Distributed Coordination Optimization Method for Virtual Power Plants to Participate in the Primary and Distribution Two-Level Markets Based on Reputation Control Mechanisms

Abstract

The present invention discloses a distributed coordinated optimization method for a virtual power plant to intervene in the primary and distribution two-level markets based on a credit control mechanism, which decomposes the primary and distribution two-level markets into a main grid wholesale market sub-problem and multiple distribution system retail market sub-problems. To make full use of the distributed generation resources aggregated by the virtual power plant and the flexible adjustable load to reduce the operating costs of a strongly uncertain power system, the present invention improves and perfects the organizational implementation process for the virtual power plant to participate in the primary and distribution two-level power markets. In addition, considering that there is a large uncertainty in the actual power of the virtual power plant and the strong individual profit-seeking nature of the virtual power plant agent, which easily leads to default behavior of the virtual power plant, a dynamic credit index for evaluating the contract performance completion degree of the virtual power plant is proposed, and the market default behavior of the virtual power plant is inhibited by setting market access thresholds and paying liquidated damages.

Description

Technical Field

[0001] The present invention relates to electric energy storage technology and market coordination optimization technology, and particularly relates to a distributed coordination optimization method for a virtual power plant to intervene in the main and distribution two-level markets based on a reputation control mechanism. Background Art

[0002] Under the market-dominated mechanism, in order to eliminate the power imbalance that occurs during the operation between the wholesale market on the main grid side and the retail market on the distribution side, it is necessary to give full play to the coordination role of the main and distribution system operators (wholesale market operator on the main grid side / retail market operator on the distribution side) between different levels of markets, enhance the power flow and price interaction between the main and distribution two-level markets, so as to further improve the safe and economic operation efficiency of the power system.

[0003] It can be foreseen that in the future, more and more distributed new energy power generation devices will enter the low-voltage distribution system. However, due to the large number and small capacity of distributed energy sources, it is not suitable to directly participate in market transactions. The distributed energy and flexibility resources in the low-voltage distribution network participate in the market through a virtual power plant (VPP). Considering problems such as the large volatility and uncertainty of the power generation power of distributed power sources within the VPP, and the strong individual profit-seeking nature of market agents, etc., it leads to the fact that market entities such as VPPs are extremely prone to default behaviors, increasing the safety risk of power grid operation, and also triggering the trust loss of system operators towards VPPs, ultimately affecting the enthusiasm of VPPs to participate in market transactions.

[0004] In order to promote the integration of VPPs into power market operations, it is urgent to establish a market management model for evaluating the contract performance completion degree of market entities such as VPPs to ensure the safe operation of the power grid and improve the enthusiasm of VPPs to participate in the market. By introducing reputation evaluation indicators and clarifying the update rules of the reputation values of each trading entity, the reputation of VPPs is evaluated, recorded and updated to encourage users to spontaneously abide by the rules. Considering that the trading volume of the wholesale market on the main grid side is large, the market participants on the main grid side need to meet certain access capacity and grid connection conditions. The retail market on the distribution side is close to various distributed energy and flexibility resources, and the trading volume is small. For VPPs, the access threshold for participating in the wholesale market for trading is significantly higher than that of the retail market. Thus, the level of the VPP agent's reputation value becomes the only critical criterion for it to participate in the retail market on the distribution side or the wholesale market on the main grid side.

[0005] There is almost no coordination between traditional main grid side wholesale market operators and distribution network side retail market operators. The fragmented market management model is no longer applicable to the situation where a large number of distributed power sources are connected to the distribution network. The reasons are as follows. On the one hand, main grid side wholesale market operators must cope with the inevitable uncertainties, fluctuations and intermittencies of large-scale new energy power generation. To quickly adjust the balance between the power generation side and the demand side, it is required that main grid side wholesale market operators have more abundant flexibility resources and additional ancillary services. On the other hand, the access of a large number of distributed energy sources such as controllable distributed generation, photovoltaic power generation, wind power and energy storage to the distribution network makes the distribution network no longer a simple passive network, but a system with stronger flexibility resources. The distributed power sources connected to the distribution network can not only solve problems such as line congestion and voltage quality of the distribution network itself, but also provide flexibility and ancillary services to the main grid. To promote the development of the national unified power market, it is urgent to improve and perfect the coordination strategy between the main grid side wholesale market and the distribution side retail market intervened by VPP to make full use of the flexibility resources of the low-voltage distribution network to reduce the operation cost of the highly uncertain power system. Summary of the Invention

[0006] Object of the Invention: The object of the present invention is to solve the deficiencies existing in the prior art and provide a distributed coordination optimization method for virtual power plants intervening in the main and distribution two-level markets based on a reputation control mechanism; to ensure that each trading entity can actively perform trading contracts, a reputation evaluation index based on the contract performance degree is established to restrain the default behavior of market entities in the actual electricity delivery stage. The present invention mainly associates the reputation value with the deposit to be paid and the market access threshold to inhibit the default behavior of each entity. The coordination between the main grid wholesale market and each distribution network retail market is conducive to clarifying the coupling relationship between supply and demand and prices in the transmission and distribution network and solving the problem of trading barriers caused by market liquidity between regions.

[0007] Technical Solution: A distributed coordination optimization method for virtual power plants intervening in the main and distribution two-level markets based on a reputation control mechanism. For each market entity in the virtual power plant VPP, first calculate their respective corresponding reputation values according to the power deviation rate, and then judge the access situation of the virtual power plant participating in the distribution side retail market and the main grid side wholesale market according to the reputation values. Market entities with actual reputation values lower than the preset value are prohibited from participating in the main and distribution two-level markets; for market entities that are allowed to participate in the main and distribution two-level markets, the following optimization and coordination methods are executed:

[0008] Step 1: Set the main grid side wholesale market operator as the energy provider, issue the price signals of the delivered energy and reserve products, and the distribution side retail market operator feeds back the electricity energy demand and the reserve service sharing value to the main grid side wholesale market operator; construct a coordination optimization model of the main and distribution two-level market operators as shown in formulas (3)-(7):

[0009] Objective function:

[0010] Constraint

[0011] g T (u T ,x T )≤0(5)

[0012]

[0013]

[0014] In the above formula, x TSO and are the power variables inside the main grid side and the distribution side of k respectively; is the power variable of the boundary node between the main grid side and the distribution side of k; u TSO and are the price variables inside the main grid and the distribution network of k respectively; F, f, and g are the objective function, equality constraint, and inequality constraint respectively; the superscript TSO and the subscript D k are the main grid and the distribution network of k respectively, k = 1, 2,..., N D , N D is the total number of distribution networks under the main grid;

[0015] Step 2: Decouple the objective function formula (3) according to the boundary information to obtain an optimization clearing sub-problem for the main grid side wholesale market and multiple optimization clearing sub-problems for the distribution side retail market; for the optimization clearing sub-problem of the main grid side wholesale market, construct an optimization model for the operator of the main grid side wholesale market, and for each optimization clearing sub-problem of the distribution side retail market, construct an optimization model for the operator of the distribution network side retail market;

[0016] Step 2.1: The objective of the optimization model for the operator of the distribution network side retail market is to minimize the total power purchase cost of the operator of the kth distribution side retail market , that is, the objective function is as follows:

[0017]

[0018] In the above formula, is the cost of the operator of the kth distribution side retail market for purchasing electrical energy and reserve from the operator of the main grid side wholesale market; is the cost of the electrical energy purchased by the operator of the distribution side retail market from the controllable distributed generation device DG; is the cost of the electrical energy purchased by the operator of the distribution side retail market from each virtual power plant VPP; is an auxiliary function, which characterizes the influence of the optimization result of the operator of the main grid side wholesale market on the operator of the distribution side retail market;

[0019] Step 2.2. The objective of the operator optimization model in the main grid wholesale market is to minimize the total purchase cost F of the operator in the main grid wholesale market, that is, the objective function is: T Minimize, that is, the objective function is:

[0020]

[0021] In the above formula, is the cost of the operator in the main grid wholesale market for purchasing electrical energy and reserves, and κ auxT is the auxiliary objective function;

[0022] Step 2.3. Since the coordinated optimization of the main and distribution two-level markets is achieved by exchanging boundary information, consistency constraints are imposed on the boundaries of the main and distribution two-level markets, that is, the power at the boundary connection between the main grid side and each distribution side is equal; for the distribution network, the main and distribution boundary nodes are the balancing nodes of the distribution network;

[0023]

[0024] In the formula, are the active powers connected between the main grid side and the operator of the kth distribution side retail market respectively; is the active power injected by the operator of the kth distribution side retail market at the balancing node.

[0025] Furthermore, when the virtual power plant participates in the distribution side retail market and the main grid side wholesale market, it includes the following four stages:

[0026] Stage 1. The virtual power plant agent reports the bidding prices and output ranges of various market products to the operator in the main grid side wholesale market and the operator in the distribution side retail market respectively; and the virtual power plant agent determines its bidding method for participating in the power market according to the types and parameters of its internal distributed energy;

[0027] The operator in the main grid side wholesale market and the operator in the distribution side retail market collect the parameters of all market entities participating in the wholesale market and the retail market respectively according to the pre-released trading time, and verify the market product trading prices and trading limit amounts with the upper-level main grid side wholesale market;

[0028] Stage 2. Day-ahead market stage:

[0029] The virtual power plant agent determines the market level it participates in according to the size of its own credit value; according to its day-ahead distributed energy prediction results, it reports its day-ahead bidding plan and formulates a scheduling plan or price incentive method for its internal flexible resources;

[0030] The main grid-side wholesale market operator and the distribution-side retail market operator collect information of all market participants including virtual power plants, including the output range and cost characteristics of various units, clarify the product trading prices, trading demands and power limit quotas agreed between the main grid-side wholesale market operator and the distribution-side retail market operator, and conduct distributed coordinated optimization clearing for the day-ahead wholesale market and the retail market according to the product demands at different levels and the system security operation constraints. The distribution-side retail market operator reports the time-sharing interaction plan of various market products to the main grid-side wholesale market operator, and the main grid-side wholesale market operator issues the time-sharing price of the products to the distribution-side retail market operator;

[0031] Finally, determine the winning bid prices and quantities of each virtual power plant, the direct control equipment scheduling plan, and the product interaction plan between the distribution system and the main grid-side wholesale market. The main grid-side wholesale market operator and the distribution-side retail market operator respectively release the day-ahead market time-sharing clearing results to all market participants;

[0032] Phase 3, Real-time Market Phase:

[0033] The virtual power plant agent updates the real-time operation plan according to the real-time distributed energy prediction results; the main grid-side wholesale market operator and the distribution-side retail market operator update the real-time information of all market participants, conduct real-time optimization clearing, determine the real-time market product interaction plan and product prices between the main grid-side wholesale market operator and the distribution-side retail market operator, and release the real-time market clearing results to all market participants;

[0034] Phase 4, Settlement Phase: The main grid-side wholesale market operator and the distribution-side retail market operator settle all market participants according to the day-ahead and real-time clearing results, update the credit values of virtual power plant nodes, and return the margin or collect liquidated damages; the virtual power plant agent audits the settlement results.

[0035] Furthermore, the specific process of credit value evaluation is as follows:

[0036] Step 1): Set the initial credit value of each virtual power plant to 100, and at the end of each round of transactions, the blockchain platform automatically updates the credit value according to its actual power delivery situation in this round of transactions;

[0037] Step 2): Calculate the credit value of each virtual power plant node according to the power deviation rate;

[0038] If the actual delivered power curve of the trading entity does not match the agreed power curve, the credit value needs to be calculated according to the contract fulfillment rate; to encourage market entities to adopt consistent trading strategies, dynamic calculation of the credit value is realized based on the historical credit value, and by adjusting the weights of historical data, it is ensured that the credit value of market entities depends on the latest data; the specific calculation method is:

[0039]

[0040] Where: R i,t is the credit value of market entity i in the transaction at the t-th period; is the contract power of market entity i in the transaction at the t-th period; is the actual transaction power of market entity i in the transaction at the t-th period; λ is the credit penalty coefficient;

[0041] Step 3): Each market entity needs to pay a margin before the transaction at the t-th period. If it defaults, part of the margin will be deducted as a default penalty; the calculation method of the liquidated damages is as follows:

[0042]

[0043] Where: C i,t is the liquidated damages that entity i needs to pay at the start of the transaction at the t-th period; R i,t-1 is the credit value of entity i after the transaction in the (t - 1)-th period; R max is the maximum value of the credit value; is the market node electricity price of entity i in the t-th period;

[0044] It can be seen from Equation (1) and Equation (2) that the larger the power deviation amount actually delivered by the trading entity, the more the credit value is deducted, and the more liquidated damages need to be paid; in addition, if the credit value of the trading entity is lower than the market-prescribed access threshold, it will be prohibited from participating in the market; therefore, each trading entity needs to actively fulfill the contract to improve its own credit value, avoid paying excessive liquidated damages and obtain market access permission.

[0045] Furthermore, in the operator optimization model of the main network side wholesale market, the power purchase cost of the wholesale market entity providing electrical energy is calculated as follows,

[0046]

[0047] The superscript "E&R" is for electrical energy and reserve, and the subscript "g" is for thermal power generating units; T - D k is the direction of power flowing from the main network to the distribution network; u g,t and v g,t are the unit start-stop state variables; and are the start-stop costs of unit g at the t-th period respectively; and are the purchase cost functions of the main network side wholesale market operator obtaining electrical energy and reserve from the generating unit and the k-th distribution side retail market operator respectively; P g,t is the power purchase amount of the main network side wholesale market operator from generating unit g; is the electrical energy obtained by the main - grid - side wholesale market operator from the k - th distribution - side retail market operator; N G is the total number of main - grid generators; N D is the number of distribution - side retail market operators; T is the total number of time periods;

[0048] The auxiliary objective function κ auxT characterizes the impact of the optimization result of the distribution - side retail market operator on the main - grid - side wholesale market operator to ensure the optimality condition of the global model. The calculation method is as follows:

[0049] κ auxT =(ζ D ) T x B (10)

[0050]

[0051] In the formula, the vector ζ BD is the sensitivity matrix of the distribution - network objective function and constraints with respect to the boundary variable ; is the objective function of the distribution network of the k - th distribution - side retail market operator; is the Lagrange multiplier vector of the equality constraint (4), is the Lagrange multiplier vector of the inequality constraint (5);

[0052] Furthermore, the optimization model of the main - grid - side wholesale market operator needs to satisfy the following four constraints:

[0053] Constraint I, the main - grid system power - balance constraint

[0054]

[0055] In the formula, is the active power generated by the generator at the i - th node of the transmission network; is the active power participated by the virtual power plant; is the active power output by the RES of the main - grid system; is the active power of the load at the i - th node;

[0056] Constraint II, the generator active - power output constraint, the ramp - rate constraint

[0057]

[0058] In the formula, and are the active - power output and its upper and lower limits of the generator at node i, respectively; r i dn 、r i upare the upward and downward ramp rates of the unit respectively;

[0059] Constraint III, line capacity constraint

[0060]

[0061] In the formula, f l max is the upper limit of the power flow transmission of the main grid line l;

[0062] Constraint IV, main grid system reserve capacity demand constraint

[0063]

[0064] In the formula, are the positive and negative reserve capacity demands of generator g; is the total sum of the uncertain power of the main grid.

[0065] The operator optimization model of the distribution network side retail market adopts a radiation network model based on MI-SOCP. In the distribution network, distributed energy DG mainly includes new energy represented by wind and light and flexibility resources represented by micro diesel engines and micro gas turbines; the specific calculation method of the objective function formula (24) is as follows:

[0066] For the cost of the kth distribution network side retail market operator to purchase electric energy and reserve from the main grid side wholesale market operator is equal to the product of the price and the quantity value. The price refers to the nodal marginal price of the electric energy wholesale market of the main grid side wholesale market operator, and the quantity value is the power provided by the market entity of the main grid side wholesale market operator. The expression is as follows;

[0067]

[0068] In the formula, is the price at which the main grid side wholesale market operator provides active power and reserve capacity to the kth distribution network side retail market operator at time t; is the active power and reserve capacity purchased by the kth distribution network side retail market operator from the main grid side wholesale market operator at time t;

[0069] For the cost of the electric energy purchased by the distribution network side retail market operator from the controllable distributed generation device DG Its expression is as follows:

[0070]

[0071] In the formula, is the bid price of the active power of DG at the ith node of the kth distribution network at time t; $P_{i,t}^{DG,k}$ is the active power output of the DG at the $i$-th distribution network node in the $k$-th distribution network at time $t$; $N_{k}$ is the total number of nodes of the $k$-th distribution-side retail market operator;

[0072] For the cost of the distribution-side retail market operator purchasing electrical energy from each virtual power plant (VPP) Its expression is as follows:

[0073]

[0074] In the formula, $b_{i,t}^{VPP,k}$ is the bid price of the virtual power plant at the $i$-th node of the $k$-th distribution-side retail market operator for providing active power and reserve capacity at time $t$; $P_{i,t}^{VPP,k}$ is the active power and reserve capacity purchased by the $k$-th distribution-side retail market operator from the virtual power plant at the $i$-th node at time $t$;

[0075] For the auxiliary function It characterizes the influence of the optimization result of the main grid-side wholesale market operator on the $k$-th distribution-side retail market operator, and its expression is:

[0076]

[0077] In the formula, $\lambda$ T,k is the Lagrange multiplier vector of the equality constraint (8), $D_{k}$ is the power demand of the $k$-th distribution-side retail market operator at the boundary node.

[0078] Furthermore, the optimization model of the distribution-side retail market operator needs to satisfy the following five constraints:

[0079] Constraint i, power balance constraint:

[0080]

[0081] In the formula, $L_{i}^{k}$ is the load demand of the node in the $k$-th distribution network.

[0082] Constraint ii, power constraint transmitted by the main grid-side wholesale market operator to the distribution-side retail market operator:

[0083]

[0084] In the formula, $P_{max}^{k}$ is the maximum power purchased by the distribution-side retail market operator from the main grid-side wholesale market operator;

[0085] Constraint iii, output power constraint of distributed generation (DG) in the virtual power plant, demand response constraint of migratable load:

[0086]

[0087] In the formula, is the ramp value of the DG power at time period t; is the maximum cuttable power of the k-th distribution-side retail market operator; T p is the set of all allowed response time periods during which the DG is cuttable;

[0088] Constraint iv, power flow constraint of the internal branches of the distribution-side retail market operator:

[0089]

[0090] In the formula, is the injection power at node i of the k-th distribution-side retail market operator; is the upper limit of the power flow transmission of branch l of the distribution network k;

[0091] Constraint vi, reserve capacity demand constraint of the distribution-side retail market operator:

[0092]

[0093] In the formula, is the cut-off amount of the migratable load at node i of the k-th distribution-side retail market operator; is the increase or decrease amount of the distributed power generation; are the positive and negative reserve demands of the k-th distribution-side retail market operator.

[0094] Beneficial effects: The present invention utilizes the virtual power plant credit mechanism to improve market efficiency and reduce systemic risks. The behavior of the virtual power plant participating in the market is more effectively supervised and incentivized, avoiding malicious default or excessive reliance on uncontrollable resources. This mechanism not only improves the transparency of the power market but also strengthens the dispatching trust of the main grid-side wholesale market operator and the distribution-side retail market operator in distributed resources, thereby improving the stability and reliability of the coordinated operation of the two-level market. The coordinated operation of the main and distribution two-level markets considering the dynamic credit value of the virtual power plant proposed by the present invention has significant advantages in the overall economic operation benefit of the power system, which helps to further promote the construction and development of the national unified power market. Description of the Drawings

[0095] Figure 1 is the overall flowchart of the distributed coordinated optimization of the main and distribution two-level markets of the present invention;

[0096] Figure 2 is the schematic diagram of the layout of each server in the two-level market of the present invention;

[0097] Figure 3 is the schematic diagram of the process in different stages of the present invention;

[0098] Figure 4 Schematic diagram of the topological structure of the main and distribution systems of the present invention;

[0099] Figure 5 Schematic diagram of the electricity energy demand and the available reserve capacity of two distribution - side retail market operators in the embodiment;

[0100] Figure 6 Schematic diagram of the winning bid power and actual power results of the virtual power plant in the embodiment;

[0101] Figure 7 Schematic diagram of the dynamic credit value and default cost results of the virtual power plant in the embodiment. Detailed implementation manners

[0102] The technical solution of the present invention will be described in detail below, but the protection scope of the present invention is not limited to the described embodiments.

[0103] As Figure 1 shown, taking an example that there are four distribution networks under one transmission network, Figure 1 it shows the algorithm flow chart of the distributed coordinated optimization of the main and distribution two - level markets. First, the optimization solution of the distribution - side retail market operator is carried out, and then the optimization solution of the main - network wholesale market operator is carried out.

[0104] As Figure 2 shown, for the virtual power plant intervention method of distributed coordinated optimization of the main and distribution two - level markets based on the credit control mechanism in this embodiment, for each virtual power plant, first calculate the respective corresponding credit value according to the power deviation rate, and then judge the access situation of the corresponding market entity to participate in the distribution - side retail market and the main - network wholesale market according to the credit value. For the virtual power plants with the actual credit value lower than the preset value, they are prohibited from participating in the main and distribution two - level markets; for the market entities that are allowed to participate in the main and distribution two - level markets, the following optimization and coordination methods are executed:

[0105] Step 1: Set the main - network wholesale market operator as the energy provider, issue the price signals of electricity energy and reserve products, and the distribution - side retail market operator feeds back the electricity energy demand and the reserve service sharing value to the main - network wholesale market operator; construct the coordinated optimization model of the main and distribution two - level market operators as shown in formulas (3) - (7). Among them, formula (3) is the objective function of the coordinated optimization of the main and distribution two - level market operators, and formulas (4) and (5) are respectively the equality constraints and inequality constraints of the optimization model of the main - network wholesale market operator; formulas (6) and (7) are respectively the equality constraints and inequality constraints of the optimization model of the distribution - side retail market operator.

[0106] Objective function:

[0107] Constraints

[0108] g T (u T ,x T )≤0(5)

[0109]

[0110] In the above formula, x TSO and are the power variables inside the main grid side and the distribution side k respectively; is the power variable of the boundary node between the main grid side and the distribution side k; u TSO and are the price variables inside the main grid and the distribution network k respectively; F, f, and g are the objective function, equality constraint, and inequality constraint respectively; the superscript TSO and the subscript D k are the main grid and the distribution network k respectively, k = 1, 2,..., N D , N D is the total number of distribution networks under the main grid;

[0111] Step 2: Decouple the objective function formula (3) according to the boundary information to obtain an optimization clearing sub-problem for the main grid side wholesale market and multiple optimization clearing sub-problems for the distribution side retail market; for the optimization clearing sub-problem of the main grid side wholesale market, construct an operator optimization model for the main grid side wholesale market, and for each optimization clearing sub-problem of the distribution side retail market, construct an operator optimization model for the distribution network side retail market;

[0112] Step 2.1: The objective of the operator optimization model for the distribution network side retail market is to minimize the total power purchase cost of the kth distribution network side retail market operator , that is, the objective function is as follows:

[0113]

[0114] In the above formula, is the cost of the kth distribution network operator purchasing electric energy and reserve from the main grid side wholesale market operator; is the cost of the distribution network operator purchasing electric energy from the controllable distributed generation device DG; is the cost of the distribution network operator purchasing electric energy from each virtual power plant VPP; is an auxiliary function, representing the influence of the optimization result of the main grid side wholesale market operator on the distribution network operator;

[0115] Step 2.2: The objective of the operator optimization model for the main grid side wholesale market is to minimize the total purchase cost F T of the main grid side wholesale market operator, that is, the objective function is:

[0116]

[0117] In the above formula, is the cost for the wholesale market operator to purchase electricity energy and reserve, and κ auxT is the auxiliary objective function, which characterizes the impact of the optimization result of the distribution-side retail market operator on the main-grid-side wholesale market operator;

[0118] Step 2.3: Since the coordinated optimization of the two-level market is achieved by interacting boundary information, consistency constraints are imposed on the boundaries of the main and distribution two-level markets, that is, the power at the boundary connection between the main grid side and each distribution side is equal; for the distribution network, the main-distribution boundary node is the balancing node of the distribution network;

[0119]

[0120] In the formula, are the active powers connected between the main grid side and the k-th distribution network operator respectively; is the injected active power of the k-th distribution network operator at the balancing node.

[0121] As Figure 3 shown, when the virtual power plant agent participates in the distribution-side retail market and the main-grid-side wholesale market in this embodiment, it includes the following four stages:

[0122] Stage 1: The virtual power plant agent reports the bidding prices and output ranges of various market products to the main-grid-side wholesale market operator and the distribution-side retail market operator respectively; and the virtual power plant agent determines its bidding method for participating in the power market according to the types and parameters of its internal distributed energy.

[0123] The main-grid-side wholesale market operator and the distribution-side retail market operator collect market entity parameters such as bidding prices and bidding powers for participating in the wholesale market and the retail market respectively according to the pre-released trading time, and verify the market product trading prices and trading limit amounts with the superior main-grid-side wholesale market.

[0124] Stage 2: Day-ahead market stage:

[0125] The virtual power plant agent determines the market level it participates in according to the size of its own credit value; according to its day-ahead distributed energy prediction results, it reports its day-ahead bidding plan and formulates a scheduling plan or price incentive method for internal flexible resources.

[0126] The main grid-side wholesale market operator and the distribution-side retail market operator collect information on all market participants including virtual power plants, clarify the product trading prices, trading demands, and power limit quotas agreed upon between the main grid-side wholesale market operator and the distribution-side retail market operator, and perform distributed coordinated optimization clearing for the day-ahead wholesale market and retail market according to the product demands at different levels and the constraints of system security operation. The distribution network operator reports the time-sharing interaction plan of various market products to the main grid-side wholesale market operator, and the main grid-side wholesale market operator issues the time-sharing price of the products to the distribution network operator;

[0127] Finally, determine the winning bid prices and quantities of each virtual power plant. The main grid-side wholesale market operator and the distribution-side retail market operator respectively release the day-ahead market time-sharing clearing results to all market participants;

[0128] Phase 3, Real-time Market Phase:

[0129] The virtual power plant agent updates the real-time operation plan according to the real-time distributed energy prediction results; the main grid-side wholesale market operator and the distribution-side retail market operator update the real-time information of all market participants, perform real-time optimization clearing, determine the real-time market product interaction plan and product price between the main grid-side wholesale market operator and the distribution network operator, and release the real-time market clearing results to all market participants;

[0130] Phase 4, Settlement Phase: The main grid-side wholesale market operator and the distribution-side retail market operator settle all virtual power plant agents according to the day-ahead and real-time clearing results, update the credit values of virtual power plant nodes, and return the margin or collect liquidated damages; the virtual power plant agent audits the settlement results.

[0131] The specific process of credit value evaluation in this embodiment is as follows:

[0132] Step 1): Set the initial credit value of the virtual power plant to 100, and at the end of each round of transactions, the blockchain platform automatically updates the credit value according to its actual power delivery situation in this round of transactions;

[0133] Step 2): Calculate the credit value of each virtual power plant VPP node according to the power deviation rate;

[0134] If the actual delivered power curve of the virtual power plant does not match the agreed power curve, the credit value is dynamically calculated; the specific calculation method is as follows:

[0135]

[0136] In the formula: R i,t is the credit value of virtual power plant i in the transaction at the t-th time period; is the contract power of virtual power plant i in the transaction at the t-th time period; is the actual trading power of virtual power plant i in the t-th period; λ is the reputation penalty coefficient;

[0137] Step 3): Each virtual power plant needs to pay a margin before the start of trading in the t-th period. If it defaults, part of the margin will be deducted as a default penalty. The calculation method of the liquidated damages is as follows:

[0138]

[0139] In the formula: C i,t is the liquidated damages that virtual power plant i needs to pay at the start of trading in the t-th period; R i,t-1 is the reputation value of virtual power plant i after trading in the (t - 1)-th period; R max is the maximum value of the reputation value; is the market node electricity price of virtual power plant i in the t-th period.

[0140] In this embodiment, in the operator optimization model of the main network side wholesale market, the purchase cost of the wholesale market entity providing electrical energy is calculated as follows.

[0141]

[0142] In the formula: E&R is electrical energy and reserve, g is a thermal power generating unit; T - D k is the direction of power flowing from the main network to the distribution network; u g,t and v g,t are the unit start-stop state variables; and are the start-up and shutdown costs of unit g in the t-th period respectively; and are the purchase cost functions for the main network side wholesale market operator to obtain electrical energy and reserve from the generating unit and the k-th distribution network operator respectively; P g,t is the power purchased by the main network side wholesale market operator from generating unit g; is the electrical energy obtained by the main network side wholesale market operator from the k-th distribution network operator; N G is the total number of main network generators; N D is the number of distribution network operators; T is the total number of periods;

[0143] The calculation method of the auxiliary objective function κ auxT is as follows:

[0144] κ auxT =(ζ D ) T x B (10)

[0145]

[0146] In the formula, the vector ζ BD is the sensitivity matrix of the objective function and constraint conditions of the distribution network with respect to the boundary variables ; is the objective function of the k-th distribution network operator; is the Lagrange multiplier vector of the equality constraint (4), and is the Lagrange multiplier vector of the inequality constraint (5).

[0147] The operator optimization model of the main network side wholesale market in this embodiment needs to satisfy the following four constraints:

[0148] Constraint I, main network system power balance constraint

[0149]

[0150] In the formula, is the active power generated by the generator at the i-th node of the transmission network; is the active power participated by the virtual power plant; is the active power output by the RES of the main network system; is the active load at the i-th node;

[0151] Constraint II, generator output active power constraint, ramp rate constraint

[0152]

[0153]

[0154] In the formula, and are respectively the active power output and its upper and lower limits of the generator at node i; r i dn , r i up are respectively the upward and downward ramp rates of the unit;

[0155] Constraint III, line capacity constraint

[0156]

[0157] In the formula, is the node injection power of the main network at node i; is the active power generated by the generator at the i-th node; is the active power participated by the virtual power plant; is the active power output by the RES of the main network system; is the active load at the i-th node; V is the set of all nodes in the main network; Ψ l is the power transfer distribution factor of the main network, which is a known quantity; is a column vector composed of the node injection power of all nodes in the main grid; f l max is the upper limit of the power flow transmission of the main grid line l;

[0158] Constraint IV, the constraint on the reserve capacity demand of the main grid system. Formulas (20)-(21) are the rated capacity constraints of the generators, and Formulas (22)-(23) are the positive and negative spinning reserve capacity constraints of the system.

[0159]

[0160] In the formula, are the positive and negative reserve capacity demands of generator g; are the maximum and minimum power generation powers of generator g; is the total sum of the uncertain power in the main grid.

[0161] The specific calculation method of the objective function formula (24) in this embodiment is as follows:

[0162] For the cost of the k-th distribution-side retail market operator purchasing electrical energy and reserve from the main-grid-side wholesale market operator is equal to the product of the price and the quantity value. The price refers to the nodal marginal price of the electrical energy wholesale market of the main-grid-side wholesale market operator, and the quantity value is the power provided by the market entity of the main-grid-side wholesale market operator. The expression is as follows;

[0163]

[0164] In the formula, is the price at which the main-grid-side wholesale market operator provides active power and reserve capacity to the k-th distribution-side retail market operator at time t; is the active power and reserve capacity purchased by the k-th distribution-side retail market operator from the main-grid-side wholesale market operator at time t;

[0165] For the cost of the distribution-side retail market operator purchasing electrical energy from the controllable distributed generation device DG Its expression is as follows:

[0166]

[0167] In the formula, is the bidding price of the active power of DG at the i-th node of the k-th distribution network at time t; is the active power output by DG at the i-th node of the k-th distribution network at time t; is the total number of nodes of the k-th distribution-side retail market operator;

[0168] For the cost of the distribution-side retail market operator purchasing electrical energy from each virtual power plant VPP Its expression is as follows:

[0169]

[0170] In the formula, is the bidding price for the active power and reserve capacity provided by the virtual power plant at node i of the k-th distribution-side retail market operator at time t; is the active power and reserve capacity purchased by the k-th distribution-side retail market operator from the virtual power plant at node i at time t;

[0171] For the auxiliary function characterizes the impact of the optimization result of the main grid-side wholesale market operator on the k-th distribution-side retail market operator, and its expression is:

[0172]

[0173] In the formula, λ T,k is the Lagrange multiplier vector of the equality constraint (8), is the power demand of the k-th distribution-side retail market operator at the boundary node.

[0174] The operator optimization model of the distribution network-side retail market in this embodiment needs to satisfy the following five constraints:

[0175] Constraint i, power balance constraint:

[0176]

[0177] In the formula, is the load demand of the nodes in the distribution network k;

[0178] Constraint ii, power constraint transmitted from the main grid-side wholesale market operator to the distribution-side retail market operator:

[0179]

[0180] In the formula, is the maximum power purchased by the distribution-side retail market operator from the main grid-side wholesale market operator;

[0181] Constraint iii, output power constraint of distributed generation DG in the virtual power plant, demand response constraint of migratable load:

[0182]

[0183] In the formula, is the ramp value of the DG power at time t; is the maximum cuttable power of the k-th distribution-side retail market operator; T pis the set of all allowed response time periods;

[0184] Constraint iv. Internal branch power constraints of retail market operators on the distribution side:

[0185]

[0186] In the formula, is the injected power at the kth distribution-side retail market operator node i; is the upper limit of power flow transmission of branch l of distribution network k;

[0187] Constraint vi: Reserve capacity demand constraints for retail market operators on the distribution side:

[0188]

[0189] In the formula, is the shedding amount of the migratable load at the k-th distribution-side retail market operator node i; The increase or decrease of the power generated by the distributed generation; is the positive and negative reserve demand of the kth distribution side retail market operator.

[0190] The settings of the distributed coordination optimization simulation example of the main and distribution two-level market in this embodiment are as follows: the main network adopts the IEEE39-node transmission system, 8 nodes are connected to the IEEE33-node distribution system, and 16 nodes are connected to the IEEE69-node distribution system. The topology is as follows: Figure 4 The market is cleared in 1h cycles, and the total number of clearing periods is 24.

[0191] Figure 5 The results of the power purchased by distribution network 1 and distribution network 2 from the transmission network are shown, as well as the reserve capacity that distribution network 1 and distribution network 2 can bear respectively. Figure 5 It can be seen that due to the existence of distributed energy generation inside distribution network 1 and distribution network 2, the demand for distribution network 1 and distribution network 2 to purchase electricity from the transmission network is not much; at the same time, since the virtual power plant can provide backup capacity, the lower-level distribution network can take the initiative to assume part of the backup capacity required by the main network.

[0192] Figure 5 It shows that the optimization method of the present invention can promote information interaction between markets at different levels, enhance the liquidity of electric energy and reserve in the main and distribution markets, and increase the enthusiasm of virtual power plants within the distribution network to participate in the market, thereby reducing the operating cost of the entire power system.

[0193] Figure 6Shows the winning bid power and actual power of virtual power plants 1 to 5. The greater the difference between the actual power and the winning bid power, the worse the credit value of the virtual power plant, and the more default penalty amount needs to be paid; Figure 7 Shows the dynamic credit value and default cost of VPP1-6. Combining Figure 6 and Figure 7 it can be seen that compared with virtual power plants 1 and 5, the difference between the actual power and the winning bid power of virtual power plant 2 is relatively large, resulting in the fastest decline in its credit value and the need to pay more default costs.

[0194] Figure 7 In (a) is the schematic diagram of the dynamic credit value and default cost of VPP1-VPP3, Figure 7 In (b) is the schematic diagram of the dynamic credit value and default cost of VPP4-VPP6; Figure 7 Verifies the effectiveness of the market operator coordination optimization method considering the credit value of virtual power plants proposed in the present invention.

Claims

1. A distributed coordination optimization method for virtual power plants intervening in the main and distribution two-level markets based on a credit control mechanism, characterized in that: For each virtual power plant, first calculate their respective dynamic credit values based on the power deviation rate, and then determine the access status of the corresponding market entities to the distribution-side retail market and the main-grid-side wholesale market according to the historical cumulative credit values. Virtual power plants with actual credit values lower than the preset value are prohibited from participating in the main and distribution-level markets. For market entities that are allowed to participate in the main and distribution-level markets, the following optimization and coordination methods are implemented: Step 1: Set the main-grid-side wholesale market operator as the energy provider, and issue price signals for electricity energy and reserve products. The distribution-side retail market operator feeds back the electricity energy demand and the reserve service sharing value to the main-grid-side wholesale market operator. Based on the generalized master-slave decomposition and coordination method, construct the coordination optimization model of the main and distribution-level market operators as shown in equations (3)-(7). Among them, equation (3) is the objective function of the coordination optimization of the main and distribution-level market operators, and equations (4) and (5) are the equality constraints and inequality constraints of the optimization model of the main-grid-side wholesale market operator respectively; equations (6) and (7) are the equality constraints and inequality constraints of the optimization model of the distribution-side retail market operator respectively. Objective function: Constraint g T (u T ,x T )≤0(5) In the above formula, x TSO and are the power variables within the main grid side and the distribution side k respectively; is the power variable at the boundary node between the main grid side and the distribution side k; u TSO and are the price variables within the main grid and the distribution network k respectively; F, f and g are the objective function, equality constraint and inequality constraint respectively; the superscript TSO and the subscript D k represent the main grid and the distribution network k respectively, k = 1, 2,..., N D , N D is the total number of distribution networks under the main grid; Step 2: Decouple the objective function formula (3) according to the boundary information to obtain a main-grid-side wholesale market optimization clearing sub-problem and multiple distribution-side retail market optimization clearing sub-problems. For the main-grid-side wholesale market optimization clearing sub-problem, construct the optimization model of the main-grid-side wholesale market operator, and for each distribution-side retail market optimization clearing sub-problem, construct the optimization model of the distribution-side retail market operator. Step 2.

1. The objective of the operator optimization model in the distribution network side retail market is to minimize the total power purchase cost of the k-th distribution network side retail market operator, that is, the objective function is as follows: Minimize, that is, the objective function is as follows: In the above formula, is the cost for the k-th distribution network operator to purchase electrical energy and reserve from the main grid side wholesale market operator; is the cost of electrical energy purchased by the distribution network operator from the controllable distributed generation device DG; is the cost of electrical energy purchased by the distribution network operator from each virtual power plant; is an auxiliary function, representing the impact of the optimization result of the main grid side wholesale market operator on the distribution network operator; Step 2.

2. The objective of the operator optimization model in the main network side wholesale market is to minimize the total purchase cost F of the operator in the main network side wholesale market, that is, the objective function is: T Minimize, that is, the objective function is: In the above formula, is the cost for the wholesale market operator to purchase electric energy and reserve, and κ auxT is the auxiliary objective function, which characterizes the impact of the optimization result of the distribution-side retail market operator on the main-grid-side wholesale market operator; Step 2.3: Since the coordination optimization of the main and distribution-level markets is achieved by exchanging boundary information, impose consistency constraints on the main and distribution-level market boundaries, that is, the power at the boundary connection between the main grid and each distribution network is equal. For the distribution network, the main and distribution boundary nodes are the balancing nodes of the distribution network. Wherein, are respectively the active powers connected between the main grid side and the k-th distribution network operator; is the active power injected by the k-th distribution network operator at the balancing node.

2. The distributed coordinated optimization method for the virtual power plant to intervene in the primary and secondary distribution markets based on the reputation control mechanism according to claim 1, wherein, When the virtual power plant agent participates in the distribution-side retail market and the main-grid-side wholesale market, it includes the following four stages: Stage 1: The virtual power plant agent reports the bid prices and output ranges of various market products to the main-grid-side wholesale market operator and the distribution-side retail market operator respectively; and the virtual power plant agent determines its bidding method for participating in the electricity market according to the types and parameters of its internal distributed energy sources. The main-grid-side wholesale market operator and the distribution-side retail market operator collect market entity parameters such as bid prices and bid powers of the participants in the wholesale market and the retail market respectively according to the pre-announced trading time, and verify the market product trading prices and trading limit amounts with the upper-level main-grid-side wholesale market. Stage 2: Day-ahead market stage: The virtual power plant agent determines the market level it participates in according to the size of its own credit value; according to its day-ahead distributed energy prediction results, reports the day-ahead bidding plan, and formulates the scheduling plan or price incentive method for internal flexibility resources. The main grid side wholesale market operator and the distribution side retail market operator collect information of all market participants including virtual power plants, clarify the product trading price, trading demand and power limit amount agreed between the main grid side wholesale market operator and the distribution side retail market operator, and conduct distributed coordinated optimization clearing of the day-ahead wholesale market and retail market according to the product demands at different levels and the system security operation constraints. The distribution network operator reports the time-sharing interaction plan of various market products to the main grid side wholesale market operator, and the main grid side wholesale market operator issues the time-sharing price of the products to the distribution network operator; Finally, determine the winning bid price and quantity of each virtual power plant. The main grid side wholesale market operator and the distribution side retail market operator respectively release the day-ahead market time-sharing clearing results to all market participants; Phase 3, Real-time Market Phase: The virtual power plant agent updates the real-time operation plan according to the real-time distributed energy prediction results. The main grid side wholesale market operator and the distribution side retail market operator update the real-time information of all market participants, conduct real-time optimization clearing, determine the real-time market product interaction plan and product price between the main grid side wholesale market operator and the distribution network operator, and release the real-time market clearing results to all market participants; Phase 4, Settlement Phase: The main grid side wholesale market operator and the distribution side retail market operator settle all virtual power plant agents according to the day-ahead and real-time clearing results, update the credit value of the virtual power plant nodes and return the margin or collect liquidated damages; the virtual power plant agent audits the settlement results.

3. The distributed coordinated optimization method for the virtual power plant to intervene in the primary and secondary markets based on the credit control mechanism according to claim 1 or 2, characterized in that, The specific process of credit value evaluation is as follows: Step 1): Set the initial credit value of the virtual power plant to 100, and at the end of each round of transactions, the blockchain platform automatically updates the credit value according to its actual electricity delivery situation in this round of transactions; Step 2): Calculate the credit value of each virtual power plant VPP node according to the power deviation rate; If the actual delivered power curve of the virtual power plant does not deliver according to the agreed power curve, the credit value is dynamically calculated; the specific calculation method is: Where: R i,t is the credit value of virtual power plant i in the t-th period of trading; is the contract power of virtual power plant i in the t-th period of trading; is the actual trading power of virtual power plant i in the t-th period of trading; λ is the credit penalty coefficient; Step 3): Each virtual power plant needs to pay a margin before the start of trading in the t-th period. If it defaults, part of the margin will be deducted as a default penalty; the calculation method of liquidated damages is: Where: C i,t is the penalty fee that the virtual power plant i needs to pay at the start of trading in the t-th period; R i,t-1 is the credit value of the virtual power plant i after trading in the (t - 1)-th period; R max is the maximum value of the credit value; is the market node electricity price of the virtual power plant i in the t-th period.

4. The distributed coordinated optimization method for virtual power plants to intervene in the primary and secondary distribution markets based on the reputation control mechanism according to claim 1, characterized in that In the operator optimization model of the main network side wholesale market, the electricity purchase cost of the wholesale market entity for providing electrical energy is calculated as follows: Where: E&R is electric energy and reserve, and g is a thermal power generating unit; T-D k is the direction of power flowing from the main grid to the distribution network; u g,t and v g,t are the unit start-stop state variables; and are the start-up and shut-down costs of unit g in period t, respectively; and are the purchase cost functions for the main grid-side wholesale market operator to obtain electric energy and reserve from the generating unit and the kth distribution network operator, respectively; P g,t is the power purchased by the main grid-side wholesale market operator from generating unit g; is the electric energy obtained by the main grid-side wholesale market operator from the kth distribution network operator; N G is the total number of main grid generators; N D is the number of distribution network operators; T is the total number of periods; Auxiliary objective function κ auxT is calculated as follows: κ auxT = (ζ D ) T x B (10) where the vector ζ BD is the sensitivity matrix of the objective function and constraints of the distribution network with respect to the boundary variables ; is the objective function of the k-th distribution network operator; is the Lagrange multiplier vector of the equality constraint (4), and is the Lagrange multiplier vector of the inequality constraint (5).

5. The distributed coordinated optimization method for a virtual power plant to intervene in the primary and secondary distribution markets based on a reputation control mechanism according to claim 1 or 4, characterized in that, The optimization model of the main grid side wholesale market operator needs to meet the following four constraints: Constraint I, Main grid system power balance constraint In the formula, is the active power generated by the generator at the i-th node of the transmission grid; is the active power participated by the virtual power plant; is the active power output by the RES of the main grid system; is the active power of the load at the i-th node; Constraint II, Generator output active power constraint, ramp constraint where and are the active power output of the generator at node i and its upper and lower limits, respectively; r i dn and r i up are the upward and downward ramp rates of the unit, respectively; Constraint III, Line capacity constraint In the formula, is the nodal injection power of the i-th node of the main grid; is the active power generated by the generator at the i-th node; is the active power participated by the virtual power plant; is the active power output by the RES of the main grid system; is the active power of the load at the i-th node; V is the set of all nodes of the main grid; Ψ l is the power transfer distribution factor of the main grid, which is a known quantity; is the column vector composed of the nodal injection powers of all nodes of the main grid; f l max is the upper limit of the power flow transmission of the main grid line l; Constraint IV, Main grid system reserve capacity demand constraint. Formulas (20)-(21) are the rated capacity constraints of generators, and formulas (22)-(23) are the positive and negative spinning reserve capacity constraints of the system; In the formula, is the positive and negative reserve capacity requirements of generator g; is the maximum and minimum power generation of generator g; is the total uncertainty power of the main grid.

6. The distributed coordinated optimization method for the virtual power plant to intervene in the primary and distribution two-level markets based on the reputation control mechanism according to claim 1, characterized in that: The specific calculation method of the objective function formula (24) is as follows: The cost for the k-th distribution-side retail market operator to purchase electric energy and reserve from the main grid-side wholesale market operator is equal to the product of the price and the quantity. The price refers to the nodal marginal price of the electric energy wholesale market of the main grid-side wholesale market operator, and the quantity is the power provided by the market entity of the main grid-side wholesale market operator. The expression is as follows; In the formula, is the price at which the main grid-side wholesale market operator provides active power and reserve capacity to the k-th distribution-side retail market operator during the t period; is the active power and reserve capacity purchased by the k-th distribution-side retail market operator from the main grid-side wholesale market operator during the t period; For the cost of the electric energy purchased by the distribution-side retail market operator from the controllable distributed generation device DG Its expression is as follows: wherein, is the bidding price of the active power of the DG at the k-th distribution network node i during the t-th period; is the active power output by the DG at the k-th distribution network node i during the t-th period; is the total number of nodes of the k-th distribution-side retail market operator; For the cost of the distribution-side retail market operator to purchase electric energy from each virtual power plant (VPP) Its expression is as follows: wherein is the bidding price for the active power and reserve capacity provided by the virtual power plant at node i of the k-th distribution-side retail market operator in period t; is the active power and reserve capacity purchased by the k-th distribution-side retail market operator from the virtual power plant at node i in period t; For the auxiliary function Characterize the impact of the optimization result of the main network side wholesale market operator on the k-th distribution side retail market operator, and its expression is as follows: where λ T,k is the Lagrange multiplier vector of the equality constraint (8), and [[0000202]] is the power demand of the k-th distribution-side retail market operator at the boundary node.

7. The distributed coordinated optimization method for virtual power plants to intervene in the primary and secondary distribution markets based on the reputation control mechanism according to claim 1 or 6, characterized in that, The optimization model of the distribution network side retail market operator needs to meet the following five constraints: Constraint i, Power balance constraint: wherein, is the load demand of the nodes in the distribution network k; Constraint ii, Power constraint transmitted from the main grid side wholesale market operator to the distribution network side retail market operator: In the formula, is the maximum power purchased by the distribution-side retail market operator from the main-grid-side wholesale market operator; Constraint iii, Output power constraint of distributed generation DG inside the virtual power plant, demand response constraint of migratable loads: In the formula, is the ramp value of the DG power at time period t; is the maximum available shedding power of the k-th distribution-side retail market operator; T p is the set of all allowed response time periods; Constraint iv, Internal branch power constraint of the distribution network side retail market operator: wherein, is the injection power at the i-th node of the k-th distribution-side retail market operator; is the upper limit of the power flow transmission of the branch l of the distribution network k; Constraint vi, reserve capacity requirement constraint for distribution-side retail market operators: In the formula, is the curtailment of the transferable load at the \(i\)-th node of the \(k\)-th distribution-side retail market operator; is the increase or decrease in the distributed generation power; are the positive and negative reserve demands of the \(k\)-th distribution-side retail market operator.

Images