the solutions of all sub-problems. Based on the basic solution process of the ADMM algorithm, in this embodiment, the consistency problem is decomposed into sub-problems of a single seller and a single buyer, which specifically include:

[0082] Decompose the first set of sub-problems F1 of the seller set from the consistency problem. The expression formula of F1 is:

[0083]

[0084] in, Auxiliary variables for sellers to set, and Λ represents the transaction price matrix agreed upon by the buyer and seller, also p s The matrix of dual variables, ρ represents the preset penalty parameter, Denotes Λ and The Hadamard product of express and Since the optimization purpose of the ADMM algorithm is usually to minimize the objective function, in this embodiment, the transaction model F is appropriately negatively deformed to adapt to the ADMM algorithm.

[0085] Thanks to the separable structure of the model constructed in this embodiment, in the first set of sub-problems, the objective function and constraints can be further decomposed into the objectives and constraints of a single seller, so that the centralized solution of the original model can be transformed into the parallel solution of M sub-problems of a single seller. Based on the first set of sub-problems F1, the sub-problems of seller i are further decomposed into for:

[0086]

[0087] Among them, C s,i (p s,i ) represents the cost function of seller i, p s,i represents the amount of electric energy injected by seller i, express The i-th row of represents the transpose of the i-th row of П, e i,* represents the i-th row of E, Λ i,* represents the i-th row of Λ.

[0088] Similar to the seller, the buyer specifically includes: decomposing the second set of sub-problems F2 of the buyer set from the consistency problem. The expression formula of F2 is:

[0089]

[0090] in, Auxiliary variables for buyers to set, and Φ represents the transaction price matrix reached by the buyer and seller, also p b The matrix of dual variables, ρ represents the preset penalty parameter, represents Φ and The Hadamard product of express and Hadamard product of;

[0091] Based on the second set of subproblems F2, we further decompose the subproblems of buyer j. for:

[0092]

[0093] Among them, C b,j (p b,j ) represents the utility function of buyer j, p b,j represents the electricity consumption of buyer j, express The jth column of *,j represents the transpose of the jth column of П, e *,j represents the j-th column of E, represents the transpose of the j-th column of Φ.

[0094] Then the ADMM iterative solution cycle process is carried out, specifically:

[0095] Initialize, set the network fee matrix and penalty parameters, and start the ADMM iteration cycle;

[0096] Based on the ADMM algorithm, the seller's auxiliary variables and buyer's auxiliary variables in the sub-problems of a single seller and a single buyer are solved respectively, specifically:

[0097] For each seller i, its power injection amount p at the kth iteration s,i (k) and seller auxiliary variables The i-th row Both:

[0098]

[0099]

[0100] For each buyer j, the energy consumption p at the kth iteration b,j (k) and buyer-aided variables The i-th row Both:

[0101]

[0102]

[0103] After each iteration, seller i will Sent to all buyers, buyer j will It is sent to all sellers, and then the transaction volume matrix E as a global variable is updated by calculating the average value of the seller's auxiliary variables and the buyer's auxiliary variables to obtain E(k). Specifically, two updating methods can be used: row-based and column-based.

[0104] When updating in rows, sum each row of the seller auxiliary variable and the buyer auxiliary variable, and update each row of the transaction volume matrix as follows:

[0105]

[0106] Among them, e i,* (k+1) represents the i-th row of the volume matrix in the k+1th iteration, represents the i-th row of the seller auxiliary variable in the k-th iteration, represents the i-th row of the buyer auxiliary variable in the k-th iteration.

[0107] When updating in columns, sum each column of the seller auxiliary variable and the buyer auxiliary variable, and update each column of the transaction volume matrix as follows:

[0108]

[0109] Among them, e *,j (k+1) represents the jth column of the volume matrix in the k+1th iteration, represents the j-th column of the seller auxiliary variable in the k-th iteration, represents the j-th column of the buyer auxiliary variable in the k-th iteration.

[0110] If the difference between the trading volume matrix E and the seller's auxiliary variables and the buyer's auxiliary variables is less than the preset limit, the iteration is stopped and the trading volume matrix E at this time is used as the trading volume optimization result. Otherwise, the updated trading volume matrix E will be substituted into the sub-problem to continue the k+1th iteration.

[0111] In this embodiment, the difference between the transaction volume matrix E and the seller auxiliary variable and the buyer auxiliary variable is expressed as the original residual r k and the dual residual s k , the specific calculation is:

[0112]

[0113] s k =(E(k)-E(k-1))°(E(k)-E(k-1));

[0114] represents the seller auxiliary variable solved in the kth iteration, represents the buyer auxiliary variable solved at the kth iteration, E(k) represents the updated transaction volume matrix at the kth iteration, and E(k-1) represents the updated transaction volume matrix at the k-1th iteration. In this embodiment, when the original residual and the dual residual are reduced to a certain preset limit, it can be considered that the ADMM algorithm has converged and the iteration can be stopped.

[0115] The serial numbers in the above embodiments are for description only and do not represent the sequence of the components during assembly or use.

[0116] The above descriptions are merely embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for adjusting multi-agent distributed transaction volume in distribution network based on ADMM algorithm, characterized in that: The distribution network multi-agent distributed interaction method comprises: Obtain the transaction volume request data of the producer and consumer, generate the transaction volume matrix based on the transaction volume request data, and input the transaction volume matrix into the transaction model to generate the consistency problem in the ADMM algorithm; The consistency problem is decomposed into sub-problems of a single seller and a single buyer. The seller's auxiliary variables and the buyer's auxiliary variables are obtained by performing ADMM iterations on the sub-problems. The transaction volume matrix is ​​updated by calculating the average values ​​of the seller's auxiliary variables and the buyer's auxiliary variables. When the difference between the seller's auxiliary variable, the buyer's auxiliary variable and the updated transaction volume matrix is ​​less than the preset limit, the iteration is stopped, and the power consumption and power injection of the prosumer are adjusted according to the updated transaction volume matrix; The updating of the transaction volume matrix by calculating the average value of the seller's auxiliary variable and the buyer's auxiliary variable includes: Sum each row of the seller auxiliary variable and the buyer auxiliary variable respectively, and update each row of the transaction volume matrix as follows: ； in, represents the i-th row of the volume matrix in the k+1th iteration, represents the i-th row of the seller auxiliary variable in the k-th iteration, represents the i-th row of the buyer auxiliary variable in the k-th iteration; The method of updating the transaction volume matrix by calculating the average value of the seller auxiliary variable and the buyer auxiliary variable also includes: Sum each column of the seller auxiliary variable and the buyer auxiliary variable respectively, and update each column of the trading volume matrix as follows: ； in, represents the jth column of the volume matrix in the k+1th iteration, represents the j-th column of the seller auxiliary variable in the k-th iteration, represents the jth column of the buyer's auxiliary variable in the kth iteration; The process of obtaining the transaction volume request data of the producer and consumer, generating a transaction volume matrix according to the transaction volume request data, and inputting the transaction volume matrix into the constructed transaction model to generate consistency problems in the ADMM algorithm includes: Obtaining transaction volume request data of prosumers, marking prosumers as buyers or sellers according to their identity types, and forming a transaction volume matrix according to the transaction volume request data based on the corresponding relationship between buyers and sellers; The transaction volume matrix is ​​input into the transaction model F, which is: ； in, represents the utility function of the buyer set, represents the buyer’s electricity consumption, represents the cost function for the set of sellers, Indicates the amount of electric energy injected by the seller, represents a unit column vector of dimension s, express The transpose of represents a unit column vector of dimension b, represents the network fee matrix, express and The transaction volume matrix of buyers and sellers is composed of express and The Hadamard product of represents the set of sellers, and S represents the set of buyers; Set the constraints of transaction model F as follows: ； in, , They represent the upper and lower limits of the amount of electricity injected by the prosumer, , They represent the upper and lower limits of the electricity consumption of prosumers respectively; The transaction model F after inputting the transaction volume matrix is ​​inverted to obtain the consistency problem in the ADMM algorithm.

2. The method for adjusting the multi-agent distributed transaction volume of a distribution network based on the ADMM algorithm according to claim 1 is characterized in that: The dimension of the network fee matrix is , is the total number of sellers, is the total number of buyers, and the value ranges of M and N are both positive integers.

3. The method for adjusting multi-agent distributed transaction volume in distribution network based on ADMM algorithm according to claim 1 is characterized in that: The cost function of the seller set It is expressed as: ； in, Indicates the amount of electric energy injected by the seller, , These are the preset coefficients for the linear term and the quadratic term respectively.

4. The method for adjusting the multi-agent distributed transaction volume of a distribution network based on the ADMM algorithm according to claim 1 is characterized in that: The utility function of the buyer set is It is expressed as: ； in, represents the buyer’s electricity consumption, , These are the preset coefficients for the linear term and the quadratic term respectively.

5. The method for adjusting multi-agent distributed transaction volume in distribution network based on ADMM algorithm according to claim 1 is characterized in that: The consistency problem is decomposed into sub-problems of a single seller and a single buyer, including: Decomposing the first set of subproblems of the seller set from the consistency problem , The expression formula is: ； in, is the seller's auxiliary variable to be solved, and , represents the transaction price matrix reached by the buyer and seller, represents the preset penalty parameter, express and The Hadamard product of express and Hadamard product of; Based on the first set of sub-problems , further decompose the sub-problem of seller i for: ； in, represents the cost function of seller i, represents the amount of electric energy injected by seller i, express The i-th row of express The transpose of the i-th row of express The i-th row of express The i-th row of .

6. The method for adjusting the multi-agent distributed transaction volume of distribution network based on ADMM algorithm according to claim 1 is characterized in that: The consistency problem is decomposed into sub-problems of a single seller and a single buyer, including: Decomposing the second set of sub-problems of the buyer set from the consistency problem , The expression formula is: ； in, is the buyer's auxiliary variable to be solved, and , represents the transaction price matrix reached by the buyer and seller, represents the preset penalty parameter, express and The Hadamard product of express and Hadamard product of; Based on the second set of sub-problems , further decompose the sub-problem of buyer j for: ； in, represents the utility function of buyer j, represents the electricity consumption of buyer j, express The jth column of express The transpose of the jth column of express The jth column of express The transpose of the j-th column of .

Images