Power distribution network voltage coordination method and system considering p2p transaction risk

Abstract

The present application belongs to the field of power distribution network optimization scheduling, and provides a power distribution network voltage coordination method and system considering P2P transaction risk, the method comprising: constructing a producer-consumer P2P energy transaction model and an optimization decision model of a power distribution system operator DSO; constructing a coordination model based on the fusion of a dynamic operating envelope DOE and a risk network usage charge RC-NUC guide; based on the coordination model based on the fusion of the dynamic operating envelope DOE and the risk network usage charge RC-NUC guide, constructing a double-layer coordination optimization framework of multi-agent interaction between the power distribution system operator DSO and the producer-consumer; solving the double-layer coordination optimization framework of multi-agent interaction between the power distribution system operator DSO and the producer-consumer to obtain the coordination decisions of the multi-agent. The DOE and RC-NUC are used to guide the producer-consumer to adjust the transaction strategy to ensure network safety, only the price and power information are exchanged between the multi-agents, the privacy and autonomy of the transaction are taken into account, and the coordinated optimization of the power distribution network safety and the economic benefits of the market participants is realized.

Description

TECHNICAL FIELD

[0001] The application belongs to the field of optimal dispatching of distribution networks, and relates to a distribution network voltage coordination method and system considering P2P transaction risks. BACKGROUND

[0002] The rapid popularization of distributed energy promotes new modes of power system operation and market transactions, especially the rise of peer-to-peer (P2P) markets. Prosumers with power generation and consumption capabilities participate in energy transactions and sharing as independent subjects, which can effectively balance supply and demand and improve energy efficiency. However, the demand of multiple subjects for privacy and autonomy drives the market to be decentralized. In a free market environment, prosumers driven by the desire for profit are prone to risks such as voltage out-of-limit after transactions. Therefore, integrating and exploiting the flexibility of local systems, coordinating the economic efficiency of prosumers and the safety of distribution networks, has become a key challenge.

[0003] Existing research introduces a distribution system operator (DSO) to supervise and guide prosumer transactions to ensure network safety. Some research uses sensitivity analysis to prevent transactions that endanger network safety, or optimizes power flow through network reconfiguration, on-load tap changers, and reactive power management to meet network constraints, but the above direct control methods rely on DSO to centrally obtain transaction information and lack an incentive mechanism. Therefore, indirect guidance methods based on price signals are proposed, such as designing network usage charge (NUC) based on electrical distance to increase transaction costs, or using distribution locational marginal pricing (DLMP) to reflect power flow distribution, and accurately guiding prosumer transactions through price decomposition, taking into account network safety and transaction privacy. However, when P2P transactions result in no solution to the optimal power flow (OPF), DSO cannot extract the Lagrange multiplier to calculate DLMP and NUC, limiting the applicability of this method. SUMMARY

[0004] The purpose of the present application is to provide a distribution network voltage coordination method and system considering P2P transaction risks, to solve the problem of P2P transaction risks of prosumers in distribution networks.

[0005] To achieve the above purpose, the present application adopts the following technical solutions:

[0006] In a first aspect, the present application provides a distribution network voltage coordination method considering P2P transaction risks, comprising:

[0007] According to the internal power surplus or deficit of the producer and consumer, the power is purchased or sold to the main grid or shared with other producers and consumers, a producer and consumer P2P energy transaction model is constructed; the distribution network is modeled with the target of minimizing the total operation cost of the distribution system operator DSO, and an optimization decision model of the distribution system operator DSO is obtained;

[0008] A coordination model based on the fusion of dynamic operation envelope DOE and risk crossing network fee RC-NUC guidance is constructed;

[0009] Based on the coordination model based on the fusion of dynamic operation envelope DOE and risk crossing network fee RC-NUC guidance, a double-layer coordination optimization framework of multi-agent interaction of the distribution system operator DSO and the producer and consumer is constructed;

[0010] The double-layer coordination optimization framework of multi-agent interaction of the distribution system operator DSO and the producer and consumer is solved in combination with the producer and consumer P2P energy transaction model and the optimization decision model of the distribution system operator DSO, and the coordination decision of the multi-agent is obtained.

[0011] Optionally, the construction of the producer and consumer P2P energy transaction model according to the internal power surplus or deficit of the producer and consumer, the purchase or sale of the power to the main grid or the sharing of the power with other producers and consumers comprises the following steps.

[0012] Each producer and consumer is configured with photovoltaic power generation, micro gas turbine, energy storage and demand response load, and for a certain producer and consumer The optimal operation model of the P2P energy transaction is expressed as follows:

[0013] Objective function

[0014] (1)

[0015] (2)

[0016] (3)

[0017] (4)

[0018] (5)

[0019] (6)

[0020] In the formula, , and are cost coefficients of MT; is a unit non-adequate cost of load deviation; is a battery degradation cost coefficient; and are on-grid electricity price and time-of-use electricity price; is the original load; is the producer-consumer transaction object set. is the decision variable of producer-consumer k; MT output power, actual power load, BES charge-discharge power, buy-sell power with DSO, P2P transaction power and price; the total objective function is composed of MT power generation cost, DR inadaptation cost, BES degradation cost, P2P and P2G transaction cost;

[0021] constraint conditions

[0022] (7)

[0023] (8)

[0024] (9)

[0025] (10)

[0026] (11)

[0027] (12)

[0028] (13)

[0029] (14)

[0030] (15)

[0031] (16)

[0032] wherein, is the maximum output power of the MT; the maximum / minimum power demand of the demand response load at time period t is set; is the charge-discharge efficiency; is the capacity of the energy storage unit at time t; is the state of charge; is the maximum and minimum limit of the energy storage capacity; is the initial and ending capacity of the energy storage; is the maximum charge-discharge power; ​​​​​​​​​​​The net power injection is given by constraint (7), which is the power output limit of MT. Constraint (8) is the upper and lower bounds of the actual load power. Constraint (9) is the relationship between actual and expected load demand. For BES, constraint (10) is the relationship between the measured SoC and the charging / discharging power. Constraints (11) and (12) are the constraints and initial / final conditions of SoC, respectively. Constraint (13) is the charging and discharging power. Constraint (14) is that the amount of P2P transaction power negotiated between two producers and consumers is equal and the signs are opposite. Constraints (15) and (16) are the active power balance and net power injection equations.

[0033] Optionally, the step of modeling the distribution network to obtain an optimization decision-making model for the distribution system operator (DSO) with the objective of minimizing the total operating cost of the DSO includes:

[0034] The distribution network is modeled using the DistFlow model, and the radial distribution network is represented by a tree diagram. The sets of nodes and branches are respectively and The root node is connected to the main network and numbered 1; the remaining nodes are each connected to a parent node. and child node set ; will node arrive The branch number is j; the specific model of DSO is shown below:

[0035] objective function

[0036] (17)

[0037] In the formula, The marginal electricity price at the node; The active power injection is for the root node; the objective function represents minimizing the total operating cost of the DSO; the first term represents the cost of purchasing electricity from the wholesale market; the last two terms give the revenue from P2G transactions;

[0038] Constraints

[0039] (18)

[0040] (19)

[0041] (20)

[0042] (twenty one)

[0043] (twenty two)

[0044] (twenty three)

[0045] (24)

[0046] (25)

[0047] (26)

[0048] where, and denote the square of the voltage magnitude at node i and its parent, respectively; denotes the square of the branch j current magnitude; parameters and denote the resistance and reactance of branch j, respectively; and denote the active and reactive load at node i, respectively; and are the actual and maximum output power of generator m; is the net injection power of a producer-consumer connected at node i; and denote the active and reactive power flow of branch i at time period t; and denote the upper and lower bound of the square of the voltage magnitude at node i; constraints (19) and (20) are the active and reactive power balance; constraints (21) and (22) are the power limits at the receiving and sending nodes of each line; constraint (23) is the voltage drop of distribution lines; the branch current constraint (24) is formulated in the standard second-order cone form; constraint (25) limits the voltage magnitude at nodes. The generator constraints are given by (26).

[0049] Optionally, the constructing the coordination model based on a fused dynamic operating envelope (DOE) comprises:

[0050] The DSO calculates the DOE, sets a specific safety domain for the net injection power of each producer-consumer, and the calculation is performed in two steps: in the first step, the DSO checks the network safety of the expected import and export power of the producer-consumer, and if the voltage constraint is not violated, the expected transaction power is approved; if the rule is violated, the DOE is calculated in the second step, and the DOE result is directly delivered to the producer-consumer;

[0051] The objective function is designed to minimize the sum of squares of the difference between the expected net injection power of the producer-consumer and the DOE defined value Specifically, as follows:

[0052] (27)

[0053] (28)

[0054] ​ (29)

[0055] (30)

[0056] Optionally, the risk crossing fee RC-NUC guided coordination model comprises:

[0057] The NUC is generated by calculating the difference between the DLMP of the nodes, as a price signal to guide the user's electricity behavior, reflecting the risk of node voltage out of limits.

[0058] The NUC is improved by the voltage safety related risk coefficient RC to form RC-NUC, which is used to guide producers and consumers to carry out "grid-friendly" transactions, and a safety margin value is set , defined as the absolute value of the node voltage deviation, when , it is considered that the node has a risk, and the greater the value exceeds, the higher the potential risk of voltage out of limits;

[0059] Definition:

[0060] (31)

[0061] (32)

[0062] In the formula, is the risk coefficient of producer i and consumer j; is the node set that is greatly affected by the transaction between producer i and consumer j; as can be seen from formula (31), when the node has a risk, the value of risk coefficient is greater than 1; the VSC between the two nodes is obtained by calculation:

[0063] (33)

[0064] In the formula, is the 0-1 variable of the node greatly affected by the transaction between producer i and producer j, if its value is 1, it means that the node voltage is greatly affected by the transaction, and is included in ; is the voltage change of node k caused by the transaction between producer i and producer j, is the VSC of the voltage amplitude at node k to the active power injection at node i; is the adjustable risk control parameter;

[0065] The expression of RC-NUC is set as:

[0066] (34)

[0067] In the formula: DLMP for node i.

[0068] Optionally, based on the integrated dynamic operating envelope (DOE) and risk over network charge (RC-NUC) guided coordination model, a double-layer coordination optimization framework for the interaction between the distribution system operator (DSO) and the prosumers is constructed, including:

[0069] The lower layer of the double-layer coordination optimization framework optimizes the scheduling resources of the prosumers, negotiates the P2P transaction according to the demand and transaction preference, or transacts with the DSO according to the time-of-use price and the network access price; the upper layer of the double-layer coordination optimization framework performs safety check based on the network information and the expected import and export power of the prosumers, if the safety constraint is violated, the DOE is calculated and issued, then the DSO calculates the DLMP and the risk over network charge to guide the prosumers to adjust the transaction decision.

[0070] Optionally, the P2P energy transaction model of the prosumers and the optimization decision model of the distribution system operator (DSO) are combined to solve the double-layer coordination optimization framework for the interaction between the distribution system operator (DSO) and the prosumers, and the coordinated decision of the multiple parties is obtained, including:

[0071] The objective function and the constraint condition of the P2P energy transaction model of the prosumers are updated as follows:

[0072] (35)

[0073] (36)

[0074] (37)

[0075] In the formula, represents the DOE limit issued by the DSO to the prosumer k in the period t; is positive, indicating that it is limited by import; is negative, indicating that it is limited by export;

[0076] The ADMM algorithm is used to iteratively solve the P2P transaction problem, first, an auxiliary variable is introduced to decouple the consistency constraint (14):

[0077] (38)

[0078] In the formula, represents the auxiliary variable introduced, which is regarded as the P2P transaction amount estimate value of the prosumer k; is the Lagrange multiplier, representing the shadow price of the transaction power, defined as the P2P price;

[0079] Then, the augmented Lagrangian function is established, and the P2P transaction problem is further decomposed into a single sub-problem that can be independently solved by each prosumer k:

[0080] (39)

[0081] wherein, is a positive penalty factor; when the lower-level optimal solution is obtained, the sum of P2P transaction costs of all producers and consumers is zero;

[0082] Let v be the iteration number of the ADMM algorithm, the updating methods of the Lagrange multiplier and the auxiliary variable are as follows:

[0083] (40)

[0084] (41)

[0085] The convergence criterion of the original residual error and the dual residual error is as follows:

[0086] (42)

[0087] (43)

[0088] wherein, and are the set error precisions, respectively;

[0089] An adaptive updating method of the penalty factor is adopted to accelerate the iteration convergence speed:

[0090] (44)

[0091] wherein, is a constant for judging the relationship between the original and dual residual errors; and are the expansion and reduction multiples of the penalty factor, respectively;

[0092] Let z be the iteration number of the upper and lower layers; after the lower-level loop converges, the DSO updates the DLMP and the RC-NUC according to the OPF result; when the variables of the DSO and the producers and consumers satisfy formula (45) and (46), the decisions of the two parties no longer change, and the benefits are optimal;

[0093] (45)

[0094] (46).

[0095] In the second aspect, the application provides a power distribution network voltage coordination system considering P2P transaction risks, comprising:

[0096] A transaction model and decision model construction module is configured to construct a P2P energy transaction model of the producers and consumers according to the internal power surplus or deficit of the producers and consumers, and to purchase or sell power from the main grid or share power with other producers and consumers.

[0097] A coordination model construction module is configured to construct a coordination model based on the fusion of the dynamic operating envelope DOE and the risk of crossing the network fee RC-NUC guidance.

[0098] A double-layer coordination optimization framework construction module is configured to construct a double-layer coordination optimization framework of the multi-agent interaction between the DSO and the producers and consumers based on the coordination model of the fusion of the dynamic operating envelope DOE and the risk of crossing the network fee RC-NUC guidance.

[0099] An output solving module is configured to solve the double-layer coordination optimization framework of the multi-agent interaction between the DSO and the producers and consumers based on the P2P energy transaction model of the producers and consumers and the optimization decision model of the DSO, and to obtain the coordination decision of the multi-agent.

[0100] In a third aspect, the present application provides a computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the power distribution network voltage coordination method considering P2P transaction risks.

[0101] In a fourth aspect, the present application provides a computer readable storage medium, which stores a computer program, wherein the computer program is executable on a processor to implement the steps of the power distribution network voltage coordination method considering P2P transaction risks.

[0102] Compared with the prior art, the present application has the following technical effects:

[0103] The present application discloses a power distribution network voltage management method considering P2P transaction risks, constructs a DSO-producer and consumer double-layer optimization model based on DOE and RC-NUC, and designs an improved distributed algorithm for solving, thereby realizing the coordinated optimization of the power distribution network safety and the economic efficiency of the producers and consumers in the P2P energy market. The DOE constraint ensures the solvability of the power flow, the RC-NUC is used to further guide the producers and consumers to participate in the friendly interaction of the power grid, effectively balances the market benefits and system safety, and improves the reliability of the P2P transaction risk management.

[0104] Compared with the traditional method, the present application has the following advantages:

[0105] 1) A DSO-consumer two-level optimization framework is proposed, which takes into account system safety and market efficiency. The upper layer DSO is guided by system safety, and the DOE and RC-NUC methods are used to guide the consumer transaction; the lower layer constructs a decentralized P2P market, and the consumer participates in the transaction to maximize economic benefits, realizes multi-party privacy protection and efficiency collaborative improvement;

[0106] 2) The RC-NUC method is proposed, which quantifies the impact of decentralized P2P transactions on power flow in distribution networks through price signals, accurately reflects the influence degree of each transaction on node voltage risk, and effectively guides the optimization of transaction behavior of producers and consumers, and improves the reliability of system risk management in the P2P transaction scenario.

[0107] 3) A DOE calculation method suitable for P2P transactions is proposed, which ensures that the upper layer has a solution while allowing producers and consumers to freely determine transaction power and price under DOE restrictions. The lower layer P2P transaction problem is solved by using the adaptive alternating direction method of multipliers (ADMM), which accelerates convergence by dynamically adjusting parameters, realizes independent optimization under limited information sharing, and meets the privacy and autonomy needs of transactions. BRIEF DESCRIPTION OF DRAWINGS

[0108] Figure 1 The DSO-consumer coordination optimization framework of the present application;

[0109] Figure 2 The solving step diagram of the present application.

[0110] Figure 3 The flowchart of the present application. DETAILED DESCRIPTION

[0111] The present application is further described below in conjunction with the accompanying drawings:

[0112] Example 1, please refer to Figure 3 The present application provides a power distribution network voltage coordination method considering P2P transaction risk, which includes:

[0113] According to the internal power surplus or deficit of the producer and consumer, the power is purchased or sold to the main network or shared with other producers and consumers, and a producer and consumer P2P energy transaction model is constructed; the power distribution system operator DSO is modeled to minimize the total operation cost, and the optimization decision model of the power distribution system operator DSO is obtained;

[0114] A coordination model based on fusion of dynamic operation envelope DOE and risk over network fee RC-NUC guidance is constructed;

[0115] A coordination model based on fusion of dynamic operation envelope (DOE) and risk crossing network fee (RC-NUC) guidance is constructed to build a double-layer coordination optimization framework of multi-agent interaction between distribution system operator (DSO) and producers and consumers.

[0116] A double-layer coordination optimization framework of multi-agent interaction between DSO and producers and consumers is solved by combining a P2P energy transaction model of producers and consumers and an optimized decision model of DSO to obtain coordinated decisions of multi-agent.

[0117] The application proposes a power distribution network voltage management method considering P2P transaction risk, and constructs a DSO-producer and consumer coordination optimization method based on DOE and RC-NUC guidance in a P2P market transaction environment. The method considers network constraints, uses DOE and RC-NUC to guide producers and consumers to adjust transaction strategies to ensure network safety, and only exchanges price and power information between multi-agent, taking into account the privacy and autonomy of transactions, and successfully realizes the coordinated optimization of power distribution network safety and economic benefits of market participants. The method has the following advantages:

[0118] 1) The proposed RC-NUC fully considers the influence of P2P transaction on power flow of the power distribution network, accurately guides the producers and consumers to adjust the transaction behavior, and effectively improves the system risk response capability;

[0119] 2) The proposed DOE method dynamically optimizes import and export power, allows producers and consumers to independently determine transaction power and price within the limit range, and at the same time ensures that the optimal power flow problem of the power distribution network has a solution;

[0120] 3) The adaptive ADMM algorithm improves the convergence performance and guarantees the independence and privacy of the transaction process of the producers and consumers.

[0121] In embodiment 2, the application provides a power distribution network voltage coordination method considering P2P transaction risk, which includes:

[0122] A DSO-producer and consumer double-layer optimization model based on DOE and RC-NUC is constructed, and an improved distributed algorithm is designed to solve it, realizing the coordinated optimization of power distribution network safety and economic benefits of producers and consumers in the P2P energy market.

[0123] The specific implementation steps are as follows:

[0124] Step 1: Construct a double-layer coordination optimization framework of multi-agent interaction between DSO and producers and consumers.

[0125] First, the interaction relationship between DSO and producers and consumers in the power distribution network is described, and the constructed double-layer coordination optimization framework is as follows: Figure 1The lower layer of producers and consumers optimizes scheduling resources, negotiates P2P transactions according to demand and transaction preferences, and can also trade with the DSO according to time-of-use electricity prices and on-grid electricity prices. All transactions are autonomous behaviors of producers and consumers, ensuring their autonomy and privacy. The upper layer of DSO performs security verification based on network information and expected import and export power of producers and consumers. If the security constraints are violated, the DSO calculates and issues the DOE. Then the DSO calculates the DLMP and the risk of over-network fees to guide producers and consumers to adjust their transaction decisions and reduce the risk of P2P transactions to the distribution network.

[0126] Step 2: Construct the mathematical model of producers and consumers and DSO.

[0127] Step 2.1: Prosumer P2P energy transaction model.

[0128] Prosumers are active participants in the P2P market. It is assumed that each prosumer is equipped with flexible resources such as photovoltaic (PV), micro-turbines (MT), battery energy storage (BES), and demand response (DR) loads. In any time period, a prosumer can buy or sell electricity to the main grid or share electricity with other prosumers according to its internal power surplus or deficit. For a certain prosumer , the optimal operation model of its P2P energy transaction is represented as follows.

[0129] (1) Objective function

[0130] (1)

[0131] (2)

[0132] (3)

[0133] (4)

[0134] (5)

[0135] (6)

[0136] In the formula, , and are the cost coefficients of MT; is the unit cost of load deviation; is the battery degradation cost coefficient; and are the on-grid electricity price and time-of-use electricity price; is the original load; is the set of transaction objects for producers and consumers. is the decision variable of producer and consumer k. , , / , / , and denote the MT output power, actual power load, BES charge and discharge power, buy and sell power with DSO, P2P transaction power and price. The total objective function (1) is composed of MT generation cost, DR inadaptation cost, BES degradation cost, P2P and P2G transaction cost. The specific formula of each cost function is represented as (2)-(6).

[0137] (2) Constraint conditions

[0138] (7)

[0139] (8)

[0140] (9)

[0141] (10)

[0142] (11)

[0143] (12)

[0144] (13)

[0145] (14)

[0146] (15)

[0147] (16)

[0148] In the formula, is the maximum output power of MT; the maximum / minimum electricity demand of demand response load at t period is set; and are charge and discharge efficiencies; is the capacity of energy storage unit at t time; is the state of charge; and are maximum and minimum limits of energy storage capacity; and are initial and ending capacities of energy storage; and are maximum charge and discharge power;​​ The net injection power. Constraint (7) is the power output limit of MT. Constraint (8) is the upper and lower bound of the actual load power. Constraint (9) is the relationship between the actual and expected load demand. Constraint (10) is the relationship between the measured SoC and the charging / discharging power for BES. Constraints (11) and (12) are the SoC constraints and initial / final conditions, respectively. Constraint (13) is the charging and discharging power. Constraint (14) is that the P2P transaction power quantity negotiated between the two producers and consumers is equal, but the signs are opposite. Constraints (15) and (16) are the active power balance and net power injection equations.

[0149] The P2P transaction power and price can be determined by solving the above model. The DSO receives the net power injection information submitted by the producers and consumers and performs OPF to minimize the operating cost and ensure that the distribution network can operate within the constraint range.

[0150] Step 2.2: DSO's optimization decision model.

[0151] The distribution network is modeled using the DistFlow model. The radial distribution network is represented by a tree graph , and the node and branch sets are and , respectively. The root node is connected to the main network and numbered 1, and the remaining nodes are connected to the parent node and child node set , respectively. The branches from node to are numbered j. The specific model of the DSO is as follows:

[0152] (1) Objective function

[0153] (17)

[0154] where is the node marginal price. is the active power injection of the root node. The objective function (17) represents the minimization of the total operating cost of the DSO. The first term represents the cost of purchasing electricity from the wholesale market. The last two terms give the income of the P2G transaction.

[0155] (2) Constraint conditions

[0156] (18)

[0157] (19)

[0158] (20)

[0159] (21)

[0160] (22)

[0161] (23)

[0162] (24)

[0163] (25)

[0164] (26)

[0165] where, and denote the voltage magnitude squared at node i and its parent node, respectively; denotes the square of branch j current magnitude; parameters and denote the resistance and reactance of branch j, respectively; and denote the active and reactive load at node i, respectively; and are the actual and maximum output power of generator m; is the net injection power of a producer-consumer connected at node i; and denote the active and reactive power flow of branch i at time period t; and denote the upper and lower bound of voltage magnitude squared at node i. Constraints (19) and (20) are the active and reactive power balance. Constraints (21) and (22) are the power limits at the receiving and sending nodes of each line. Constraint (23) is the voltage drop of distribution lines. The branch current constraint (24) is formulated in the standard second-order cone form. Constraint (25) limits the voltage magnitude at nodes. The generator constraints are given by (26).

[0166] The DSO can only perform OPF calculation based on the net injection power submitted by the producer-consumers, and cannot directly control the P2P transaction power and price, which may lead to voltage out-of-limit problems. Therefore, it is necessary to coordinate the relationship between the DSO and the producer-consumers, and to achieve the overall consideration of the safety of the distribution network and the privacy and economy of the producer-consumers.

[0167] Step 3: Construct a coordination model based on DOE and RC-NUC guidance.

[0168] Step 3.1: DOE calculation model.

[0169] To reduce the risk of P2P transactions on the distribution network, the DSO first calculates the DOE, which sets a specific safety domain for each producer-consumer's net injection power. The calculation is done in two steps: first, the DSO checks the network safety of the producer-consumer's expected import and export power. If the voltage constraint is not violated, the expected transaction power is approved; if there is a violation, the DOE is calculated in the second step. The DOE result is directly passed to the producer-consumer, allowing them to autonomously manage their assets while protecting their privacy and ensuring that network constraints are not violated.

[0170] To improve transaction fairness and avoid excessive power reduction for producer-consumers far from the root node, the objective function is designed to minimize the sum of the squares of the differences between the producer-consumer's expected net injection power and the DOE-defined value, as follows: between the producer-consumer's expected net injection power and the DOE-defined value, as follows:

[0171] (27)

[0172] s.t.(18), (20)-(25);

[0173] (28)

[0174] (29)

[0175] (30)

[0176] Since the DSO defines the DOE for the producer-consumer and has no direct control over their internal assets and energy transactions, certain principles need to be considered: (1) If a producer-consumer wants to import power, they cannot be forced to export power or import more power than expected. (2) If a producer-consumer wants to export power, their export can be reduced to zero, but they cannot be forced to import power or export more power than expected. The corresponding constraints are represented in (28)-(29). In addition, the power generation and load of node i remain unchanged before and after the DOE calculation. Therefore, the active power balance constraint of node i based on equation (19) is re-expressed as equation (30).

[0177] Step 3.2: Definition of RC-NUC guiding method and model.

[0178] The NUC is generated by calculating the difference in DLMP between nodes, serving as a price signal to guide user electricity behavior and reflecting the risk of node voltage exceeding the limit. However, the traditional NUC method lacks sensitivity to voltage constraints, especially when the voltage is close to the safety boundary. The uncertainty of new energy output may cause the system to transition from a safe state to a risk of exceeding the limit. In this case, the traditional NUC cannot accurately reflect the network status and effectively guide producer-consumers to adjust their transaction strategies to alleviate local voltage problems.

[0179] ​To solve the above problems, the risk coefficient (RC) related to voltage security is proposed to improve the NUC, forming the RC-NUC, which is used to guide producers and consumers to carry out "grid-friendly" transactions and improve the voltage security of the power grid in uncertain environment. The core role of RC-NUC is to reduce the risk that P2P transactions and uncertain factors may bring. In order to identify the risk, a safety margin value is set , which defines the absolute value of the node voltage deviation. When , it is considered that the node has a risk, and the greater the value exceeds, the higher the potential risk of voltage limit.

[0180] RC can fully reflect the risk of any transaction to the node voltage, which can be represented by the part exceeding the safety margin value. The greater the RC, the higher the potential risk of the system under the transaction, so the value of RC should be a positive feedback to RC-NUC, and it can represent the influence of each transaction on the node voltage. Define:

[0181] (31)

[0182] (32)

[0183] In the formula, RCi,j is the risk coefficient of producer i and consumer j; is the node set that is more affected by the transaction between producer i and consumer j. According to formula (31), when the node has a risk, the value of is greater than 1. Considering that the voltage sensitivity coefficient (VSC) represents the change of the node voltage amplitude caused by the change of the power of the power system, it can be used to describe the influence of each transaction on each node voltage. Therefore, the VSC between two nodes can be obtained by calculating:

[0184] (33)

[0185] In the formula, is a 0-1 variable of the node that is more affected by the transaction between producer i and consumer j. If its value is 1, it means that the node voltage is more affected by the transaction, which can be included in ; is the change of the node k voltage caused by the transaction between producer i and consumer j, is the VSC of the voltage amplitude at node k to the active power injection at node i; is the adjustable risk control parameter.

[0186] Therefore, to reflect the influence of regional resources on the safe operation of distribution networks in the time-space dimension, and to evaluate the severity of each transaction on the voltage risk of different nodes, the expression of RC-NUC is set as:

[0187] (34)

[0188] wherein, is the DLMP corresponding to node i, which can be obtained by the relevant dual variable in step 2.2.

[0189] In any transaction, producer-consumer i and producer-consumer j will receive the corresponding RC-NUC fairly without price discrimination. RC adjusts dynamically according to voltage changes, and when the grid voltage approaches the safety boundary, RC-NUC increases, prompting producer-consumers to adjust their transaction behavior in response to changes in grid conditions, thereby enhancing the adaptability and flexibility of the system. At the same time, the presence of RC does not affect the physical and economic meaning of the original NUC. Due to the additional transaction cost brought by RC-NUC, producer-consumers may reduce the transaction power appropriately to reduce energy costs. Therefore, the introduction of RC-NUC effectively guides producer-consumers to adjust their transaction behavior autonomously, ensuring that transactions comply with network constraints.

[0190] Step 4: Solution process of the bi-level coordination optimization framework.

[0191] According to the decision-making model of each subject constructed in step 2 and the safety guidance method in step 3, producer-consumers participate in the P2P market to minimize energy costs, while the DSO guides producer-consumers through DOE and RC-NUC to ensure network safety and maximize economic benefits. Both parties influence each other and adjust their own decisions according to the other party's strategy.

[0192] When producer-consumers are guided by the RC-NUC price signal, they need to pay an additional risk operation fee. Since both buyers and sellers share equal profits in P2P transactions, this fee is shared by both parties. Therefore, the producer-consumer model in step 2.1 needs to be modified, and the objective function and constraint conditions are updated as follows:

[0193] (35)

[0194] s.t.(2)-(16);

[0195] (36)

[0196] (37)

[0197] wherein, represents the DOE limit issued by the DSO to producer-consumer k in period t. is positive, indicating that it is subject to import restrictions; is negative, indicating that it is subject to export restrictions.

[0198] The P2P transaction problem is solved iteratively by ADMM algorithm. First, auxiliary variables are introduced to decouple the consistency constraint (14):

[0199] (38)

[0200] where is the introduced auxiliary variable, which can be regarded as the P2P transaction volume estimate of producer-consumer k. is the Lagrange multiplier, which represents the shadow price of transaction power, and is defined as the P2P price.

[0201] Then, the augmented Lagrangian function is established, and the P2P transaction problem is further decomposed into individual sub-problems that can be solved independently by each producer-consumer k:

[0202] (39)

[0203] s.t. (2)-(4), (6)-(13), (15)-(16), (36)-(37).

[0204] where is a positive penalty factor. When the lower-level optimal solution is obtained, the sum of P2P transaction costs of all producer-consumers is zero, so the objective function (39) of the decomposed sub-problem does not contain the P2P transaction cost.

[0205] Let v be the iteration number of ADMM algorithm, and the update methods of Lagrange multiplier and auxiliary variable are as follows:

[0206] (40)

[0207] (41)

[0208] The convergence criteria for the primal residual and the dual residual are as follows:

[0209] (42)

[0210] (43)

[0211] where and are the set error precisions, respectively.

[0212] Since the penalty factor has a great influence on the convergence performance of ADMM, using a fixed step size may waste computational resources and be very dependent on the choice of initial value. Therefore, an adaptive updating method for the penalty factor is adopted to speed up the iteration convergence speed:

[0213] (44)

[0214] wherein, is a constant to determine the relationship between the original and the dual residual; and are the multiples of the penalty factor expansion and reduction, respectively.

[0215] Let z be the iteration number of the upper and lower layers. After the lower layer loop converges, the DSO updates the DLMP and RC-NUC according to the OPF result. When the variables of the DSO and the producers and consumers satisfy equations (45) and (46), the decisions of both sides no longer change, the benefits are optimal, the system risk is successfully eliminated, and the benefit improvement of market subjects and the safe operation of the system under the market environment are achieved.

[0216] (45)

[0217] (46)

[0218] The solving steps of the DSO-producer and consumer bi-level optimization model based on the DOE and RC-NUC under any operating boundary are shown in Table 1. Figure 2 Table 1. Solving steps of the DSO-producer and consumer bi-level optimization model based on the DOE and RC-NUC under any operating boundary

Claims

1. A power distribution network voltage coordination method considering P2P transaction risk, characterized in that, The application relates to a power grid energy transaction model of a power consumer, and a power grid energy transaction model of a power consumer. A coordination model based on a dynamic operation envelope DOE and a risk network usage cost RC-NUC is constructed. A double-layer coordination optimization framework of multi-agent interaction between a power grid energy transaction model of a power consumer and a power grid energy transaction model of a power consumer is constructed. The double-layer coordination optimization framework of multi-agent interaction between a power grid energy transaction model of a power consumer and a power grid energy transaction model of a power consumer is solved in combination with the power grid energy transaction model of the power consumer and the optimization decision model of the power grid energy transaction model of the power consumer, so that coordinated decisions of the multi-agent are obtained. The application relates to a power grid energy transaction model of a power consumer, and a power grid energy transaction model of a power consumer. The DOE is calculated by the DSO, and a specific safety domain is set for the net injection power of each power consumer, and the calculation is carried out in two steps: in the first step, the DSO checks the network safety of the expected import and export power of the power consumer, and if the voltage constraint is not violated, the expected transaction power is approved; if the rule is violated, the DOE is calculated in the second step, and the DOE result is directly transmitted to the power consumer. The application relates to a power grid energy transaction model of a power consumer, and a power grid energy transaction model of a power consumer. The objective function is designed to minimize the producer-consumer expected net injection power The sum of the square of the difference between the DOE defined values The sum of the square of the difference between the DOE defined values (27) (28) (29) (30)； The expression of the RC-NUC is set as: A double-layer coordination optimization framework of multi-agent interaction between a power grid energy transaction model of a power consumer and a power grid energy transaction model of a power consumer is constructed based on the coordination model based on the dynamic operation envelope DOE and the risk network usage cost RC-NUC, and the double-layer coordination optimization framework comprises the following steps: The NUC is improved by a voltage safety related risk coefficient RC to form RC-NUC, which is used to guide the producers and consumers to carry out "grid-friendly" transactions, and set a safety margin value , define as the absolute value of the node voltage deviation, when , it is regarded that the node has a risk, and the greater the exceeded value, the higher the potential risk of voltage out-of-limit. The lower layer of the double-layer coordination optimization framework is an optimization scheduling resource of the power consumer, which negotiates the P2P transaction according to demand and transaction preference, or trades with the DSO according to the time-of-use price and the on-grid price; the upper layer of the double-layer coordination optimization framework is the DSO, which carries out safety checking based on network information and expected import and export power of the power consumer, and if the safety constraint is violated, the DOE is calculated and issued, then the DSO calculates the DLMP and the risk network usage cost, and guides the power consumer to adjust the transaction decision. (31) (32) wherein, is the risk coefficient of producer i and consumer j; is the set of nodes that are more affected by the transaction of producer i and consumer j; from equation (31), when a node has a risk, the risk coefficient is greater than 1; by calculating the VSC between two nodes, we get: (33) In the formula, is a 0-1 variable of the node greatly affected by the transaction between producer consumer i and producer consumer j, if its value is 1, it means that the voltage of the node is greatly affected by the transaction, and the node is listed in ; is the voltage change of node k caused by the transaction between producer consumer i and producer consumer j, is the VSC of the voltage amplitude at node k to the active power injection at node i; is the adjustable risk control parameter; The application relates to a power grid energy transaction model of a power consumer, and a power grid energy transaction model of a power consumer. (34) In the formula: is the DLMP corresponding to node i; Objective function Constraint condition 2. The power distribution network voltage coordination method considering P2P transaction risk according to claim 1, characterized in that, The application relates to a power grid energy transaction model of a power consumer, and a power grid energy transaction model of a power consumer. Each prosumer is equipped with photovoltaic generation, micro gas turbine, energy storage and demand response load, and for a certain prosumer The optimal operation model of its P2P energy transaction is represented as follows: Objective function (1) (2) (3) (4) (5) (6) wherein, , and are the cost coefficients of MTs, respectively; is the unit discomfort cost of load deviation; is the battery degradation cost coefficient; and are the on-grid electricity price and time-of-use price; is the original load; is the set of transaction objects of producers and consumers; is the decision variable of producer and consumer k; , , / , / , and represent the MT output power, actual power load, BES charging and discharging power, buying and selling power with DSO, P2P transaction power and price; the total objective function is composed of MT generation cost, DR discomfort cost, BES degradation cost, P2P and P2G transaction cost; Constraint condition (7) (8) (9) (10) (11) (12) (13) (14) (15) (16) wherein, is the maximum output power of MT; the maximum / minimum electricity demand of the demand response load at time period t is set; and is the charging / discharging efficiency; is the storage unit capacity at time t; is the state of charge; and is the maximum / minimum limit of the storage capacity; and is the initial and final capacity of the storage; and is the maximum charging / discharging power; is the net injection power; constraint (7) is the power output limit of MT; constraint (8) is the upper and lower bound of the actual load power; constraint (9) is the relationship between the actual and expected load demand; constraint (10) is the relationship between the measured SoC and the charging / discharging power for BES; constraints (11) and (12) are the constraints and initial / final conditions of SoC, respectively; constraint (13) is the charging and discharging power; constraint (14) is that the amount of P2P transaction power negotiated between the two producers and consumers is equal, and the signs are opposite; constraints (15) and (16) are the active power balance and net power injection equations.

3. The power distribution network voltage coordination method considering P2P transaction risk according to claim 1, characterized in that, The double-layer coordination optimization framework of multi-agent interaction between a power grid energy transaction model of a power consumer and a power grid energy transaction model of a power consumer is solved in combination with the power grid energy transaction model of the power consumer and the optimization decision model of the power grid energy transaction model of the power consumer, so that coordinated decisions of the multi-agent are obtained. The distribution network is modeled by using a power distribution network flow distribution DistFlow model, and a tree diagram is used to represent the radial distribution network , a node set and a branch set are respectively and ; the root node is connected with the main network and numbered as 1, and the remaining nodes are respectively connected with a parent node set and a child node set ; the branch from the node to the node is numbered as j; and a specific model of the DSO is as follows: The objective function and the constraint condition of the power grid energy transaction model of the power consumer are updated as follows: (17) wherein is the nodal marginal price; is the root node active power injection; the objective function represents the minimization of the total operating cost of the DSO; the first term represents the cost of the electricity purchased from the wholesale market; the last two terms give the revenues from the P2G transactions; An ADMM algorithm is used to iteratively solve the P2P transaction problem, and first, an auxiliary variable is introduced to decouple the consistency constraint (14): (18) (19) (20) (21) (22) (23) (24) (25) (26) In the formula, and Let i and y represent the squares of the voltage magnitudes of node i and its parent node, respectively. Represents the square of the amplitude of the current in branch j; parameter and Let J represent the resistance and reactance of branch J, respectively. and These represent the active and reactive loads of node i, respectively. and The actual and maximum output power of generator m; The net injected power of the producer-consumer connected at node i; and Let represent the active and reactive power flow of branch i during time period t; and Let represent the upper and lower bounds of the square of the voltage amplitude at node i, respectively; constraints (19) and (20) are for active and reactive power balance; constraints (21) and (22) are for power limits at the receiving and transmitting nodes of each line; constraint (23) is for the voltage drop of the distribution line; branch current constraint (24) is formulated in standard second-order cone form; constraint (25) limits the node voltage amplitude; generator constraint is given by (26).

4. The power distribution network voltage coordination method considering P2P transaction risk according to claim 2, characterized in that, ​ ​ (35) (36) (37) wherein, represents the DOE restriction for the period t issued by the DSO to the producer-consumer k; is positive indicating import restriction; is negative indicating export restriction; ​ (38) In the formula, represents the introduced auxiliary variable, which is regarded as the P2P transaction volume estimation value of producer k; is the Lagrange multiplier, which represents the shadow price of transaction power and is defined as the P2P price; Then, the augmented Lagrangian function is established, and the P2P transaction problem is further decomposed into a single sub-problem that each producer-consumer k can solve independently: (39) In the formula, is a positive penalty factor; when the lower optimal solution is obtained, the sum of P2P transaction costs of all producers and consumers is zero; Let v be the iteration number of the ADMM algorithm, and the update of the Lagrange multiplier and the auxiliary variable is as follows: (40) (41) Original residual and dual residuals The convergence criteria are as follows: (42) (43) wherein and are the set error precisions, respectively. The adaptive updating method of the penalty factor is adopted to accelerate the iteration convergence speed: (44) wherein is a constant to judge the relationship between the original and the dual residual; and are the multiples of the penalty factor enlargement and reduction, respectively; Let z be the iteration number of the upper and lower layers; after the lower layer loop converges, the DSO updates the DLMP and RC-NUC according to the OPF result; when the variables of the DSO and the producer-consumers satisfy equations (45) and (46), the decisions of both parties no longer change, and the benefits are optimal; (45) (46)。 5. A power distribution network voltage coordination system for implementing the power distribution network voltage coordination method considering P2P transaction risks according to claim 1, characterized in that, The transaction model and decision model construction module is configured to construct a producer-consumer P2P energy transaction model according to the internal power surplus or deficit of the producer-consumer, purchase or sell power from the main grid, or share power with other producer-consumers; and model the distribution network with the minimum total operation cost of the distribution system operator DSO as the target to obtain an optimization decision model of the distribution system operator DSO. The coordination model construction module is configured to construct a coordination model based on the fusion of the dynamic operation envelope DOE and the risk network usage charge RC-NUC guidance. The double-layer coordination optimization framework construction module is configured to construct a double-layer coordination optimization framework of the multi-agent interaction between the distribution system operator DSO and the producer-consumer based on the coordination model of the fusion of the dynamic operation envelope DOE and the risk network usage charge RC-NUC guidance. The output solving module is configured to solve the double-layer coordination optimization framework of the multi-agent interaction between the distribution system operator DSO and the producer-consumer based on the producer-consumer P2P energy transaction model and the optimization decision model of the distribution system operator DSO, and obtain the coordination decisions of the multi-agent. The computer program is executed by the processor to realize the steps of the power distribution network voltage coordination method considering P2P transaction risks according to any one of claims 1 to 4.

6. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, The computer program is executed by the processor to realize the steps of the power distribution network voltage coordination method considering P2P transaction risks according to any one of claims 1 to 4.

7. A computer-readable storage medium storing a computer program, wherein the computer program comprises the following steps of: receiving a request for a resource from a client; determining whether the client is authorized to access the resource; and if the client is authorized to access the resource, providing the resource to the client. ​

Images